Journal of Virology

Fusion-associated small transmembrane (FAST) proteins are viral nonstructural proteins that mediate cell-cell fusion to form multinucleated syncytia. We previously reported that human species B rotavirus NSP1-1 is a FAST protein that induces syncytia in primate epithelial cells but not rodent fibroblasts. We hypothesized that the NSP1-1 proteins of other rotavirus species could also mediate cell-cell fusion and that fusion activity might be limited to cell types derived from homologous hosts. To test this hypothesis, we predicted the structure and domain organization of NSP1-1 proteins of species B rotavirus from a human, goat, and pig, species G rotavirus from a pigeon and turkey, and species I rotavirus from a dog and cat. We cloned these sequences into plasmids and transiently expressed the NSP1-1 proteins in avian, canine, hamster, human, porcine, and simian cells. Regardless of host origin of the virus, each NSP1-1 protein induced syncytia in primate cells, while few induced syncytia in other cell types. To identify the domains that determined cell-specific fusion activity for human species B rotavirus NSP1-1, we engineered chimeric proteins containing domain exchanges with the p10 FAST protein from Nelson Bay orthoreovirus. Using the chimeric proteins, we found that the N-terminal and transmembrane domains determined the cell type specificity of fusion activity. Although the species and cell type criteria for fusion activity remain unclear, these findings suggest that rotavirus species B, G, and I NSP1-1 are functional FAST proteins whose N termini play a role in specifying the cells in which they mediate syncytia formation. IMPORTANCEMechanisms of membrane fusion and determinants of host range for pathogens remain poorly understood. Improved understanding of these concepts could open new areas for therapeutic development and shed light on virus epidemiology. Our analyses of NSP1-1 proteins from species B, G, and I rotaviruses provide insights into the diversity of domain features tolerated by functional FAST proteins. Further, the observation that all putative FAST proteins tested can induce syncytia formation in at least some cell types provides evidence that rotaviruses that encode NSP1-1 proteins are fusogenic viruses. Finally, although the criteria for their specificity remain unclear, our observations regarding fusion capacities of different NSP1-1 proteins and of chimeric FAST proteins suggest a potential role for rotavirus FAST proteins in determining the efficiency of viral replication within a given host or cell type.

Human papillomaviruses (HPV) are causative agents in ano-genital and oral cancers; HPV16 is the most prevalent type detected in human cancers. The HPV16 E6 protein targets p53 for proteasomal degradation to facilitate proliferation of the HPV16 infected cell. However, in HPV16 immortalized cells E6 is predominantly spliced (E6*) and unable to degrade p53. Here we demonstrate that human foreskin keratinocytes immortalized by HPV16 (HFK+HPV16), and HPV16 positive oropharyngeal cancers, retain significant expression of p53. In addition, p53 levels can be increased in HPV16+ head and neck cancer cell lines following treatment with cisplatin. Introduction of full-length E6 into HFK+HPV16 resulted in attenuation of cellular growth (in hTERT immortalized HFK, E6 expression promoted enhanced proliferation). An understudied interaction is that between E2 and p53 and we investigated whether this was important for the viral life cycle. We generated mutant genomes with E2 unable to interact with p53 resulting in profound phenotypes in primary HFK. The mutant induced hyper-proliferation, but an ultimate arrest of cell growth; {beta}-galactosidase staining demonstrated increased senescence, and COMET assays showed increased DNA damage compared with HFK+HPV16 wild type cells. There was failure of the viral life cycle in organotypic rafts with the mutant HFK resulting in premature differentiation and reduced proliferation. The results indicate that the E2-p53 interaction is critical during the HPV16 life cycle, and that disruption of this interaction has anti-viral potential. We discuss potential mechanisms to explain these phenotypes. ImportanceHuman papillomaviruses are causative agents in around 5% of all cancers. There are currently no antivirals available to combat these infections and cancers, therefore it remains a priority to enhance our understanding of the HPV life cycle. Here we demonstrate that an interaction between the viral replication/transcription/segregation factor E2 and the tumor suppressor p53 is critical for the HPV16 life cycle. HPV16 immortalized cells retain significant expression of p53, and the critical role for the E2-p53 interaction demonstrates why this is the case. If the E2-p53 interaction is disrupted then HPV16 immortalized cells fail to proliferate, have enhanced DNA damage and senescence, and there is premature differentiation during the viral life cycle. Results suggest that targeting the E2-p53 interaction would have therapeutic benefits, potentially attenuating the spread of HPV16.

