Deletion of the first cysteine-rich region of the varicella-zoster virus glycoprotein E ectodomain abolishes the gE and gI interaction and differentially affects cell-cell spread and viral entry

Varicella-zoster virus (VZV) glycoprotein E (gE) is the most abundant glycoprotein in infected cells and, in contrast to those of other alphaherpesviruses, is essential for viral replication. The gE ectodomain contains a unique N-terminal region required for viral replication, cell-cell spread, and secondary envelopment; this region also binds to the insulin-degrading enzyme (IDE), a proposed VZV receptor. To identify new functional domains of the gE ectodomain, the effect of mutagenesis of the first cysteine-rich region of the gE ectodomain (amino acids 208 to 236) was assessed using VZV cosmids. Deletion of this region was compatible with VZV replication in vitro, but cell-cell spread of the rOka-ΔCys mutant was reduced significantly. Deletion of the cysteine-rich region abolished the binding of the mutant gE to gI but not to IDE. Preventing gE binding to gI altered the pattern of gE expression at the plasma membrane of infected cells and the posttranslational maturation of gI and its incorporation into viral particles. In contrast, deletion of the first cysteine-rich region did not affect viral entry into human tonsil T cells in vitro or into melanoma cells infected with cell-free VZV. These experiments demonstrate that gE/gI heterodimer formation is essential for efficient cell-cell spread and incorporation of gI into viral particles but that it is dispensable for infectious varicella-zoster virion formation and entry into target cells. Blocking gE binding to gI resulted in severe impairment of VZV infection of human skin xenografts in SCIDhu mice in vivo, documenting the importance of cell fusion mediated by this complex for VZV virulence in skin.     

Varicella-zoster virus glycoprotein E is a critical determinant of virulence in the SCID mouse-human model of neuropathogenesis

Varicella-zoster virus (VZV) is a neurotropic alphaherpesvirus. VZV infection of human dorsal root ganglion (DRG) xenografts in immunodeficient mice models the infection of sensory ganglia. We examined DRG infection with recombinant VZV (recombinant Oka [rOka]) and the following gE mutants: gEΔ27-90, gEΔCys, gE-AYRV, and gE-SSTT. gEΔ27-90, which lacks the gE domain that interacts with a putative receptor insulin-degrading enzyme (IDE), replicated as extensively as rOka, producing infectious virions and significant cytopathic effects within 14 days of inoculation. Since neural cells express IDE, the gE/IDE interaction was dispensable for VZV neurotropism. In contrast, gEΔCys, which lacks gE/gI heterodimer formation, was significantly impaired at early times postinfection; viral genome copy numbers increased slowly, and infectious virus production was not detected until day 28. Delayed replication was associated with impaired cell-cell spread in ganglia, similar to the phenotype of a gI deletion mutant (rOkaΔgI). However, at later time points, infection of satellite cells and other supportive nonneuronal cells resulted in extensive DRG tissue damage and cell loss such that cytopathic changes observed at day 70 were more severe than those for rOka-infected DRG. The replication of gE-AYRV, which is impaired for trans-Golgi network (TGN) localization, and the replication of gE-SSTT, which contains mutations in an acidic cluster, were equivalent to that of rOka, causing significant cytopathic effects and infectious virus production by day 14; genome copy numbers were equivalent to those of rOka. These experiments suggest that the gE interaction with cellular IDE, gE targeting to TGN sites of virion envelopment, and phosphorylation at SSTT are dispensable for VZV DRG infection, whereas the gE/gI interaction is critical for VZV neurovirulence.     

