![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Department of Molecular Biology, Princeton University, Princeton, New Jersey 08540
SUMMARY INTRODUCTION Virus Nomenclature Herpesviridae and Alphaherpesvirinae MOLECULAR BIOLOGY OF PRV Virion Structure Genome and Gene Content Transcriptional architecture. Core genes. Capsid Proteins Tegument Proteins Tegument complexity. Envelope Proteins Glycoprotein nomenclature. Glycoproteins and endocytosis. Viral Replication Cycle Entry Herpesvirus entry mediators, cellular receptors for PRV entry. Regulation of viral glycoprotein-induced membrane fusion. Viral Transcription Activators and the Transcription Cascade IE180. EP0. Other regulators of gene expression: UL54, UL41, and UL48. Host Transcript Changes during PRV Infection Gene Expression during Latency DNA Replication Nucleotide Metabolism Capsid Formation Cyclooxygenases and capsid assembly DNA Encapsidation Egress Nuclear egress and primary envelopment. (i) US3 promotes nuclear egress Tegumentation and secondary envelopment L-particles. Functional redundancy and egress. Viral Antiapoptosis Genes US3 antiapoptosis activity TAP Inhibition by UL49.5 (gN) PRV AS A MODEL ORGANISM PRV Pathogenesis in Animal Models Rat Eye Model: Genes Required for Neuroinvasion and Neurovirulence Chicken Embryo Eye Model Mouse Skin Flank Model: New Facets of PRV Pathogenesis Two Modes of PRV Lethality Mouse Intranasal Infection Model Axonal Targeting, Transport, and Assembly of PRV in Cultured Neurons Models for Reactivation from Latency PRV AS A TRANSSYNAPTIC TRACER Viral Tracers versus Nonviral Tracers Transsynaptic Spread of PRV in the Nervous System Circuit tracing by PRV faithfully reproduces tracing deduced by well-characterized, nonviral tracers. Infection does not spread intraaxonally to synaptically unconnected but physically adjacent circuitry. Spread in the nervous system requires an intact circuit and synaptic connections. Containment of Neuronal Infection by Cytoarchitecture and Nonpermissive Cells Attenuated Strains Make Good Tracing Viruses Analyzing the Spread of PRV-Bartha in the Rat Eye Model Virulence and Neural Tracing Recent Advances in Viral Tracing Techniques Dual tracing in circuitry analysis. Electrophysiological recording of live neurons infected by a PRV tracer. Conditional replication of PRV in specific neuronal populations. Potential for monodirectional anterograde tracing with PRV. VETERINARY IMPACT OF PSEUDORABIES (AUJESZKY'S DISEASE) Historical Discovery Lethality of Pseudorabies Infection in Nonnative Hosts Acute PRV Infection of Swine PRV Latency in Swine Modified Live Vaccines DNA Vaccines Diagnostic Tests for PRV Worldwide Distribution of Pseudorabies Economic Impact of PRV in the United States Efforts to Eradicate PRV in the United States Feral Swine: Reservoirs of Sexually Transmitted PRV in the United States OUTLOOK ACKNOWLEDGMENTS REFERENCES
| SUMMARY |
|---|
 Top NextReferences |
|---|
| INTRODUCTION |
|---|
|
TopPreviousNextReferences |
|---|
This review focuses on recent reports regarding the molecular biology of PRV, the use of laboratory animal models to study viral pathogenesis, the use of PRV as a neuronal tracer, and the agricultural impact of PRV. The broad coverage of this review targets not only virologists, but also those interested in neurobiology and veterinary medicine.
The three major subfamilies differ in the cell type where latency is established and the length of their productive replication cycle (348). Alphaherpesviruses have the broadest host range, tend to replicate rapidly with cytopathic effects to produce viral particles in a matter of hours, and can establish latency in the sensory ganglia. Betaherpesviruses have the most restricted host range and the slowest rate of replication that is often accompanied by cell enlargement (cytomegalia), and establish latency in a number of tissues and cells, including secretory glands, kidneys, and lymphoreticular cells. Gammaherpesviruses infect lymphoblastoid cells and are usually specific for either T or B lymphocytes, establishing latency in lymphoid tissue.