Retroviral reverse transcription starts within the capsid and uncoating and reverse transcription are mutually dependent. There is still debate regarding the timing and cellular location of HIVs uncoating and reverse transcription and whether it occurs solely in the cytoplasm, nucleus or both. HIV can infect non-dividing cells because there is active transport of the preintegration complex (PIC) across the nuclear membrane, but Murine Leukemia Virus (MLV) is thought to depend on cell division for replication and whether MLV uncoating and reverse transcription is solely cytoplasmic has not been studied. Here, we used NIH3T3 and primary mouse dendritic cells to determine where the different stages of reverse transcription occur and whether cell division is needed for nuclear entry. Our data strongly suggest that in both NIH3T3 cells and dendritic cells (DCs), the initial step of reverse transcription occurs in the cytoplasm. However, we detected MLV RNA/DNA hybrid intermediates in the nucleus of dividing NIH3T3 cells and non-dividing DCs, suggesting that reverse transcription can continue after nuclear entry. We also found that the MLV PIC requires cell division to enter the nucleus of NIH3T3 cells. In contrast, we show that MLV can infect non-dividing primary DCs, although integration of MLV DNA in DCs still required the viral p12 protein. Knockdown of several nuclear pore proteins dramatically reduced the appearance of integrated MLV DNA in DCs but not NIH3T3 cells. Additionally, MLV capsid associates with the nuclear pore proteins NUP358 and NUP62 during infection. These findings suggest that simple retroviruses, like HIV, gain nuclear entry by traversing the nuclear pore complex in non-mitotic cells. Author SummaryIt is widely believed that gammaretroviruses like MLV require cell division to achieve nuclear entry and complete their replication. We show here that while this is true for rapidly dividing tissue culture cells, in quiescent cells like dendritic cells, the natural targets of MLV infection, the virus establishes infection without cell division. These studies show that the requirements for retrovirus infection depend on the cell type.

Gammaherpesviruses establish life-long infections within their host and have been shown to be the causative agents of devastating malignancies. Chronic infection within the host is mediated through cycles of transcriptionally quiescent stages of latency with periods of reactivation into more active lytic and productive infection. The mechanisms that regulate reactivation from latency remain poorly understood. Previously, we defined a critical role for the viral cyclin in promoting reactivation from latency. Disruption of the viral cyclin had no impact on the frequency of cells containing viral genome during latency, yet it remains unclear whether the viral cyclin influences latently infected cells in a qualitative manner. To define the impact of the viral cyclin on latent gene expression, we utilized a viral cyclin deficient variant expressing a LANA-beta-lactamase fusion protein (LANA::{beta}la), to enumerate both the cellular distribution and frequency of latent gene expression. Disruption of the viral cyclin did not affect the cellular distribution of latently infected cells, but did result in a significant decrease in the frequency of cells that expressed LANA::{beta}la across multiple tissues and in both immunocompetent and immunodeficient hosts. Strikingly, whereas the cyclin-deficient virus had a reactivation defect in bulk culture, sort purified cyclin-deficient LANA::{beta}la expressing cells were fully capable of reactivation. These data emphasize that the {gamma}HV68 latent reservoir is comprised of at least two distinct stages of infection characterized by differential latent gene expression, and that a primary function of the viral cyclin is to promote latent gene expression within infected cells in vivo.\n\nAUTHOR SUMMARYGammaherpesviruses are ubiquitous viruses with oncogenic potential that establish latency for the life of the host. These viruses can emerge from latency through reactivation, a process that is controlled by the immune system. Control of viral latency and reactivation is thought to be critical to prevent {gamma}HV-associated disease. This study focuses on a virally-encoded cyclin that is required for reactivation from latency. By characterizing how the viral cyclin influences latent infection in pure cell populations, we find that the viral cyclin has a vital role in promoting viral gene expression during latency. This work provides new insight into the function of a virally encoded cyclin in promoting reactivation from latency.

The reovirus {sigma}1 attachment protein mediates virus interaction with host cell receptors that is critical for cell entry. Reovirus tropism is controlled by properties of {sigma}1. {sigma}1 is present as trimers that are held within turrets at the icosahedral vertices of reovirus virions. However, because {sigma}1 has not been visualized on reovirus virions in high resolution structures and because the fulllength structure of purified {sigma}1 protein has not been solved, it is not clear how {sigma}1 is presented on virions. What properties of {sigma}1 are essential for its incorporation on virions is also not known. In this study, we used ColabFold to model the structure of reovirus serotype 1 (T1) and serotype 3 (T3) {sigma}1 proteins. We find that these proteins fold into similar structures with regions of flexibility between the head and body domains of {sigma}1. We also predicted the structures of chimeric {sigma}1 proteins comprised of domain swaps between T1 and T3 {sigma}1 proteins. Our analyses indicate that chimeric proteins with mismatched body and head domain have increased flexibility in this region. Characterization of particles expressing such chimeric {sigma}1 proteins demonstrated that deviation from the flexibility of parental {sigma}1 leads to a reduction in {sigma}1 incorporation on to the virion. Further, we find that even when incorporation is not affected, virus attachment to host cell receptors is influenced by altered {sigma}1 flexibility. Finally, our work demonstrates that 1 protein impacts the encapsidation pattern and receptor engagement properties of {sigma}1 and that this effect is influenced by properties of the N-terminal portion of {sigma}1. ImportanceAttachment to host cell receptors is a critical step in initiation of virus infection. Some viruses attach to cellular receptors via dedicated viral proteins. Both the number of attachment factors present on the virus and whether they are present on the virus particle in the correct form can influence cell attachment. Here, using reovirus as a model, we use a protein structure prediction algorithm to model the as yet unknown structure of full-length reovirus attachment protein {sigma}1. We find predicted regions of flexibility in the protein and identify how this flexibility is regulated. We find that the flexibility of {sigma}1 independently regulates whether it is stably incorporated into particles and if can efficiently interact with host receptors.