Varicella-zoster virus ORF63 inhibits apoptosis of primary human neurons

Virus-encoded modulation of apoptosis may serve as a mechanism to enhance cell survival and virus persistence. The impact of productive varicella-zoster virus (VZV) infection on apoptosis appears to be cell type specific, as infected human sensory neurons are resistant to apoptosis, yet human fibroblasts readily become apoptotic. We sought to identify the viral gene product(s) responsible for this antiapoptotic phenotype in primary human sensory neurons. Treatment with phosphonoacetic acid to inhibit viral DNA replication and late-phase gene expression did not alter the antiapoptotic phenotype, implicating immediate-early (IE) or early genes or a virion component. Compared to the parental VZV strain (rOKA), a recombinant virus unable to express one copy of the diploid IE gene ORF63 (rOkaΔORF63) demonstrated a significant induction of apoptosis in infected neurons, as determined by three methods: annexin V staining, deoxynucleotidyltransferase-mediated dUTP-biotin nick end label staining, and transmission electron microscopy. Furthermore, neurons transfected with a plasmid expressing ORF63 resisted apoptosis induced by nerve growth factor withdrawal. These results show that ORF63 can suppress apoptosis of neurons and provide the first identification of a VZV gene encoding an antiapoptotic function. As ORF63 is expressed in neurons during both productive and latent infection, it may play a significant role in viral pathogenesis by promoting neuron survival during primary and reactivated infections.     

Neuronal Subtype and Satellite Cell Tropism Are Determinants of Varicella-Zoster Virus Virulence in Human Dorsal Root Ganglia Xenografts In Vivo

Varicella zoster virus (VZV), a human alphaherpesvirus, causes varicella during primary infection. VZV reactivation from neuronal latency may cause herpes zoster, post herpetic neuralgia (PHN) and other neurologic syndromes. To investigate VZV neuropathogenesis, we developed a model using human dorsal root ganglia (DRG) xenografts in immunodeficient (SCID) mice. The SCID DRG model provides an opportunity to examine characteristics of VZV infection that occur in the context of the specialized architecture of DRG, in which nerve cell bodies are ensheathed by satellite glial cells (SGC) which support neuronal homeostasis. We hypothesized that VZV exhibits neuron-subtype specific tropism and that VZV tropism for SGC contributes to VZV-related ganglionopathy. Based on quantitative analyses of viral and cell protein expression in DRG tissue sections, we demonstrated that, whereas DRG neurons had an immature neuronal phenotype prior to implantation, subtype heterogeneity was observed within 20 weeks and SGC retained the capacity to maintain neuronal homeostasis longterm. Profiling VZV protein expression in DRG neurons showed that VZV enters peripherin+ nociceptive and RT97+ mechanoreceptive neurons by both axonal transport and contiguous spread from SGC, but replication in RT97+ neurons is blocked. Restriction occurs even when the SGC surrounding the neuronal cell body were infected and after entry and ORF61 expression, but before IE62 or IE63 protein expression. Notably, although contiguous VZV spread with loss of SGC support would be predicted to affect survival of both nociceptive and mechanoreceptive neurons, RT97+ neurons showed selective loss relative to peripherin+ neurons at later times in DRG infection. Profiling cell factors that were upregulated in VZV-infected DRG indicated that VZV infection induced marked pro-inflammatory responses, as well as proteins of the interferon pathway and neuroprotective responses. These neuropathologic changes observed in sensory ganglia infected with VZV may help to explain the neurologic sequelae often associated with zoster and PHN.     

Varicella-zoster virus glycoprotein I is essential for growth of virus in Vero cells.

Varicella-zoster virus (VZV) encodes at least six glycoproteins. Glycoprotein I (gI), the product of open reading frame 67, is a 58- to 62-kDa glycoprotein found in VZV-infected cells. We constructed two VZV gI deletion mutants. Immunoprecipitation of VZV gE from infected cells indicated that cells infected with VZV deleted for gI expressed a gE that was larger (100 kDa) than that expressed in cells infected with the parental virus (98 kDa). Cell-associated or cell-free VZV deleted for gI grew to lower titers in melanoma cells than did parental VZV. While VZV deleted for gI replicated in other human cells, the mutant virus replicated to very low titers in primary guinea pig and monkey cells and did not replicate in Vero cells. When compared with the parental virus, rescued viruses, in which the gI deletion was restored with a wild-type allele, showed a similarly sized gE and comparable growth patterns in melanoma and Vero cells. VZV deleted for gI entered Vero cells; however, viral DNA synthesis was impaired in these cells. The VZV gI mutant was slightly impaired for adsorption to human cells. Thus, VZV gI is required for replication of the virus in Vero cells, for efficient replication of the virus in nonhuman cells, and for normal processing of gE.     

