Search

ABSTRACTThe tegument of herpesvirus virions is a distinctive structure whose assembly and function are not well understood. The herpes simplex virus type 1 VP22 tegument protein encoded by the UL49 gene is conserved among the alphaherpesviruses. Using cell biology and viral genetics, we provide an initial characterization of the pseudorabies virus (PRV) VP22 homologue. We identified three isoforms of VP22 present in PRV-infected cells that can be resolved by polyacrylamide gel electrophoresis. The predominant form is not phosphorylated and is present in virions, while the other two species are phosphorylated and excluded from virions. VP22 localized to the nucleus by 6 h postinfection, as determined by immunofluorescence and cell fractionation. VP22 immunofluorescence in the nucleus was both diffuse and in punctate structures. The punctate nuclear localization was the most pronounced form of staining and did not localize exclusively to sites of viral DNA replication. Unexpectedly, a VP22 null mutant had no obvious phenotypes during tissue culture infections and was similar to the wild type in all respects. Moreover, the VP22 null mutant was as virulent and neuroinvasive as the wild-type virus after infection of the rodent eye and spread to the brain using both anterograde and retrograde neuronal circuits.

ABSTRACTThe alphaherpesvirus Us4 gene encodes glycoprotein G (gG), which is conserved in most viruses of the alphaherpesvirus subfamily. In the swine pathogen pseudorabies virus (PRV), mutant viruses with internal deletions and insertions in the gG gene have shown no discernible phenotypes. We report that insertions in the gG locus of the attenuated PRV strain Bartha show reduced virulence in vivo and are defective in their ability to spread from cell to cell in a cell-type-specific manner. Similar insertions in the gG locus of the wild-type PRV strain Becker had no effect on the ability of virus infection to spread between cells. Insertions in the gG locus of the virulent NIA-3 strain gave results similar to those found with the Bartha strain. To examine the role of gG in cell-to-cell spread, a nonsense mutation in the gG signal sequence was constructed and crossed into the Bartha strain. This mutant, PRV157, failed to express gG yet had cell-to-cell spread properties indistinguishable from those of the parental Bartha strain. These data indicated that, while insertions in the gG locus result in decreased cell-to-cell spread, the phenotype was not due to loss of gG expression as first predicted. Analysis of gene expression upstream and downstream of gG revealed that expression of the upstream Us3 protein is reduced by insertion of lacZor egfpat the gG locus. By contrast, expression of the gene immediately downstream of gG, Us6, which encodes glycoprotein gD, was not affected by insertions in gG. These data indicate that DNA insertions in gG have polar effects and suggest that the serine/threonine kinase encoded by the Us3 gene, and not gG, functions in the spread of viral infection between cells.

ABSTRACTThe membrane proteins gI and gE of Pseudorabies virus(PRV) are required for viral invasion and spread through some neural pathways of the rodent central nervous system. Following infection of the rat retina with wild-type PRV, virus replicates in retinal ganglion neurons and anterogradely spreads to infect all visual centers in the brain. By contrast, gI and gE null mutants do not infect a specific subset of the visual centers, e.g., the superior colliculus and the dorsal lateral geniculate nucleus. In previous experiments, we suggested that the defect was not due to inability to infect projection-specific retinal ganglion cells, because mixed infection of a gE deletion mutant and a gI deletion mutant restored the wild-type phenotype (i.e., genetic complementation occurred). In the present study, we provide direct evidence that gE and gI function to promote the spread of infection after entry into primary neurons. We used stereotaxic central nervous system injection of a fluorescent retrograde tracer into the superior colliculus and subsequent inoculation of a PRV gI-gE double null mutant into the eye of the same animal to demonstrate that viral antigen and fluorescent tracer colocalize in retinal ganglion cells. Furthermore, we demonstrate that direct injection of a PRV gI-gE double null mutant into the superior colliculus resulted in robust infection followed by retrograde transport to the eye and replication in retinal ganglion neuron cell bodies. These experiments provide additional proof that the retinal ganglion cells projecting to the superior colliculus are susceptible and permissive to gE and gI mutant viruses. Our studies confirm that gI and gE specifically facilitate anterograde spread of infection by affecting intracellular processes in the primary infected neuron such as anterograde transport in axons or egress from axon terminals.

ABSTRACTThe glycoproteins I and E of pseudorabies virus are important mediators of cell-to-cell spread and virulence in all animal models tested. Although these two proteins form a complex with one another, ascribing any function to the individual proteins has been difficult. We have shown previously, using nonsense mutations, that the N-terminal ectodomain of the gE protein is sufficient for gE-mediated transsynaptic spread whereas the cytoplasmic domain of the protein is required for full expression of virulence. These same studies demonstrated that the cytoplasmic domain of gE is also required for endocytosis of the protein. In this report, we describe the construction of viruses with nonsense mutations in gI that allowed us to determine the contributions of the gI cytoplasmic domain to protein expression as well as virus neuroinvasion and virulence after infection of the rat eye. We also constructed double mutants with nonsense mutations in both gE and gI so that the contributions of both the gE and gI cytoplasmic domains could be determined. We observed that the gI cytoplasmic domain is required for efficient posttranslational modification of the gI protein. The gE cytoplasmic domain has no effect on gE posttranslational glycosylation. In addition, we found that infection of all gE-gI-dependent anterograde circuits projecting from the rat retina requires both ectodomains and at least one of the cytoplasmic domains of the proteins. The gI cytoplasmic domain promotes transsynaptic spread of virus better than the gE cytoplasmic domain. Interestingly, both gE and gI cytoplasmic tails are required for virulence; lack of either one or both results in an attenuated infection. These data suggest that gE and gI play differential roles in mediating directional neuroinvasion of the rat; however, the gE and gI cytoplasmic domains most likely function together to promote virulence.