Varicella-zoster virus (VZV) is the etiological agent of chickenpox and herpes zoster, while herpes simplex virus 1 (HSV-1) causes oral and genital herpes. Both infections manifest with skin blisters from which the viruses are transmitted to new hosts either via aerosol (VZV) or skin microabrasions (HSV-1). VZV reaches the skin through the blood route, and in the skin epidermis it first infects undifferentiated keratinocytes of the basal layer. Conflicting evidence exists for HSV-1, making it unclear whether HSV-1 infects undifferentiated or differentiated keratinocytes. Here, we developed in vitro models of primary human epidermal keratinocytes differentiation to recapitulate infection of distinct layers of the epidermis by VZV and HSV-1. Our data show that replication of both viruses is restricted, VZV more than HSV-1, if initial infection occurs in differentiated keratinocytes, but not if initial infection occurs in basal undifferentiated keratinocytes. Like VZV, HSV-1 downregulates expression of proteins associated with keratinocyte differentiation, such as the suprabasal keratin K10. However, whereas downregulation of K10 occurs soon after VZV infection and before the virus has replicated, HSV-1-mediated K10 downregulation appears to require full viral replication. These observations provide insights into the potential for VZV and HSV-1 interactions with epidermal differentiation to yield strategies for developing host and pathogen-directed antiviral agents.

Clinical and epidemiologic evidence from the recent outbreak of Oropouche virus (OROV) has demonstrated increased severity in clinical disease and adverse pregnancy outcomes including miscarriage, stillbirth, and neonatal demise. Serological evidence suggests vertical transmission of OROV may be responsible. OROV has not been studied in the context of pregnancy and has unknown ability to infect the relevant tissues of the maternal-fetal interface, which have anti-viral properties; therefore, the mechanisms of vertical transmission are unknown. We used polarized human trophoblast stem cell organoids and human placenta explants to demonstrate that OROV (BeAn19991) infects and replicates in human tissues of the maternal-fetal interface including chorionic villi and the microbial-resistant cell, syncytiotrophoblast. Viral replication is robust within the first 24 hours post infection, and tissues from earlier gestations may be more susceptible to infection. These data indicate tissues at the maternal-fetal interface are susceptible to OROV infection and may facilitate vertical transmission, leading to adverse pregnancy outcomes.

Clade 2.3.4.4b highly pathogenic H5N1 avian influenza (HPAI) viruses started circulating widely in lactating dairy cattle in the United States at the end of 2023. Avian influenza viruses enter cells after binding to glycan receptors with terminally linked 2-3 sialic acid, whereas human influenza viruses typically bind to glycan receptors terminally linked 2-6 sialic acid in the upper respiratory tract. Here, we evaluated the receptor binding properties of hemagglutinin (HA) trimers from a clade 2.3.4.4b avian isolate (A/American Wigeon/South Carolina/22-000345-001/2021) and a cattle isolate (A/dairy cattle/Texas/24-008749-002-v/2024). Using two different methods, we found that both of the 2.3.4.4b H5s bound efficiently to glycan receptors with terminally linked 2-3 sialic acid with no detectable binding to glycan receptors with terminally linked 2-6 sialic acid. Our data suggest that clade 2.3.4.4b H5N1 viruses bind poorly to human receptors. It will be important to continue evaluating receptor binding properties of these viruses as they evolve in cattle.

Alpha herpesviruses naturally infect the peripheral nervous system, and can spread to the central nervous system causing severe deadly or debilitating disease. Because alpha herpesviruses spread along synaptic circuits, and infected neurons exhibit altered electrophysiology and increased spontaneous firing, we hypothesized that alpha herpesviruses use activity-dependent synaptic vesicle-like regulated secretory mechanisms for egress and spread from neurons. To address this hypothesis, we used a compartmentalized primary neuron culture system to measure egress and spread of pseudorabies virus (PRV), pharmacological and optogenetics approaches to modulate neuronal firing activity, and a live-cell fluorescence microscopy assay to directly visualize the exocytosis of individual virus particles from infected neurons. Using tetrodotoxin to silence neuronal activity, we observed no inhibition of virus spread, and using potassium chloride or optogenetics to elevate neuronal activity, we also show no increase in virus spread. Using a live-cell fluorescence microscopy method to directly measure virus egress from infected neurons, we observed no association between virus particle exocytosis and intracellular Ca2+ signaling. Finally, we observed virus particle exocytosis occurs in association with constitutive secretory Rab GTPases, Rab6a and Rab8a, not Rab proteins that are associated with the Ca2+-regulated secretory pathway in neurons, Rab3a and Rab11a. Therefore, we conclude that alpha herpesvirus egress and spread is independent of neuronal activity and Ca2+ signaling because virus particle exocytosis uses constitutive secretory mechanisms in neurons.\n\nAuthor SummaryAlpha herpesviruses, including important human pathogens Herpes Simplex Virus 1 and 2, and Varicella-Zoster Virus, are among the very few viruses that naturally infect the nervous system. These viruses cause recurrent herpetic and zosteriform lesions, peripheral neuropathies, and deadly or debilitating central nervous system diseases. Many of the molecular and cellular mechanisms of viral egress and spread remain unknown, particularly in the context of specialized neuronal cell biology. Our results indicate that elevated firing activity of infected neurons is not functionally or mechanistically linked to virus egress and spread; therefore, therapies targeting peripheral neuropathic symptoms, elevated neuronal activity, and synaptic vesicle secretory mechanisms are unlikely to affect virus spread in the nervous system.