Complex Formation Facilitates Endocytosis of the Varicella-Zoster Virus gE:gI Fc Receptor

Open reading frames within the unique short segment of alphaherpesvirus genomes participate in egress and cell-to-cell spread. The case of varicella-zoster virus (VZV) is of particular interest not only because the virus is highly cell associated but also because its most prominent cell surface protein, gE, bears semblance to the mammalian Fc receptor FcγRII. A previous study demonstrated that when expressed alone in cells, VZV gE was endocytosed from the cell surface through a tyrosine localization motif in its cytoplasmic tail (J. K. Olson and C. Grose, J. Virol. 71:4042–4054, 1997). Since VZV gE is normally found in association with gI in the infected cell, the present study was directed at defining the trafficking of the VZV gE:gI protein complex. First, VZV gI underwent endocytosis and recycling when it was expressed alone in cells, and interestingly, VZV gI contained a methionine-leucine internalization motif in its cytoplasmic tail. Second, VZV gI was found by confocal microscopy to colocalize with VZV gE during endocytosis and recycling in cells. Third, by a quantitative internalization assay, VZV gE:gI was shown to undergo endocytosis more efficiently (steady state, 55 to 60%) than either gE alone (steady state, ~32%) or gI alone (steady state, ~45%). Further, examination of endocytosis-deficient mutant proteins demonstrated that VZV gI exerted a more pronounced effect than gE on internalization of the complex. Most importantly, therefore, these studies suggest that VZV gI behaves as an accessory component by facilitating the endocytosis of the major constituent gE and thereby modulating the trafficking of the entire cell surface gE:gI Fc receptor complex.     

Mutational analysis of the role of glycoprotein I in varicella-zoster virus replication and its effects on glycoprotein E conformation and trafficking.

The contributions of the glycoproteins gI (ORF67) and gE (ORF68) to varicella-zoster virus (VZV) replication were investigated in deletion mutants made by using cosmids with VZV DNA derived from the Oka strain. Deletion of both gI and gE prevented virus replication. Complete deletion of gI or deletions of 60% of the N terminus or 40% of the C terminus of gI resulted in a small plaque phenotype as well as reduced yields of infectious virus. Melanoma cells infected with gI deletion mutants formed abnormal polykaryocytes with a disrupted organization of nuclei. In the absence of intact gI, gE became localized in patches on the cell membrane, as demonstrated by confocal microscopy. A truncated N-terminal form of gI was transported to the cell surface, but its expression did not restore plaque morphology or infectivity. The fusogenic function of gH did not compensate for gI deletion or the associated disruption of the gE-gI complex. These experiments demonstrated that gI was dispensable for VZV replication in vitro, whereas gE appeared to be required. Although VZV gI was dispensable, its deletion or mutation resulted in a significant decrease in infectious virus yields, disrupted syncytium formation, and altered the conformation and distribution of gE in infected cells. Normal cell-to-cell spread and replication kinetics were restored when gI was expressed from a nonnative locus in the VZV genome. The expression of intact gI, the ORF67 gene product, is required for efficient membrane fusion during VZV replication.     

Differentiated neuroblastoma cells provide a highly efficient model for studies of productive varicella-zoster virus infection of neuronal cells

Varicella-zoster virus (VZV) is a highly species-specific herpesvirus that targets sensory ganglionic neurons. This species specificity has limited the study of many aspects of VZV pathogenesis, including neuronal infection. We report development of a highly efficient neuroblastoma cell model to study productive VZV infection of neuronal cells. We show that differentiation of SH-SY5Y neuroblastoma cells yields a homogenous population of neuron-like cells that are permissive to the full VZV replicative cycle. These cells supported productive infection by both laboratory and clinical VZV isolates, including the live varicella vaccine. This model may enable rapid identification of genetic determinants facilitating VZV neurotropism.     