ABSTRACTThe protein product of the pseudorabies virus (PRV) Us9 gene is a phosphorylated, type II membrane protein that is inserted into virion envelopes and accumulates in the trans-Golgi network. It is among a linked group of three envelope protein genes in the unique short region of the PRV genome which are absent from the attenuated Bartha strain. We found that two different Us9 null mutants exhibited no obvious phenotype after infection of PK15 cells in culture. Unlike those of gE and gI null mutants, the plaque size of Us9 null mutants on Madin-Darby bovine kidney cells was indistinguishable from that of wild-type virus. However, both of the Us9 null mutants exhibited a defect in anterograde spread in the visual and cortical circuitry of the rat. The visual system defect was characterized by restricted infection of a functionally distinct subset of visual projections involved in the temporal organization of behavior, whereas decreased anterograde spread of virus to the cortical projection targets was characteristic of animals receiving direct injections of virus into the cortex. Spread of virus through retrograde pathways in the brain was not compromised by a Us9 deletion. The virulence of the Us9 null mutants, as measured by time to death and appearance of symptoms of infection, also was reduced after their injection into the eye, but not after cortical injection. Through sequence analysis, construction of revertants, measurement of gE and gI protein synthesis in the Us9 null mutants, and mixed-infection studies of rats, we conclude that the restricted-spread phenotype after infection of the rat nervous system reflects the loss of Us9 and is not an indirect effect of the Us9 mutations on expression of glycoproteins gE and gI. Therefore, at least three viral envelope proteins, Us9, gE, and gI, function together to promote efficient anterograde transneuronal infection by PRV in the rat central nervous system.

After pseudorabies virus (PRV) infection of murine L929 cells, the cell surface expression of major histocompatibility complex (MHC) class I proteins changes such that the total amount of MHC class I molecules remains relatively constant but the levels of the individual alleles Dk and Kk vary. This is an active process involving at least three PRV gene products that act in an allele-specific manner such that cell surface expression of MHC class I Dk is decreased and that of Kk is increased. Our results indicate that an early gene product mediates the overall reduction in Dk protein and a late gene product which is mutant in the attenuated PRV strain Bartha mediates the increase in Kk protein. We provide additional evidence for a third gene product involved in the regulation of the synthesis of both the Dk and Kk proteins. In addition, we show that the early decrease in the Dk protein is not due to a block in synthesis or processing of the complex through the secretory system.

ABSTRACTAfter pseudorabies virus (PRV) infection of murine L929 cells, the cell surface expression of major histocompatibility complex (MHC) class I proteins changes such that the total amount of MHC class I molecules remains relatively constant but the levels of the individual alleles Dkand Kkvary. This is an active process involving at least three PRV gene products that act in an allele-specific manner such that cell surface expression of MHC class I Dkis decreased and that of Kkis increased. Our results indicate that an early gene product mediates the overall reduction in Dkprotein and a late gene product which is mutant in the attenuated PRV strain Bartha mediates the increase in Kkprotein. We provide additional evidence for a third gene product involved in the regulation of the synthesis of both the Dkand Kkproteins. In addition, we show that the early decrease in the Dkprotein is not due to a block in synthesis or processing of the complex through the secretory system.

ABSTRACTThe Us9 protein is a phosphorylated membrane protein present in the lipid envelope of pseudorabies virus (PRV) particles in a unique tail-anchored type II membrane topology. In this report, we demonstrate that the steady-state residence of the Us9 protein is in a cellular compartment in or near the trans-Golgi network (TGN). Through internalization assays with an enhanced green fluorescent protein epitope-tagged Us9 protein, we demonstrate that the maintenance of Us9 to the TGN region is a dynamic process involving retrieval of molecules from the cell surface. Deletion analysis of the cytoplasmic tail reveals that an acidic cluster containing putative phosphorylation sites is necessary for the recycling of Us9 from the plasma membrane. The absence of this cluster results in the relocalization of Us9 to the plasma membrane due to a defect in endocytosis. The acidic motif, however, does not contain signals needed to direct the incorporation of Us9 into viral envelopes. In this study, we also investigate the role of a dileucine endocytosis signal in the Us9 cytoplasmic tail in the recycling and retention of Us9 to the TGN region. Site-directed mutagenesis of the dileucine motif results in an increase in Us9 plasma membrane staining and a partial internalization defect.

ABSTRACTThe pseudorabies virus (PRV) gE gene encodes a multifunctional membrane protein found in infected cell membranes and in the virion envelope. Deletion of the gE gene results in marked attenuation of the virus in almost every animal species tested that is permissive for PRV. A common inference is that gE mutants are less virulent because they have reduced ability to spread from cell to cell; e.g., gE mutants infect fewer cells and, accordingly, animals live longer. In this report, we demonstrate that this inference does not hold in a rat experimental model for virus invasion of the brain. We find that animals infected with gE mutants live longer despite extensive retrograde, transneuronal spread of virus in the rat brain. In this model of brain infection, virus is injected into the stomach musculature and virions spread to the brain in long axons of brain stem neurons that give rise to the tenth cranial nerve (the vagus). The infection then spreads from neuron to neuron in well-defined, and physically separated, areas of the brain involved in autonomic regulation of the viscera. We examined the progression of infection of five PRV strains in this circuitry: the wild-type PRV-Becker strain, the attenuated PRV-Bartha vaccine strain, and three gE mutants isogenic with the PRV-Becker strain. By 60 to 67 h after infection, all PRV-Becker-infected animals were dead. Analysis of Becker-infected rats killed prior to virus-induced death demonstrated that the virus had established an infection only in the primary vagal neurons connected directly to the stomach and synaptically linked neurons in the immediate vicinity of the caudal brain stem. There was little spread to other neurons in the vagus circuitry. In contrast, rats infected with PRV-Bartha or PRV-Becker gE mutants survived to at least 96 h and exhibited few overt signs of disease. Despite this long survival and the lack of symptoms, brains of animals sacrificed at this time revealed extensive transsynaptic infection not only of the brain stem but also of areas of the forebrain synaptically linked to neurons in the brain stem. This finding provides evidence that the gE protein plays a role in promoting symptoms of infection and death in animals that is independent of neuron-to-neuron spread during brain infection. When this early virulence function is not active, animals live longer, resulting in more extensive spread of virus in the brain.