Gammaherpesviruses (GHVs) are lymphotropic tumor viruses with a biphasic infectious cycle. Lytic replication at the primary site of infection is necessary for GHVs to spread throughout the host and establish latency in distal sites. Dissemination is mediated by infected B cells that traffic hematogenously from draining lymph nodes to peripheral lymphoid organs, such as the spleen. B cells serve as the major reservoir for viral latency, and it is hypothesized that periodic reactivation from latently infected B cells contributes to maintaining long-term chronic infection. While fundamentally important to an understanding of GHV biology, aspects of B cell infection in latency establishment and maintenance are incompletely defined, especially roles for lytic replication and reactivation in this cell type. To address this knowledge gap and overcome limitations of replication-defective viruses, we generated a recombinant murine gammaherpesvirus 68 (MHV68) in which ORF50, the gene that encodes the essential immediate-early replication and transcription activator protein (RTA), was flanked by loxP sites to enable conditional ablation of lytic replication by ORF50 deletion in cells that express Cre recombinase. Following infection of mice that encode Cre in B cells with this virus, splenomegaly and viral reactivation from splenocytes were significantly reduced, however the number of latently infected splenocytes was equivalent to WT MHV68. Despite ORF50 deletion, MHV68 latency was maintained over time in spleens of mice at levels approximating WT, reactivation-competent MHV68. Stimulation of polyclonal B cell activation and proliferation by treating mice with lipopolysaccharide (LPS), which promotes MHV68 reactivation ex vivo, yielded equivalent increases in the number of latently infected cells for both ORF50-deleted and WT MHV68, even when mice were simultaneously treated with the antiviral drug cidofovir. Together, these data demonstrate that lytic replication in B cells is not required for MHV68 latency establishment and maintenance and further indicate that B cell proliferation, and not reactivation per se, is a major mechanism for maintaining latent viral genomes in the host. IMPORTANCEGammaherpesviruses establish lifelong chronic infections in cells of the immune system and place infected hosts at risk for developing lymphomas and other diseases. It is hypothesized that gammaherpesviruses must initiate acute infection in these cells to establish and maintain long-term infection, but this has not been directly tested. We report here the use of a viral genetic system that allows for cell-type-specific deletion of a viral gene that is essential for replication and reactivation. We employ this system in an in vivo model to reveal that viral replication is not required to initiate or maintain infection within immune cells.

Respiratory Syncytial Virus (RSV) disease in newborns ranges from mild symptoms to severe disease requiring hospitalization. RSV is classified into two subtypes (RSVA and RSVB) based on antigenic and genetic differences. The role these genomic variations play in disease severity remains unknown. Genome sequences were obtained using next-generation RNA sequencing on archived frozen nasal swabs of young children (< 8 months-old) infected with RSV in Rochester, NY between 1977-1998. Samples were chosen from both children hospitalized with severe RSV disease (inpatient) and those presenting with mild symptoms (outpatient) during their first cold-season. Both A and B subtypes demonstrated significant differences in the phylogeny and primary-protein structure during this time period. We found a significant association between RSV phylogeny over this time period and disease severity. For both subtypes, the G-protein demonstrated the greatest amino acid substitutions, although the number of amino acid substitutions was higher in the RSVA subtype. We found a significant association between G-protein variation and disease severity for RSVA, but not RSVB. For both subtypes, variation in the M2-2 protein was significantly associated with disease severity. These results suggest that the genetic variability of RSV proteins may contribute to disease severity in humans. ImportanceEach cold-season Respiratory Syncytial Virus (RSV) infects thousands of children in the US. Some will display mild cold symptoms while others develop severe disease, sometimes resulting in lifelong lung problems or fatality. RSV initiates infection and replicates in the nasopharynx. Substitutions in the RSV genome can be found in clinically isolated nasal-swab samples of RSV infected children. Whether these genome variations contribute to severe disease is unknown. Here we found a statistically significant association between RSV phylogeny and disease severity. Furthermore, we found specific RSV proteins (G and M2-2) whose amino acid variation was statistically associated with severe disease, although which protein was associated depended on subtype. Taken together, our results suggest that RSV genotype contributed to disease severity over this time period.