Essential role played by the C-terminal domain of glycoprotein I in envelopment of varicella-zoster virus in the trans-Golgi network: interactions of glycoproteins with tegument

Varicella-zoster virus (VZV) is enveloped in the trans-Golgi network (TGN). Here we report that glycoprotein I (gI) is required within the TGN for VZV envelopment. Enveloping membranous TGN cisternae were microscopically identified in cells infected with intact VZV. These sacs curved around, and ultimately enclosed, nucleocapsids. Tegument coated the concave face of these sacs, which formed the viral envelope, but the convex surface was tegument-free. TGN cisternae of cells infected with VZV mutants lacking gI (gI(Delta)) or its C (gI(DeltaC))- or N-terminal (gI(DeltaN))-terminal domains were uniformly tegument coated and adhered to one another, forming bizarre membranous stacks. Viral envelopment was compromised, and no virions were delivered to post-Golgi structures. The TGN was not gI-immunoreactive in cells infected with the gI(Delta) or gI(DeltaN) mutants, but it was in cells infected with gI(DeltaC) (because the ectodomains of gI and gE interact). The presence in the TGN of gI lacking a C-terminal domain, therefore, was not sufficient to maintain enveloping cisternae. In cells infected with intact VZV or with gI(Delta), gI(DeltaN), or gI(DeltaC) mutants, ORF10p immunoreactivity was concentrated on the cytosolic face of TGN membranes, suggesting that it interacts with the cytosolic domains of glycoproteins. Because of the gE-gI interaction, cotransfected cells that expressed gE or gI were able to target truncated forms of the other to the TGN. Our data suggest that the C-terminal domain of gI is required to segregate viral and cellular proteins in enveloping TGN cisternae.     

Varicella-zoster virus glycoprotein M homolog is glycosylated, is expressed on the viral envelope, and functions in virus cell-to-cell spread

Although envelope glycoprotein M (gM) is highly conserved among herpesviruses, the varicella-zoster virus (VZV) gM homolog has never been investigated. Here we characterized the VZV gM homolog and analyzed its function in VZV-infected cells. The VZV gM homolog was expressed on virions as a glycoprotein modified with a complex N-linked oligosaccharide and localized mainly to the Golgi apparatus and the trans-Golgi network in infected cells. To analyze its function, a gM deletion mutant was generated using the bacterial artificial chromosome system in Escherichia coli, and the virus was reconstituted in MRC-5 cells. VZV is highly cell associated, and infection proceeds mostly by cell-to-cell spread. Compared with wild-type VZV, the gM deletion mutant showed a 90% reduction in plaque size and 50% of the cell-to-cell spread in MRC-5 cells. The analysis of infected cells by electron microscopy revealed numerous aberrant vacuoles containing electron-dense materials in cells infected with the deletion mutant virus but not in those infected with wild-type virus. However, enveloped immature particles termed L particles were found at the same level on the surfaces of cells infected with either type of virus, indicating that envelopment without a capsid might not be impaired. These results showed that VZV gM is important for efficient cell-to-cell virus spread in cell culture, although it is not essential for virus growth.     

Glycoprotein I of varicella-zoster virus is required for viral replication in skin and T cells

Varicella-zoster virus (VZV) glycoprotein I (gI) is dispensable in cell culture; the SCIDhu model of VZV pathogenesis was used to determine whether gI is necessary in vivo. The parental and repaired viruses grew in human skin and thymus/liver implants, but the gI deletion mutant was not infectious. Thus, gI is essential for VZV infectivity in skin and T cells.     

Productive varicella-zoster virus infection of cultured intact human ganglia

Varicella-zoster virus (VZV) is a species-specific herpesvirus which infects sensory ganglia. We have developed a model of infection of human intact explant dorsal root ganglia (DRG). Following exposure of DRG to VZV, viral antigens were detected in neurons and nonneuronal cells. Enveloped virions were visualized by transmission electron microscopy in neurons and nonneuronal cells and within the extracellular space. Moreover, rather than remaining highly cell associated during infection of cultured cells, such as fibroblasts, cell-free VZV was released from infected DRG. This model enables VZV infection of ganglionic cells to be studied in the context of intact DRG.     