ABSTRACTThe role of alphaherpesvirus membrane protein internalization during the course of viral infection remains a matter of speculation. To determine the role of internalization of the pseudorabies virus (PRV) gE and gI proteins, we constructed viral mutants encoding specific mutations in the cytoplasmic tail of the gE gene that inhibited internalization of the gE-gI complex. We used these mutants to assess the role of gE-gI endocytosis in incorporation of the proteins into the viral envelope and in gE-mediated spread or gE-promoted virulence. In addition, we report that another viral mutant, PRV 25, which encodes a gE protein defective in endocytosis, contains an additional, previously uncharacterized mutation in the gE gene. We compared PRV 25 to another viral mutant, PRV 107, that does not express the cytoplasmic tail of the gE protein. The gE protein encoded by PRV 107 is also defective in endocytosis. We conclude that efficient endocytosis of gE is not required for gE incorporation into virions, gE-mediated virulence, or spread of virus in the rat central nervous system. However, we do correlate the defect in endocytosis to a small-plaque phenotype in cultured cells.

We previously demonstrated that intraocular injections of virulent and attenuated strains of pseudorabies virus (PRV) produce transneuronal infection of functionally distinct central visual circuits in the rat. The virulent Becker strain of PRV induces two temporally separated waves of infection that ultimately target all known retinorecipient neurons; the attenuated Bartha strain only infects a functionally distinct subset of these neurons. In this study, we demonstrate that deletion of a single viral gene encoding glycoprotein gI is sufficient to reproduce both the novel pattern of infectivity and the reduced neurovirulence of the Bartha strain of PRV. Glycoprotein gIII, a major viral membrane protein required for efficient adsorption of virus in cell culture, has no obvious role in determining the pattern of neuronal infectivity, but appears to function with gI to influence neurovirulence. These data suggest that neuroinvasiveness and virulence are the products of an interaction of viral envelope glycoproteins with as yet unidentified cellular receptors.

In the Alphaherpesvirinae subfamily, the gE and gI genes are conserved and encode membrane glycoproteins required for efficient pathogenesis (virulence). The molecular mechanism(s) responsible is not well understood, but the existence of similar phenotypes of gE and gI mutations in diverse Alphaherpesvirinae implies conservation of function(s). In this report, we describe construction of pseudorabies virus (PRV) recombinants that efficiently express the bovine herpesvirus 1 (BHV-1) membrane proteins gI and gE at the PRV gG locus. Each BHV-1 gene was cloned in a PRV mutant lacking both the PRV gI and gE coding sequences. All recombinant viruses expressed the BHV-1 proteins at levels similar to or greater than that observed after infection with parental BHV-1, and there were no observable differences in processing or ability to form gE-gI oligomers. The important observation resulting from this report is that the BHV-1 gE and gI proteins functioned together to complement the virulence defect of PRV lacking its own gE and gI genes in a rodent model, despite being derived from a highly restricted host range virus with a different pathogenic profile.

Pseudorabies virus glycoproteins gE and gI are required to infect some, but not all, regions of the rodent central nervous system after peripheral injection. After infection of the retina, pseudorabies virus mutants lacking either gE or gI can subsequently infect neural centers involved in the control of circadian function but cannot infect visual circuits mediating visual perception or the reflex movement of the eyes. In this study, we used genetic complementation to test the hypothesis that gE and gI are required for entry into the specific retinal ganglion cells that project to visual centers. These data strongly suggest that gE and gI must function after the viruses enter primary neurons in the retina.

The in vivo activity of the recBC nuclease was assayed by transfection of isogenic rec+ and rec minus spheroplasts with bacteriophage DNA of various origin and structure. The results indicate that the recBC nuclease can limit transfection at several stages during the production of an infective center; such limitations depend primarily on whether the DNA is in, or assumes, a nuclease-sensitive structure. The first stage of limitation can occur when a nuclease-sensitive transfecting molecule enters the spheroplast. Other potential limitation points occur during replication and maturation of the bacteriophage DNA. The initial stage can be bypassed by using recBC nuclease-resistant molecules such as circular forms. Through analysis of results with other DNA structures, we found that in vivo the effects of the double-strand exonucleolytic activity of the recBC nuclease predominated. The effects of the single-strand nuclease activities seem to be modified from those observed for the purified enzyme in vitro (Karu et al., 1974). Inside the cell, the single-strand exonuclease activity is very weak and the single-strand endonuclease activity is abolished almost completely.

The mouse L-cell mutant gro29 is defective for egress of herpes simplex virus type 1 (HSV-1) virions and is significantly reduced in HSV-1 glycoprotein export (B. W. Banfield and F. Tufaro, J. Virol. 64:5716-5729, 1990). In this report, we demonstrate that pseudorabies virus (PRV), a distantly related alphaherpesvirus, shows a distinctive set of defects after infection of gro29 cells. Specifically, we identify defects in the rate and extent of viral glycoprotein export, infectious particle formation, plaque formation, and virus egress. The initial rate of viral glycoprotein synthesis was unaffected in gro29 cells, but the extent of export from the endoplasmic reticulum to the Golgi apparatus was impaired and export through the Golgi apparatus became essentially blocked late in infection. Moreover, by using a secreted variant of a viral membrane protein, we found that export from the Golgi apparatus out of the cell was also defective in gro29 cells. PRV does not form plaques on gro29 monolayers. A low level of infectious virus is formed and released early after infection, but further virus egress is blocked. Taken together, these observations suggest that the gro29 phenotype involves either multiple proteins or a single protein used at multiple steps in viral glycoprotein export and virus egress from cells. Moreover, this host cell protein is required by both HSV and PRV for efficient propagation in infected cells.