Human cytomegalovirus (HCMV) infects various cell types in its human host, and this broad tropism plays a vital role in viral transmission, dissemination, and pathogenesis. HCMV strains differ in their ability to infect and replicate in different cell types, but the genetic determinants of cell tropism have only begun to be understood. A widely used HCMV strain, TB40/E, contains a mixture of genetically distinct virus variants. Only a few passages in ARPE-19 epithelial cells resulted in the selective enrichment of a substrain, termed TB40/EE, which infected epithelial cells more efficiently than the parental TB40/E and induced the formation of large multinucleated syncytia. Herein, we used sequence comparison and genetic engineering of a TB40/E-derived bacterial artificial chromosome clone, TB40-BAC4, to demonstrate that the high infectivity of TB40/EE and its ability to induce syncytia in epithelial cells depends on two single-nucleotide variants (SNVs) affecting the envelope glycoprotein UL128 and the major viral transactivator protein, IE2. While the intronic SNV in UL128 increased splicing of the UL128 transcript, it surprisingly decreased viral infectivity and replication in epithelial cells. The additional introduction of the IE2 SNV reversed this phenotype, increasing infectivity and syncytium formation. This SNV resulted in a D390H substitution and increased the levels of several early and late viral proteins, suggesting that it altered the ability of IE2 to activate viral genes. The same two SNVs increased the ability to infect THP-1-derived macrophages and JEG-3 trophoblast cells. These results demonstrate that HCMV cell tropism depends on both envelope glycoproteins and regulatory proteins. ImportanceDifferent genetic versions of human cytomegalovirus (HCMV) affect its ability to infect various human cell types. Here we focused on a commonly used strain, TB40/E, which contains a mix of virus variants. After growing it in epithelial cells, a specific variant called TB40/EE became dominant. This variant infected epithelial cells more effectively and caused the formation of large, fused cells (syncytia). In this study, we discovered that two small genetic changes were responsible for this behavior. One change affected a protein on the viral envelope (UL128) by altering how its RNA was processed. Surprisingly, this change reduced the viruss ability to spread, but a second change in a regulatory protein (IE2) reversed that effect. Together, these changes enhanced the viruss ability to infect not only epithelial cells but also macrophages and placental cells. This study highlights how small genetic tweaks can influence how HCMV targets different types of human cells.

Human cytomegalovirus (HCMV) infects a wide range of cell types in the body, including a variety of epithelial cell types. Despite the significance of epithelial cells during infection, HCMV has been difficult to study in epithelial cells. In this study, we examined HCMV infection in mammary and prostate epithelial cell lines, finding that the virus establishes a semi-permissive, biosynthetically abortive state. Building on previous work, we hypothesized that shifting epithelial cells to a mesenchymal cell state would restore HCMV biosynthesis and progeny production. To test this hypothesis, we induced epithelial-to-mesenchymal transition (EMT) using TGF-{beta} and the EMT-transcription factor (EMT-TF) SNAIL. We found that shifting strongly epithelial cell lines to a mesenchymal cell state shifted HCMV infection from a semi-permissive to fully permissive state. This effect appeared to involve two distinct mechanisms: EMT-sensitive enhancement of viral entry and EMT-sensitive enhancement of viral mRNA translation. Although the precise mechanisms remain elusive, our findings identify the epithelial-mesenchymal cell state axis as an important regulator of HCMV infection and provide new insights into how cellular differentiation states influence viral replication. They also raise the possibility that the EMT pathway, a fundamental pathway involved in development and cancer metastasis, could regulate HCMV infection in-vivo, potentially contributing to viral persistence or pathogenesis in epithelial tissues.