Humans harbor three alphaherpesviruses: varicella-zoster virus (VZV) and herpes simplex virus type 1 (HSV-1) and type 2 (HSV-2). The monkey B virus can also be transmitted to humans with lethal consequences. Despite its significant homology to human alphaherpesviruses and its broad host range, PRV is not transmitted to humans. Anecdotal reports of rare human PRV infections have not been substantiated, and likely reflect the low cross-reactivity of antibodies against HSV-1 gB to PRV gB (344, 437). Owing to the significant homology between members of the Alphaherpesvirinae, information derived from the study of PRV provides a powerful opportunity for comparative molecular virology (118). Accordingly, research interests in PRV reflect more than the agricultural impact of the disease it causes. Other well-studied alphaherpesviruses are animal pathogens important to agriculture, including bovine herpesviruses (BHV-1 and BHV-5), equine herpesviruses (EHV-1 and EHV-4), ovine herpesvirus, and avian herpesviruses such as Marek's disease virus and infectious laryngotracheitis virus (ILTV) (348).
Based on molecular criteria and sequence analysis, the Alphaherpesvirinae subfamily can be subdivided further into the genera Simplexvirus (HSV-1), Varicellovirus (VZV), Marek's disease-like virus, and ILTV-like virus (265, 346, 348). PRV, and its closely related homologs BHV-1, EHV-1 and EHV-4, feline herpesvirus type 1, and canine herpesvirus type 1, are all members of the Varicellovirus genus.
| MOLECULAR BIOLOGY OF PRV |
|---|
|
TopPreviousNextReferences |
|---|
|
|
|
The genome of PRV is largely colinear with those of HSV-1 and other alphaherpesviruses, except for a large internal inversion in the UL region situated between UL46 and UL26.5 (32, 46, 110). PRV genes ORF1, ORF1.2, and UL3.5 are not found in HSV-1, while at least 16 HSV-1 genes are not found in PRV (219). Three origins of replication have been mapped in PRV: OriL, located in the UL region (Fig. 2), and two copies of OriS, each located in the inverted repeats (132, 440). The PRV origins of replication are structured as two inverted binding sites for the origin-binding protein UL9 (GTTCGCAC), separated by a 43-bp A-T-rich segment (76% A+T). This basic structure is found once in OriL, while three imperfect repeats of this basic arrangement constitute OriS. Likewise, HSV-1 contains two copies of oriS and one of oriL, consisting of palindromes of 45 bp and 144 bp, respectively, centered around an A-T-rich region of 18 bp (oriS) or 20 bp (oriL). The HSV-1 A-T regions are also flanked by one or two inverted binding sites for UL9 that vary in binding affinity (37, 241, 347).
Transcriptional architecture. Many features of the PRV transcriptional architecture (Fig. 2) are conserved in the related alphaherpesviruses HSV-1 and BHV-1 (219). Many of the same genes form families of 3'-coterminal transcripts (266, 219, 331). The few genes found to be spliced in alphaherpesviruses are usually immediate-early genes or latency transcripts. PRV contains two known spliced transcripts (US1 and the Large Latency Transcript, LLT) and a putative spliced transcript (UL15), and all three homologs are spliced in HSV-1 (Fig. 2) (347, 219). Many of the core transcription elements are predicted to be shared between genes, with TATA boxes initiating divergent transcripts, or TATA boxes also functioning as polyadenylation signals for genes upstream. These features may be related to the high gene density and limited intergenic sequences found in alphaherpesvirus genomes, such as PRV. PRV also contains multiple short DNA repeat elements, often located between converging transcripts (Fig. 2), and these may serve to prevent transcription from one gene into an oppositely transcribed gene.