We have described three mutant strains of Pseudorabies virus that contain mutations in the signal sequence coding region of a nonessential envelope glycoprotein, gIII. The alterations disrupt, truncate, or eliminate the hydrophobic core domain of the signal sequence. Each mutant was assayed for its ability to promote the translocation of gIII across the endoplasmic reticulum membrane and the subsequent localization of the mature form of the glycoprotein to the infected cell surface or the virus envelope. Our results confirm and extend findings in other systems that the overall hydrophobicity of the signal sequence core region is a major determinant of translocation efficiency. We were unable to correlate simply the length of the core or the average hydrophobicity of core residues with export efficiency. Because our work involved the use of infectious virus mutants, we were able to identify a virus defect associated with a complete block in gIII export. This defect will facilitate a pseudoreversion analysis of gIII signal sequence function.

We describe in this paper the isolation and characterization of a class of mutations, designated puq, that allow phage lambda to grow better under conditions that limit the synthesis of the phage Q gene product. These mutations were located between phage genes P and Q, a region of the lambda chromosome containing two gene N-independent mutations, nin5 and byp, that we also show to be puq mutations. Whereas the puq-3 and puq-16 mutations probably map under the nin5 deletion, the byp mutation maps between this deletion and the Q lambda-Q phi 80 crossover point. These mutations likely act by increasing the synthesis of the Q gene product. We demonstrate that the clear-plaque phenotype and reduced lysogenization frequency of byp mutants depend on increased Q gene activity. The significance of these results in understanding how transcription proceeds through the P-Q region of the lambda genome is discussed.

We have identified integrated proviral DNA sequences of m1 and HT-1 isolates of Moloney sarcoma virus (MuSV) in EcoRI digests of transformed mink cell genomic DNA and have cloned these fragments in bacteriophage lambda. Both the lambda-HT1 phage recombinant, containing a 12.3-kilobase MuSV pair (kb) fragment, and the lambda-m1 phage recombinant, containing a 7.0-kb fragment, possess full copies of the sarcoma viruses along with 5' and 3' host flanking sequences. The MuSV proviral DNA sequences, 6.7 kb for HT-1 and 5.2 kb for m1, are colinear by heteroduplex microscopy with the 1.5-kb difference in size accounted for by two approximately equal to 0.8-kb deleted regions in m1. Both integrated viral genomes are terminally redundant and have integrated at the same site in the provirus but at different sites on the host chromosome. The host sequence flanking integrated HT-1 MuSV have been identified as a single EcoRI restriction fragment of 5.6 kb in normal mink cells.

The site-specific recombination system of phage lambda promotes crossovers at its attachment site (att). In this report we show that when phage are crossed in conditions where only the site-specific recombination system is active, a low frequency of crossovers can also be detected in a region that is close to but does not contain att. These crossovers require the phage int gene, the host hip gene, and the integrity of att. They are not detected if one of the parents carries a substitution of a heterologous attachment site (attB instead of attP). To explain these findings we suggest that site-specific recombination can proceed by exchange of single strands between the participating chromosomes at att and migration of the resulting junction outside of att.

A simple, efficient method for introducing recombinant DNA into a herpesvirus vector and retrieving it at a later time has been developed. By using the Cre-lox site-specific recombination system of coliphage P1, DNA can be readily inserted in vitro into a pseudorabies virus (PRV) vector containing the lox recombination site. The vector PRV42 contains a lox site within the nonessential gIII gene, which encodes a virion envelope glycoprotein. Incubation in vitro of PRV42 DNA with Cre protein and a circular plasmid containing a lox site generates approximately 5% recombinant molecules having the plasmid integrated into the PRV genome at the lox site. Transfection of the reaction mixture into cultured cells allows recovery of the infectious recombinant virus, which is readily identified by a nondestructive "black-plaque assay" using a gIII-specific monoclonal antibody. PRV42 plaques stain black when treated with the gIII monoclonal antibody and a peroxidase-linked second anti-antibody because the lox site placed within the gIII gene of PRV42 does not destroy the gIII epitope. However, Cre-mediated integration of heterologous DNA at the lox site disrupts the gIII epitope so that the resulting recombinant virus produces white plaques. The recombinant virus is infectious, stable, and grows as well as the parental PRV42 vector. The inserted plasmid can be efficiently excised (greater than 50%) from viral DNA by Cre and recovered by transformation of Escherichia coli.

We have developed a general approach to the cloning of specific segments of the mammalian genome that involves a two-step purification of EcoRI fragments of mammalian DNA and their in vitro insertion into a suitably constructed EK2 derivative of bacteriophage lambda. The combination of fragment purification, exclusion of parental-type recombinants, and simple phage screening techniques permits the isolation of virtually any gene segment for which there is an identifying hybridization probe. We illustrate the approach by describing the cloning of an approximately 7000-base-long segment of mouse DNA containing globin and surrounding gene sequences.

Integrated Moloney murine sarcoma provirus (MSV) has direct terminal repeat sequences (TRS). We determined the nucleotide sequence of both 588-base-pair TRS elements and the adjacent host and viral junctions of an integrated MSV cloned in bacteriophage lambda. Sequences were identified corresponding to the tRNAPro primer binding site in genomic RNA and the reverse-transcribed minus strong stop DNA. Each 588-base-pair repeat contains putative sites for promoting RNA synthesis and RNA polyadenylylation. The first and last 11 nucleotides of the TRS are inverted with respect to each other, and the same four-nucleotide host sequence is found bracketing integrated MSV. Some similarities of TRS and prokaryotic insertion sequence elements are discussed.

Transneuronal transport of pseudorabies virus (PRV) from the retina to visual centers that mediate visual discrimination and reflexes requires specific genes in the unique short region of the PRV genome. In contrast, these same viral genes are not required to infect retinorecipient areas of the brain involved in circadian rhythm regulation. In this report, we demonstrate that viral mutants carrying defined deletions of the genes encoding glycoprotein gI or gp63, or both, result in the same dramatic transport defect. Efficient export of either gI or gp63 from the endoplasmic reticulum to the Golgi apparatus in a fibroblast cell line requires the presence of both proteins. We also show that gI and gp63 physically interact, as demonstrated by pulse-chase and sucrose gradient sedimentation experiments. Complex formation is rapid compared with homodimerization of PRV glycoprotein gII. We suggest that gI and gp63 function in concert to affect neurotropism in the rat visual circuitry and that a heterodimer is likely to be the unit of function.