Equid alphaherpesvirus 1 (EHV1) is a DNA virus that causes severe disease outcomes in equids. Some EHV1 strains are neurotropic and cause disease in the central nervous system, whereas others are non-neurotropic and can cause negative reproductive outcomes. The molecular mechanisms that govern pathotype of individual EHV1 strains are not understood. However, EHV1 replication in the presence of epigenetic inhibitors suggests that neurotropic and non-neurotropic EHV1 are differentially susceptible to epigenetic silencing. Aside from this evidence, little is known about EHV1 chromatin or its regulation. Here, we used fluorescence recovery after photobleaching to characterize EHV1 lytic chromatin dynamics. Infection with neurotropic or non-neurotropic EHV1 mobilized all histones. Canonical (H2A, H2B, H3.1, H4) or variant (H2A.B, H2A.Z, H2A.X, macroH2A, H3.3) core and linker H1.2 histones were equally mobilized by either strain. Thus, there were no vast differences in histone mobility during neurotropic or non-neurotropic EHV1 infection. All histones except for H2A.B were more mobile within EHV1 replication compartments (RCs) than the surrounding infected-cell chromatin. The differential mobility of histones within domains enriched for viral or cellular chromatin is consistent with distinct mechanisms to assemble and regulate the chromatin associated with viral or host DNA. Histones were further mobilized within RCs in cells in which infection had further progressed. Such mobilization indicates that increased levels of EHV1 transcription, DNA replication, or protein expression directly or indirectly mobilize histones. The high histone mobility within EHV1 RCs is consistent with assembly of EHV1 genomes in very dynamic and unstable nucleosomes. These data support a model in which EHV1 limits genome silencing by preventing stable chromatin assembly, or destabilizing the chromatin assembled, with viral genomes during lytic infection. We propose that manipulation of histone dynamics represents a novel mechanism of epigenetic regulation adopted by alphaherpesviruses to maintain genome accessibility and prevent gene silencing. Author summaryDNA viruses are subjected to epigenetic regulation that silences or promotes gene expression. Multiple epigenetic mechanisms contribute to stabilize chromatin to silence gene expression or destabilize it to promote gene expression. Knowledge of the mechanisms whereby viruses prevent or overcome genome silencing and promote expression of their genes is important to understand how viruses, including alphaherpesviruses, take over the host cell to establish productive infection. Here we show that EHV1 broadly mobilizes histones within nuclear domains enriched in viral chromatin. Histone mobilization destabilizes chromatin and is consistent with the assembly of EHV1 genomes in dynamic, unstable nucleosomes. The manipulation of histone mobility is a phenomenon first described for the alphaherpesvirus herpes simplex virus 1 (HSV1). The conserved approach to dysregulate chromatin dynamics and mobilize histones represents a unique means whereby herpesviruses destabilize chromatin. Understanding the mechanisms that mobilize histones during infection will increase our general understanding of epigenetic regulation, which is important in the pathogenesis of infectious diseases and also of developmental or genetic ones. Moreover, knowledge of the processes whereby herpesviruses destabilize chromatin will support the development of novel therapeutics to maintain viral genomes in stable, silenced chromatin to prevent productive infection and development of associated diseases.

Surveillance for emerging human influenza virus clades is important for identifying changes in viral fitness and assessing antigenic similarity to vaccine strains. While fitness and antigenic structure are both important aspects of virus success, they are distinct characteristics and do not always change in a complementary manner. The 2019-20 Northern Hemisphere influenza season saw the emergence of two H1N1 clades: A5a.1 and A5a.2. While several studies indicated that A5a.2 showed similar or even increased antigenic drift compared with A5a.1, the A5a.1 clade was still the predominant circulating clade that season. Clinical isolates of representative viruses from these clades were collected in Baltimore, Maryland during the 2019-20 season and multiple assays were performed to compare both antigenic drift and viral fitness between clades. Neutralization assays performed on serum from healthcare workers pre- and post-vaccination during the 2019-20 season show a comparable drop in neutralizing titers against both A5a.1 and A5a.2 viruses compared with the vaccine strain, indicating that A5a.1 did not have antigenic advantages over A5a.2 that would explain its predominance in this population. Plaque assays were performed to investigate fitness differences, and the A5a.2 virus produced significantly smaller plaques compared with viruses from A5a.1 or the parental A5a clade. To assess viral replication, low MOI growth curves were performed on both MDCK-SIAT and primary differentiated human nasal epithelial cell cultures. In both cell cultures, A5a.2 yielded significantly reduced viral titers at multiple timepoints post-infection compared with A5a.1 or A5a. Receptor binding was then investigated through glycan array experiments which showed a reduction in receptor binding diversity for A5a.2, with fewer glycans bound and a higher percentage of total binding attributable to the top three highest bound glycans. Together these data indicate that the A5a.2 clade had a reduction in viral fitness, including reductions in receptor binding, that may have contributed to the limited prevalence observed after emergence.