Upregulation of CXCL10 in human dorsal root ganglia during experimental and natural varicella-zoster virus infection

Varicella-zoster virus (VZV) reactivation causes herpes zoster, which is accompanied by an influx of lymphocytes into affected ganglia, but the stimulus for this infiltrate is not known. We report that VZV infection of ganglia leads to increased CXCL10 production in vitro, in an explant ganglion model and in naturally infected dorsal root ganglia (DRG) during herpes zoster. Lymphocytes expressing the receptor for CXCL10, CXCR3, were also observed throughout naturally infected ganglia during herpes zoster, including immediately adjacent to neurons. This study identifies VZV-induced CXCL10 as a potential driver of T lymphocyte recruitment into DRG during herpes zoster.     

Herpes Simplex Virus Membrane Proteins gE/gI and US9 Act Cooperatively To Promote Transport of Capsids and Glycoproteins from Neuron Cell Bodies into Initial Axon Segments

Herpes simplex virus (HSV) and other alphaherpesviruses must move from sites of latency in ganglia to peripheral epithelial cells. How HSV navigates in neuronal axons is not well understood. Two HSV membrane proteins, gE/gI and US9, are key to understanding the processes by which viral glycoproteins, unenveloped capsids, and enveloped virions are transported toward axon tips. Whether gE/gI and US9 function to promote the loading of viral proteins onto microtubule motors in neuron cell bodies or to tether viral proteins onto microtubule motors within axons is not clear. One impediment to understanding how HSV gE/gI and US9 function in axonal transport relates to observations that gE⁻, gI⁻, or US9⁻ mutants are not absolutely blocked in axonal transport. Mutants are significantly reduced in numbers of capsids and glycoproteins in distal axons, but there are less extensive effects in proximal axons. We constructed HSV recombinants lacking both gE and US9 that transported no detectable capsids and glycoproteins to distal axons and failed to spread from axon tips to adjacent cells. Live-cell imaging of a gE⁻/US9⁻ double mutant that expressed fluorescent capsids and gB demonstrated >90% diminished capsids and gB in medial axons and no evidence for decreased rates of transport, stalling, or increased retrograde transport. Instead, capsids, gB, and enveloped virions failed to enter proximal axons. We concluded that gE/gI and US9 function in neuron cell bodies, in a cooperative fashion, to promote the loading of HSV capsids and vesicles containing glycoproteins and enveloped virions onto microtubule motors or their transport into proximal axons.     

Varicella-zoster virus (VZV) infection of neurons derived from human embryonic stem cells: direct demonstration of axonal infection, transport of VZV, and productive neuronal infection

Study of the human neurotrophic herpesvirus varicella-zoster virus (VZV) and of its ability to infect neurons has been severely limited by strict viral human tropism and limited availability of human neurons for experimentation. Human embryonic stem cells (hESC) can be differentiated to all the cell types of the body including neurons and are therefore a potentially unlimited source of human neurons to study their interactions with human neurotropic viruses. We report here reproducible infection of hESC-derived neurons by cell-associated green fluorescent protein (GFP)-expressing VZV. hESC-derived neurons expressed GFP within 2 days after incubation with mitotically inhibited MeWo cells infected with recombinant VZV expressing GFP as GFP fusions to VZV proteins or under an independent promoter. VZV infection was confirmed by immunostaining for immediate-early and viral capsid proteins. Infection of hESC-derived neurons was productive, resulting in release into the medium of infectious virions that appeared fully assembled when observed by electron microscopy. We also demonstrated, for the first time, VZV infection of axons and retrograde transport from axons to neuronal cell bodies using compartmented microfluidic chambers. The use of hESC-derived human neurons in conjunction with fluorescently tagged VZV shows great promise for the study of VZV neuronal infection and axonal transport and has potential for the establishment of a model for VZV latency in human neurons.     

Glycoproteins E and I facilitate neuron-to-neuron spread of herpes simplex virus.