The ICP18.5 gene (UL28) of herpes simplex virus type 1 is a member of a well-conserved gene family among herpesviruses and is thought to play a role in localization of viral glycoproteins. We have cloned, sequenced, and expressed the entire pseudorabies virus (PRV) ICP18.5 open reading frame in Escherichia coli as a Cro-ICP18.5 fusion protein. Rabbit antiserum against Cro-ICP18.5 immunoprecipitated a 79-kDa protein from PRV-infected cells as well as a 79-kDa protein from in vitro translation of a T7 RNA polymerase transcript of the ICP18.5 gene. ICP18.5 could be detected in infected cells by 2 h postinfection. Analysis by indirect immunofluorescence demonstrated that ICP18.5 became associated with the nucleus. Subcellular fractionation confirmed that ICP18.5 synthesized during a pulse-chase experiment appeared in the nuclear fraction with time and was stable for at least 2.5 h after synthesis. Pulse-chase analysis revealed that ICP18.5 was synthesized as a monomer during a 2-min pulse labeling but formed faster sedimenting complexes which were sensitive to sodium dodecyl sulfate (SDS) treatment. The majority of ICP18.5 appeared in complexes with an antigenically unrelated 70-kDa protein. Immunoblot analysis of total infected-cell extracts using polyvalent anti-ICP18.5 serum demonstrated that a 74-kDa cellular protein in addition to the 79-kDa ICP18.5 was detected. This cellular protein was present at similar levels in uninfected cells and in PRV-infected cells at least 12 h into the infectious cycle.

The pseudorabies virus gII gene shares significant homology with the gB gene of herpes simplex virus type 1. Unlike gB, however, gII is processed by specific protease cleavage events after the synthesis of its precursor. The processed forms are maintained in an oligomeric complex that includes disulfide linkages. In this report, we demonstrate the kinetics of modification, complex formation, and subsequent protease processing. In particular, we suggest that gII oligomer formation in the endoplasmic reticulum is an integral part of the export pathway and that protease cleavage occurs only after oligomers have formed. Furthermore, through the use of glycoprotein gene fusions between the gIII glycoprotein and the gII glycoprotein genes of pseudorabies virus, we have mapped a functional cleavage domain of gII to an 11-amino-acid segment.

Pseudorabies virus (PRV) vaccine strain Bartha has a diminished capacity to cause disease and harbors a variety of mutations affecting virulence. It has been reported that PRV Bartha produces virions with reduced amounts of the major envelope glycoprotein gIII. One hypothesis was that this phenotype was due to reduced expression of the gIII gene. In this report, we demonstrate that the reduced amount of gIII in virions was not mediated at the level of transcription, but rather reflected a defect in protein localization. We describe experiments with gene replacement technology to prove that the expression defect was closely linked to the gIII gene itself. Using pulse-chase experiments, we found a defect similar to that observed for certain signal sequence mutations of PRV Becker gIII. The Bartha gIII protein was translated, but was inefficiently introduced into the membrane protein export pathway. Consequently, only a fraction of the primary Bartha gIII translation product was glycosylated and matured. The remaining fraction stayed presumably in the cytoplasm, where it never became glycosylated or inserted into cell or virus membranes. The result was that Bartha-infected cells produced virions with reduced amounts of gIII in their envelopes. Comparison of the DNA sequence of the promoter and amino-terminal coding regions of Becker and Bartha gIII genes revealed a single base pair difference in Bartha, changing codon 14 of the signal sequence from a leucine (CTC) to a proline (CCC) codon. We suggest that the signal sequence mutation is responsible for the apparent reduced expression phenotype of this attenuated strain. This mutation represents, to our knowledge, the first reported natural signal sequence mutation in a herpesvirus glycoprotein.

ABSTRACTWe describe use of developing chicken embryos as a model to study neuronal spread and virulence of pseudorabies virus (PRV). At embryonic day 12, ß-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 ß-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.

ABSTRACTThe Us9 gene is highly conserved among the alphaherpesviruses sequenced to date, yet its function remains unknown. In this report, we demonstrate that the pseudorabies virus (PRV) Us9 protein is present in infected cell lysates as several phosphorylated polypeptides ranging from 17 to 20 kDa. Synthesis is first detected at 6 h postinfection and is sensitive to the DNA synthesis inhibitor phosphonoacetic acid. Unlike the herpes simplex virus type 1 Us9 homolog, which was reported to be associated with nucleocapsids in the nuclei of infected cells (M. C. Frame, D. J. McGeoch, F. J. Rixon, A. C. Orr, and H. S. Marsden, Virology 150:321–332, 1986), PRV Us9 localizes to the secretory pathway (predominately to the Golgi apparatus) and not to the nucleus. By fusing the enhanced green fluorescent protein (EGFP) reporter molecule to the carboxy terminus of Us9, we demonstrated that Us9 not only is capable of targeting a Us9-EGFP fusion protein to the Golgi compartment but also is able to direct efficient incorporation of such chimeric molecules into infectious viral particles. Moreover, through protease digestion experiments with Us9-EGFP-containing viral particles, we demonstrated that the Us9 protein is inserted into the viral envelope as a type II, tail-anchored membrane protein.