The retroviral Gag protein of human immunodeficiency virus type 1 (HIV-1) plays a central role in the selection of unspliced viral genomic RNA for packaging into new virions. Previously, we demonstrated that full-length HIV-1 Gag undergoes nuclear trafficking where it associates with unspliced viral RNA (vRNA) at transcription sites. To further explore the kinetics of HIV-1 Gag nuclear localization, we used biochemical and imaging techniques to examine the timing of HIV-1 entry into the nucleus. We also aimed to determine more precisely Gags subnuclear distribution to test the hypothesis that Gag would be associated with euchromatin, the transcriptionally active region of the nucleus. We observed that HIV-1 Gag localized to the nucleus shortly after its synthesis in the cytoplasm, suggesting that nuclear trafficking was not strictly concentration-dependent. Furthermore, we found that HIV-1 Gag preferentially localized to the transcriptionally active euchromatin fraction compared to the heterochromatin-rich region in a latently-infected CD4+ T cell line (J-Lat 10.6) treated with latency-reversal agents. Interestingly, HIV-1 Gag was more closely associated with transcriptionally-active histone markers near the nuclear periphery, where the HIV-1 provirus was previously shown to integrate. Although the precise function of Gags association with histones in transcriptionally-active chromatin remains uncertain, together with previous reports, this finding is consistent with a potential role for euchromatin-associated Gag molecules to select newly transcribed unspliced vRNA during the initial stage of virion assembly. ImportanceThe traditional view of retroviral assembly posits that HIV-1 Gag selection of unspliced vRNA begins in the cytoplasm. However, our previous studies demonstrated that HIV-1 Gag enters the nucleus and binds to unspliced HIV-1 RNA at transcription sites, suggesting that genomic RNA selection may occur in the nucleus. In the present study, we observed nuclear entry of HIV-1 Gag and co-localization with unspliced viral RNA within 8 hours post-expression. In CD4+ T cells (J-Lat 10.6) treated with latency reversal agents, as well as a HeLa cell line stably expressing an inducible Rev-dependent provirus, we found that HIV-1 Gag preferentially localized with histone marks associated with enhancer and promoter regions of transcriptionally active euchromatin near the nuclear periphery, which favors HIV-1 proviral integration sites. These observations support the hypothesis that HIV-1 Gag hijacks euchromatin-associated histones to localize to active transcription sites, promoting capture of newly synthesized genomic RNA for packaging.

The cellular enzyme poly (ADP-ribose) polymerase-1 (PARP-1) is required for NF-{kappa}B to activate inflammatory and immune response gene expression. NF-{kappa}B is also an important transcription factor in HIV-1 gene expression during active replication and latency reactivation. Therefore, enhancing NF-{kappa}B signaling is an alternative for HIV-1 latency reactivation, but significant systemic side effects related to the NF-{kappa}B role in inflammatory and immune responses are predictable. To verify this prediction, we determined whether PARP-1 is required in NF-{kappa}B-dependent HIV-1 gene expression in a human CD4+ T lymphoblastoid cell line (SUP-T1) and HEK 293T cells. Our findings indicated that PARP-1 knockout does not impair HIV-1 infection or gene expression. Specifically, NF-{kappa}B-dependent HIV-1 gene expression was not impaired by PARP-1 deficiency, highlighting an important transcriptional regulatory difference between HIV-1 and inflammatory and immune activation genes. Our findings define a negligible role of PARP-1 in HIV-1 gene expression, suggesting that PARP-1 antagonism could ameliorate the expected inflammatory response with latency-reactivating agents that act through the NF-{kappa}B signaling pathway. ImportancePARP-1 is required for NF-{kappa}B to activate the expression of inflammatory and immune response genes. NF-{kappa}B is also an important transcription factor in HIV-1 gene expression during active replication and latency reactivation. Enhancing NF-{kappa}B signaling is expected to cause HIV-1 latency reactivation, but significant systemic side effects related to the NF-{kappa}B role in inflammatory and immune responses are predictable. The role of PARP-1 in NF-{kappa}B-mediated activation of HIV-1 gene expression and in viral infection has not been determined in the context of HIV-1 infection of CD4+ T cells. Our data indicate that PARP-1 is dispensable for NF-{kappa}B-mediated activation of HIV-1 gene expression in a human CD4+ T lymphoblastoid cell line. These findings suggest that the pharmacological antagonism of PARP-1 could diminish the inflammatory effects of latency-reactivating agents that activate NF-{kappa}B signaling without impairing their effect on HIV-1 gene expression.

The envelope glycoproteins (Env) on HIV-1 virions are the sole target of broadly neutralizing antibodies (bNAb) and the focus of vaccines. However, many cross-reactive conserved epitopes are often occluded on virus particles, contributing to the evasion of humoral immunity. This study aimed to identify the Env epitopes that are exposed/occluded on HIV-1 particles and to investigate the mechanisms contributing to their masking. Using a flow cytometry-based assay, three HIV-1 isolates, and a panel of antibodies, we show that only select epitopes including V2i, gp120-g41 interface, and gp41-MPER are accessible on HIV-1 particles, while V3, V2q, and select CD4bs epitopes are masked. These epitopes become accessible after allosteric conformational changes are induced by pre-binding of select Abs, prompting us to test if similar conformational changes are required for these Abs to exhibit their neutralization capability. We tested HIV-1 neutralization where virus-mAb mix was pre-incubated/not pre-incubated for one hour prior to adding the target cells. Similar levels of neutralization were observed under both assay conditions, suggesting that the interaction between virus and target cells sensitizes the virions for neutralization via bNAbs. We further show that lectin-glycan interactions can also expose these epitopes. However, this effect is dependent on the lectin specificity. Given that, bNAbs are the ideal for providing sterilizing immunity and are the goal of current HIV-1 vaccine efforts, these data offer insight on how HIV-1 may occlude these vulnerable epitopes from the host immune response. In addition, the findings can guide the formulation of effective antibody combinations for therapeutic use. ImportanceThe human immunodeficiency virus (HIV-1) envelope (Env) glycoprotein mediates viral entry, and is the sole target of neutralizing antibodies. Our data suggest that antibody epitopes including V2q (e.g., PG9, PGT145), CD4bs (e.g., VRC01, 3BNC117) and V3 (2219, 2557) are masked on HIV-1 particles. The PG9 and 2219 epitopes became accessible for binding after conformational unmasking was induced by pre-binding of select mAbs. Attempts to understand the masking mechanism led to the revelation that interaction between virus and host cells is needed to sensitize the virions for neutralization by broadly neutralizing antibodies (bNAbs). These data provide insight on how bNAbs may gain access to these occluded epitopes to exert their neutralization effects and block HIV-1 infection. These findings have important implications for the way we evaluate the neutralizing efficacy of antibodies and can potentially guide vaccine design.