Two herpes simplex virus (HSV) glycoproteins E and I (gE and gI) form a heterooligomer which acts as an Fc receptor and also facilitates cell-to-cell spread of virus in epithelial tissues and between certain cultured cells. By contrast, gE-gI is not required for infection of cells by extracellular virus. HSV glycoproteins gD and gJ are encoded by neighboring genes, and gD is required for both virus entry into cells and cell-to-cell spread, whereas gJ has not been shown to influence these processes. Since HSV infects neurons and apparently spreads across synaptic junctions, it was of interest to determine whether gD, gE, gI and gJ are also important for interneuronal transfer of virus. We tested the roles of these glycoproteins in neuron-to-neuron transmission of HSV type 1 (HSV-1) by injecting mutant viruses unable to express these glycoproteins into the vitreous body of the rat eye. The spread of virus infection was measured in neuron-rich layers of the retina and in the major retinorecipient areas of the brain. Wild-type HSV-1 and a gJ- mutant spread rapidly between synaptically linked retinal neurons and efficiently infected major retinorecipient areas of the brain. gD mutants, derived from complementing cells, infected only a few neurons and did not spread in the retina or brain. Mutants unable to express gE or gI were markedly restricted in their ability to spread within the retina, produced 10-fold-less virus in the retina, and spread inefficiently to the brain. Furthermore, when compared with wild-type HSV-1, gE- and gI- mutants spread inefficiently from cell to cell in cultures of neurons derived from rat trigeminal ganglia. Together, our results suggest that the gE-gI heterooligomer is required for efficient neuron-to-neuron transmission through synaptically linked neuronal pathways.     

The amino terminus of varicella-zoster virus (VZV) glycoprotein E is required for binding to insulin-degrading enzyme, a VZV receptor

Varicella-zoster virus (VZV) glycoprotein E (gE) is required for VZV infection. Although gE is well conserved among alphaherpesviruses, the amino terminus of VZV gE is unique. Previously, we showed that gE interacts with insulin-degrading enzyme (IDE) and facilitates VZV infection and cell-to-cell spread of the virus. Here we define the region of VZV gE required to bind IDE. Deletion of amino acids 32 to 71 of gE, located immediately after the predicted signal peptide, resulted in loss of the ability of gE to bind IDE. A synthetic peptide corresponding to amino acids 24 to 50 of gE blocked its interaction with IDE in a concentration-dependent manner. However, a chimeric gE in which amino acids 1 to 71 of VZV gE were fused to amino acids 30 to 545 of herpes simplex virus type 2 gE did not show an increased level of binding to IDE compared with that of full-length HSV gE. Thus, amino acids 24 to 71 of gE are required for IDE binding, and the secondary structure of gE is critical for the interaction. VZV gE also forms a heterodimer with glycoprotein gI. Deletion of amino acids 163 to 208 of gE severely reduced its ability to form a complex with gI. The amino portion of IDE, as well an IDE mutant in the catalytic domain of the protein, bound to gE. Therefore, distinct motifs of VZV gE are important for binding to IDE or to gI.     

A Chicken Embryo Eye Model for the Analysis of Alphaherpesvirus Neuronal Spread and Virulence

We describe use of developing chicken embryos as a model to study neuronal spread and virulence of pseudorabies virus (PRV). At embryonic day 12, β-galactosidase-expressing PRV strains were injected into the vitreous humor of one eye, and virus replication and spread from the eye to the brain were measured by β-galactosidase activity and the recovery of infectious virus from tissues. The wild-type PRV strain, Becker, replicated in the eye and then spread to the brain, causing extensive pathology characterized by edema, hemorrhage, and necrosis that localized to virally infected tissue. The attenuated vaccine strain, Bartha, replicated in the eye and spread throughout specific regions of the brain, producing little to no overt pathology. Becker mutants lacking membrane proteins gE or gI replicated in the eye and were able to spread to the brain efficiently. The pathology associated with replication of these mutants in the brain was intermediate to that induced by Becker or Bartha. Mixed infection of a gE deletion mutant and a gI deletion mutant restored the pathogenic phenotype to wild-type levels. These data indicate that the replication of virus in embryonic brain tissue is not sufficient to induce the characteristic pathological response and that the gE and gI gene products actively affect pathological responses in the developing chicken brain.