ABSTRACTSeveral groups have reported that certain herpesvirus envelope proteins do not remain on the surface of cells that express them but rather are internalized by endocytosis in a recycling process. The biological function of membrane protein endocytosis in the virus life cycle remains a matter of speculation and debate. In this report, we demonstrate that some, but not all, membrane proteins encoded by the alphaherpesvirus pseudorabies virus (PRV) are internalized after reaching the plasma membrane. Glycoproteins gE and gB are internalized from the plasma membrane of cells, while gI and gC are not internalized efficiently. We show for gE that the cytoplasmic domain of the protein is required for endocytosis. While the gI protein is incapable of endocytosis on its own, it can be internalized when complexed with gE. We demonstrate that endocytosis of the gE-gI complex and gB occurs early after infection of tissue culture cells but that this process stops completely after 6 h of infection, a time that correlates with significant shutoff of host protein synthesis. We also show that gE protein internalized at 4 h postinfection is not present in virions formed at a later time. We discuss the differences in PRV gE and gI endocytosis compared to that of the varicella-zoster virus homologs and the possible roles of glycoprotein endocytosis in the virus life cycle.

Glycoprotein E (gE) gene of pseudorabies virus (PRV) is conserved among diverse alphaherpesviruses and therefore is predicted to be important for virus survival. gE contributes to viral spread from cell to cell in a variety of hosts and is responsible, in part, for increased virulence or pathogenesis of the virus. Virulence and spread mediated by gE are thought to be highly correlated. We initiated this study to explore the hypothesis that these two phenotypes might reflect separate functions of the gE protein. We did so by focusing on the role of the gE carboxy terminus in neuronal spread. Viruses harboring nonsense mutations affecting the expression of the gE cytoplasmic domain had several notable phenotypes. First, the truncated gE proteins expressed from these mutants are not found in virion envelopes. Second, the mutants retain the ability to spread to all retinorecipient regions of the rodent brain after retinal infection of rats. Third, the mutants have the reduced virulence phenotype of a gE deletion mutant in rats. Finally, the mutants have distinct plaque-size phenotypes on MDBK cells but not PK15 cells. Based on these observations, we suggest that gE-mediated virulence and spread may reflect separate functions that are not mediated by gE on virus particles.

The membrane glycoproteins gE and gI are encoded by pseudorabies virus (PRV), a neurotropic, broad-host-range alphaherpesvirus of swine. PRV gE and gI are required for anterograde spread to a restricted set of retinorecipient neurons in the brain after infection of the rat retina. A related alphaherpesvirus, encoding gE and gI homologs, is called bovine herpesvirus 1.1 (BHV-1.1). BHV-1.1 is a respiratory pathogen of highly restricted host range and, in contrast to PRV, is unable to propagate in or cause disease in rodents. We have shown previously that the BHV-1.1 gE and gI proteins are capable of complementing the virulence functions of PRV gE and gI in a rodent model (A. C. Knapp and L. W. Enquist, J. Virol. 71:2731-2739, 1997). We examined the ability of the BHV-1.1 gE and gI homologs to direct circuit-specific invasion of the rat central nervous system by PRV. Both complete open reading frames were cloned into a PRV mutant lacking the PRV gE and gI genes. Recombinant viruses were analyzed for the ability to invade the rat brain after infection of the retina. Surprisingly, in a portion of the animals tested, the BHV-1.1 gE and gI proteins functioned autonomously to promote spread of PRV to a subset of retinorecipient regions of the brain. First, the presence of BHV-1.1 gI alone, but not PRV gI alone, promoted viral invasion of the optic tectum. Second, expression of BHV-1.1 gE alone facilitated PRV infection of a subset of neurons in the hippocampus not normally infected by PRV. When both BHV-1.1 proteins were expressed in a coinfection, all retinorecipient regions of the rat brain were infected. Therefore, depending on the viral source, homologs of gE and gI differentially affect spread between synaptically connected neurons in the rat.

This study reports the identification and initial characterization of the precursors, modified forms, and oligomers of bovine herpesvirus 1 (BHV-1) gI and gE proteins with polyvalent rabbit serum specific for gI or gE. Our experiments used the Colorado strain of BHV-1 and mutant viruses with insertions of the Escherichia coli lacZ gene into the predicted gE and gI reading frames. We also translated the gE and gI open reading frames in vitro and expressed them in uninfected cells using eukaryotic expression vectors. Precursor-product relationships were established by pulse-chase analysis and endoglycosidase H and glycopeptidase F digestions. Like the homologous glycoproteins of herpes simplex virus type 1, pseudorabies virus, and varicella-zoster virus, BHV-1 gI and gE are modified by N-linked glycosylation and associate with each other soon after synthesis, forming a noncovalent complex in infected and transfected cells. An analysis of mutant and wild-type-virus-infected cells and transfected COS cells expressing gE or gI alone suggested that gE-gI complex formation is necessary for efficient processing of the gE precursor to its mature form. One new finding was that unlike the other alphaherpesvirus gI homologs, a fraction of pulse-labeled gI synthesized in BHV-1-infected cells apparently is cleaved into two relatively stable fragments 2 to 4 h after the pulse. Finally, we incubated BHV-1-infected cell extracts with nonimmune mouse, rabbit, horse, pig, and calf sera and found no evidence that gE or gI functioned as Fc receptors as reported for the herpes simplex virus type 1 and varicella-zoster virus homologs.

The transmembrane and anchor region of pseudorabies virus gIII is postulated to be in the 35 hydrophobic amino acids (residues 436 to 470) found near the carboxy terminus of the 479-amino-acid envelope protein. In this study, we used a genetic approach to localize the functional gIII membrane anchor between amino acids 443 and 466. Mutant gIII proteins lacking the membrane anchor were not associated with virus particles, indicating that membrane retention is a prerequisite for virion localization. Unexpectedly, the specific hydrophobic gIII sequence defined by these deletions was not required for membrane anchor function since the entire region could be replaced with leucine residues without affecting gIII membrane retention, export, or virion localization. The hydrophobic region appears to encode more than the membrane anchor domain since both efficiency of posttranslational processing and localization to virions are affected by mutations in this region. We speculate that the composition of the hydrophobic domain influences the overall conformation of gIII, which in turn effects the efficiency of gIII export and processing. The virion localization phenotype is probably indirect and reflects the efficiency of protein processing. This conclusion provides insight into the mechanism of glycoprotein incorporation into virions.