Powassan virus (POWV) is an emerging tick-borne flavivirus that causes disease in humans. POWV has considerable genetic and phenotypic diversity, including highly variable replication in vitro and pathogenesis in mice. This study sought to define the extent of variability in pathogenesis within POWV lineage II in mice and investigate possible viral determinants. Relative to other strains, two New York-derived isolates, NY.19.12 and NY.19.32, caused earlier clinical signs and earlier detection of viral RNA (vRNA) in the spleen and brain compared to mice infected with virus derived from a lineage II infectious clone (WI.97.ic). Sequencing revealed these strains share three amino acid substitutions in envelope, NS1, and NS5 compared to other lineage II strains, which were engineered into a mutant infectious clone. At early time points post-infection, clinical signs, vRNA detection in the cerebellum, and viral distribution in the brain were similar between NY.19.12 and the mutant clone, suggesting these mutations may play a role in disease progression and early neuroinvasion. However, NY.19.12 vRNA was detected in the spleen at significantly higher rates compared to both WI.97.ic and the mutant clone, indicating factors other than these mutations are responsible for increased spleen infection. Importantly, this study highlights the complexity of POWV pathogenesis and suggests that POWV lineage II strains have varying disease phenotypes likely driven by multiple genetic differences. ImportanceTick-borne flaviviruses exhibit considerable genetic and phenotypic diversity in nature, influencing their transmission and pathogenesis. Defining the mechanisms of pathogenesis requires understanding how inter-strain variation translates to phenotypic differences in viral dissemination and neuroinvasion. This study demonstrates that even closely related Powassan virus (POWV) lineage II strains display distinct disease phenotypes that are only partially attributable to nonsynonymous consensus mutations within the viral coding sequence. By highlighting the complexity of POWV infection dynamics in mice, these findings provide valuable insights into how POWV lineage II diversity may shape disease progression and severity in humans.

Herpes simplex virus 1 (HSV-1) transcription is restricted in latently infected neurons and the genomes are in mostly silenced chromatin, whereas all viral genes are transcribed in lytically infected cells, in which the genomes are dynamically chromatinized. Epigenetic regulation modulates HSV-1 transcription during lytic, latent, and reactivating infections, but the precise mechanisms are not fully defined. Nucleosomes are dynamic; they slide, breathe, assemble and disassemble. We and others have proposed that the most dynamic HSV-1 chromatin is transcriptionally competent whereas the least dynamic is silenced. However, the mechanisms yielding the unusually dynamic viral chromatin remain unknown. Histone variants affect nucleosome dynamics. The dynamics of H2A, H2A.X and macroH2A were enhanced in infected cells, whereas those of H2A.B uniquely decreased. We constructed stably transduced cells expressing tagged histone H2A, H2A.B, macroH2A, or H2B, which assembles the H2A/H2B nucleosome dimers with all H2A variants. All H2A variants, ectopic, and endogenous H2B, were assembled into HSV-1 chromatin evenly throughout the genome, but canonical H2A was relatively depleted from the viral chromatin whereas H2A.B was enriched in the most dynamic viral chromatin. When viral transcription was restricted, H2A.B became as depleted from the viral chromatin through the entire genome as H2A. We propose that lytic HSV-1 nucleosomes are enriched in the dynamic variant H2A.B/H2B dimers to promote HSV-1 chromatin dynamics and transcriptional competency, and conclude that the dynamics of HSV-1 chromatin are determined in part by the H2A variants. ImportanceHSV-1 transcription is epigenetically regulated during latent and lytic infections, and epigenetic inhibitors have been proposed as potential antiviral drugs to modulate latency and reactivation. However, the detailed mechanisms of regulation of HSV-1 transcription by epigenetics have not been fully characterized and may differ from those regulating cellular transcription. In particular, the lytic HSV-1 chromatin is unusually dynamic, whereas the latent silenced one is not, but the mechanisms resulting in the unique dynamics of the lytic chromatin remain unknown. Here we identify the enrichment on the highly dynamic histone 2A variant H2A in the most dynamic viral chromatin, which provides a mechanistic understanding for its unique dynamics. Future work to identify the mechanisms of enrichment in H2A.B on the viral chromatin may identify novel druggable epigenetic regulators that modulate HSV-1 latency and reactivation.