The entry of herpesviruses into cells involves two distinct stages: attachment or adsorption to the cell surface followed by internalization. The virus envelope glycoproteins have been implicated in both stages. Pseudorabies virus attaches to cells by an early interaction that involves the viral glycoprotein gIII and a cellular heparinlike substance. We examined the role of gIII in the attachment process by analysis of a set of viruses carrying defined gIII mutations. The initial attachment of gIII mutants with an internal deletion of 134 amino acids (PrV2) to MDBK cells was indistinguishable from that of wild-type virus. The adsorption of these mutants was, however, much more sensitive than that of wild-type virus to competing heparin. Furthermore, while attachment of wild-type virus to MDBK cells led to a rapid loss of sensitivity to heparin, this was not the case with PrV2, which could be displaced from the cell surface by heparin after it had attached to the cells. We conclude that glycoprotein gIII is involved in two distinct steps of virus attachment and that the second of these steps but not the first is defective in PrV2.

Glycoprotein gIII of pseudorabies virus is a member of a conserved gene family found in at least seven diverse herpesviruses. We report here that the putative cytoplasmic domain of gIII is not required for transport to the cell surface and, unlike the prototype domain from herpes simplex virus type 1 glycoprotein C, is not required for stable membrane anchoring. Furthermore, this domain does not appear to be essential for incorporation of the glycoprotein into virions.

We have constructed two pseudorabies virus (PRV) mutants, each with a unique EcoRI restriction site in the nonessential gIII envelope glycoprotein gene. Since no natural PRV isolate has been reported to contain EcoRI sites, the isolation and single-step growth curve analysis of these mutants established that PRV can carry such a site with little ill effect in tissue culture. Virus carrying these defined mutations produced novel gIII proteins that enabled us to begin functional assignment of protein localization information within the gIII gene. Specifically, one viral mutant contained an in-frame synthetic EcoRI linker sequence that was flanked on one side by the first one-third of the gIII gene and on the other side by the last one-third of the gene. The resulting protein lacked the middle one-third of the parental species, including five of eight putative N-linked glycosylation signals, but was still glycosylated and found in enveloped virions; it was not secreted into the medium. A second viral mutant contained an in-frame synthetic EcoRI linker sequence that additionally specified a nonsense codon at position 158, producing a gIII protein that was glycosylated and secreted into the medium; the fragment was not found in enveloped virions. By endoglycosidase and pulse-chase analyses, we established a precursor-product relationship between the various forms of gIII expressed in the parental and mutant strains, and perhaps determined certain features of the gIII protein that are required for its efficient export within the cell.

A pseudorabies virus (Becker strain) glycoprotein gene was located in the UL region at map position 0.40. The gene was identified by using open reading frame Escherichia coli plasmid expression vectors and specific antibody reagents. A 1.55-kilobase unspliced transcript from the gene was detected in pseudorabies virus-infected tissue culture cells. The DNA sequence revealed a single open reading frame of 1,437 base pairs encoding 479 amino acids. The predicted primary translation product has a molecular weight of 50,860 and contains features of a typical herpesvirus glycoprotein. An E. coli expression plasmid was constructed that contained essentially all of the open reading frame for this gene. Antibodies raised in rabbits against the protein expressed in bacteria by this plasmid immunoprecipitated pseudorabies virus-specific glycoproteins of 92,000 and 74,000 daltons from infected cell extracts. It is likely that these two forms represent different glycosylation states of the protein.

In this work we used brefeldin A (BFA), a specific inhibitor of export to the Golgi apparatus, to study pseudorabies virus viral glycoprotein processing and virus egress. BFA had little effect on initial synthesis and cotranslational modification of viral glycoproteins in the endoplasmic reticulum (ER), but it disrupted subsequent glycoprotein maturation and export. Additionally, single-step growth experiments demonstrated that after the addition of BFA, accumulation of infectious virus stopped abruptly. BFA interruption of virus egress was reversible. Electron microscopic analysis of infected cells demonstrated BFA-induced disappearance of the Golgi apparatus accompanied by a dramatic accumulation of enveloped virions between the inner and outer nuclear membranes and also in the ER. Large numbers of envelope-free capsids were also present in the cytoplasm of all samples. In control samples, these capsids were preferentially associated with the forming face of Golgi bodies and acquired a membrane envelope derived from the trans-cisternae. Our results are consistent with a multistep pathway for envelopment of pseudorabies virus that involves initial acquisition of a membrane by budding of capsids through the inner leaf of the nuclear envelope followed by deenvelopment and release of these capsids from the ER into the cytoplasm in proximity to the trans-Golgi. The released capsids then acquire a bilaminar double envelope containing mature viral glycoproteins at the trans-Golgi. The resulting double-membraned virus is transported to the plasma membrane, where membrane fusion releases a mature, enveloped virus particle from the cell.

gIII, the major envelope glycoprotein of pseudorabies virus (PRV), shares approximately 20% amino acid similarity with glycoprotein gC of herpes simplex virus type 1 (HSV-1) and HSV-2. We describe here our first experiments on the potential conservation of function between these two genes and gene products. We constructed PRV recombinants in which the gIII gene and regulatory sequences have been replaced with the entire HSV-1 gC gene and its regulatory sequences. The gC promoter functions in the PRV genome, and authentic HSV-1 gC protein is produced, albeit at a low level, in infected cells. The gC protein is present at the cell surface but cannot be detected in the PRV envelope.

We describe experiments using the swine herpesvirus, pseudorabies virus (PRV), as a vector for expression of hybrid membrane protein genes. In particular, we present the construction and analysis of three infectious PRV mutants expressing chimeric viral membrane proteins composed of portions of the PRV envelope glycoprotein gIII and of the human retrovirus, human immunodeficiency virus type 1 (HIV-1), envelope glycoproteins gp120 and gp41. All of the chimeric genes contain the transcription control sequences and the first 157 codons of PRV gIII (known to contain signals sufficient for efficient export of the encoded peptide out of the cell) fused to different regions of the HIV-1 envelope. The mutant viruses express novel glycosylated fusion proteins that are immunoprecipitated by polyvalent sera specific for gIII, as well as acquired immunodeficiency syndrome patient sera. The levels of expression are lower than expected due primarily to instability or altered processing of the hybrid mRNA. We could not detect cleavage of chimeric proteins carrying the gp120-gp41 protease processing site. The use of localization signals contained within herpesvirus membrane proteins to direct chimeric proteins to desired cellular locations is discussed.

We have constructed a pseudorabies virus mutant that contains virtually a complete deletion of the predicted signal sequence coding region for a nonessential envelope glycoprotein, gIII. No signal sequence mutants have been reported previously for a herpesvirus glycoprotein. Through endoglycosidase treatments and pulse-chase analysis, we have determined that the mutant gIII protein is not posttranslationally modified like the wild-type polypeptide, but rather is present as a single, stable species within the infected cell. The mutant polypeptide cannot be detected in the virus envelope, nor is it aberrantly localized to the tissue culture medium. Immunofluorescence studies have indicated that the mutant protein also is not localized to the surfaces of infected cells. In addition, Northern (RNA) and slot blot analyses, as well as in vitro translation experiments using infected-cell cytoplasmic RNA, have indicated that the mutant gIII allele is expressed at lower levels than the wild-type gene is. This is despite the fact that no alterations have been made upstream of the gIII coding sequence. From these results, it appears that the first 22 amino acids of the wild-type gIII protein define a necessary signal peptide that is responsible for at least the correct initiation of translocation and subsequent glycosylation of the gIII envelope glycoprotein within infected cells.

Glycoprotein gIII of pseudorabies virus is a major antigen found in the envelopes of virus particles as well as in and on the surfaces of infected cells. It is not an essential gene product for virus growth in tissue culture. In this report, we provide evidence that, although it is not essential, the gIII protein is required for efficient virus growth and that gIII mutants are quickly outgrown by wild-type virus in mixed infections.

We have established that in the Becker strain of pseudorabies virus (PRV), the glycoprotein gIII gene is not essential for growth in cell culture. This was accomplished by construction and analysis of viral mutants containing two defined deletion mutations affecting the gIII gene. These mutations were first constructed in vitro and introduced into Escherichia coli expression plasmids to verify structure and protein production. Each mutation was then crossed onto PRV by cotransfection of plasmid DNA and parental viral DNA by using gIII-specific monoclonal antibodies as selective and screening reagents. One resultant virus strain, PRV-2, contained an in-frame deletion of a 402-base-pair (bp) SacI fragment contained within the gIII gene. Another virus strain, PRV-10, contained a deletion of a 1,480-bp XhoI fragment removing 230 bp of the upstream, putative transcriptional control sequences and 87% of the gIII coding sequence. The deletion mutants were compared with parental virus by analysis of virion DNA, gIII specific RNA, and proteins reacting with gIII specific antibodies. Upon infection of PK15 cells, the deletion mutants did not produce any proteins that reacted with two gIII specific monoclonal antibodies. However, two species of truncated glycosylated proteins were observed in PRV-2 infected cells that reacted with antiserum raised against bacterially produced gIII protein. PRV-10 produced no detectable gIII-specific RNA or protein. PRV-10 could be propagated without difficulty in tissue culture. Virus particles lacking gIII were indistinguishable from parental PRV virus particles by analysis of infected-cell thin sections in the electron microscope. We therefore conclude that expression of the gIII gene was not absolutely essential for PRV growth in tissue culture.

Two naturally occurring variations of herpes simplex virus type 1 (Patton strain) with novel tandem DNA sequence duplications in the S component were isolated, and the DNA was characterized. These variants were identified among a number of plaque isolates by the appearance of new restriction enzyme fragments that hybridized with radiolabeled DNA from the BamHI Z fragment (map coordinates 0.936 to 0.949) located in the unique S region. One isolate, SP26-3, carried a 3.1-kilobase-pair duplication defined by recombination between a site in the BamHI Z fragment and a site near the origin of replication in the inverted repeat sequence of the S component carried by the EcoRI H fragment. The other isolate, SP22-4, carried a 3.5-kilobase-pair duplication defined by a recombination event between a tandem repeat array in the BamHI Z fragment and a site near the amino terminus of the Vmw175 gene in the S-region inverted repeat sequence contained in the EcoRI K fragment. Both duplicated segments contained the entire immediate early mRNA-5 coding region as well as the origin of replication located in the inverted repeat sequence of the S component. The DNA sequence of each duplication joint was determined.

Stocks of hybrid lambda phages carrying the complete integrated provirus of either m1 or HT1 Moloney murine sarcoma virus, as well as flanking host sequences, frequently contain significant numbers of phages carrying a specific deletion. This deletion arises from a recombination event between the terminally repeated sequences in the provirus that deletes the unique Moloney murine sarcoma virus sequences bracketed by the terminally repeated sequences. Physical mapping has shown that the deletion phage retains one complete copy of the terminally repeated sequence and the flanking mink host sequences. One such deletion, lambdaHT1r+, was used to characterize a mink genomic DNA sequence that contains an HT1 Moloney murine sarcoma virus integration site. This integration site sequence from normal mink cells was also cloned into phage lambda. An analysis of the heteroduplexes between the integration site and the lambdaHT1r+ deletion indicated that no major rearrangement of host sequences occurred upon integration of the Moloney murine sarcoma provirus.

A combination of mutations in bacteriophage ? and its host Escherichia coliK-12 provides a convenient system for the isolation of large quantities of covalently closed circular DNA molecules. We describe two procedures for the large scale preparation of ? DNA in the duplex circular form.

We have combined techniques of genetic and physical mapping with rapid DNA sequence analysis to identify the nucleotide change in lambda int mutations. These mutations define two dominant phenotypic classes: (i) recombination that is partially independent of accessory factors, and (ii) inhibition of wild-type Int by missense or nonsense proteins, i.e., negative complementation.