Core genes. A set of 40 herpesvirus genes are conserved among all Alpha-, Beta-, and Gammaherpesvirinae (Fig. 2). These "core genes" encode proteins that perform steps fundamental to the replication of herpesviruses, in part, because of the common structure of nucleocapsids, the basic requirements for viral DNA replication and packaging, and the shared steps of entry into and egress from cells. Phylogenetic analysis of mammalian and avian herpesvirus genomes suggests that an ancestral virus contributed the 40 core genes to modern alpha-, beta-, and gammaherpesviruses (100). While many of the core genes show sequence conservation within the Herpesviridae, some share only the relative genome position and protein function. Table 1 notes the PRV core genes. Like those of other herpesviruses, all PRV core genes are found in the UL region (Fig. 2).
Tegument complexity. The origin and evolution of the tegument components remain enigmatic, as many of these proteins share almost no sequence homology between alpha-, beta-, and gammaherpesviruses, and most are dispensable for viral growth in cultured cells. The tegument is composed of at least two distinct structures: an inner layer that is tightly associated with capsid proteins and an outer layer that is asymmetrically organized, heterogeneous, and interacts with the cytoplasmic domains of viral membrane proteins. Examination of purified virions reveals variability in the amount of VP22-GFP incorporated into individual particles (107). Similar findings were reported for other tegument proteins (249). Cryoelectron microscopy and tomography revealed that the tegument of HSV-1 virions was pleimorphic, forming an asymmetric cap that occupied two thirds of the volume enclosed within the envelope (159). The innermost layer of tegument shows a more ordered icosahedral morphology, probably imparted by UL36 (VP1/2), an essential tegument protein that may associate with the major capsid protein UL19 (VP5), and UL37, a tegument protein found to interact with UL36 in PRV (136, 213, 249, 272, 452).
Glycoprotein nomenclature. The standard nomenclature of PRV and HSV envelope glycoproteins was adopted at the 18th International Herpesvirus Workshop in 1993. Most reports published before 1995 refer to PRV glycoproteins gB, gC, gD, gE, gG, and gI as gII, gIII, gp50, gI, gX, and gp63, respectively.
Glycoproteins and endocytosis. Newly synthesized glycoproteins travel from the endoplasmic reticulum via the Golgi to the plasma membrane. However, several PRV envelope glycoproteins are subsequently internalized, either spontaneously or in response to binding of antigen-specific antibodies (125, 126, 304, 401, ). Antibody-mediated internalization of viral proteins from the cell surface may modulate the immune response and protect PRV-infected monocytes from efficient antibody-dependent, complement-mediated lysis (413). The contribution of antibody-independent, glycoprotein endocytosis to a successful viral life cycle is uncertain, but has been proposed to play a role in immune modulation, delivery of viral cell surface proteins to the intracellular compartment where viral envelopment takes place, and redirection of viral proteins to specific membrane surfaces (such as the apical, lateral, or basal surfaces of polarized cells) to facilitate cell to cell spread (reviewed in reference 49). The internalization of many alphaherpesvirus envelope proteins is mediated by tyrosine-based (YXXØ) or dileucine-based (LL) endocytosis motifs located in their cytoplasmic domain (305, 126, 94) (reviewed in reference 49). In addition, acidic clusters containing phosphorylation sites important in endocytosis of cell surface molecules can also occasionally be found. These motifs are known clathrin-mediated endocytosis motifs used by cellular receptors. Indeed, the internalization of PRV gB is mediated by the interaction between its endocytosis motif and the cellular clathrin-associated AP-2 adaptor complex (415).
|
Herpesvirus entry mediators, cellular receptors for PRV entry. Five cellular gD receptors, also known collectively as herpesvirus entry mediators, have been identified based on their capacity to allow entry of herpes simplex viruses into cells: HveA (TNFRSF14), HveB (PRR2, nectin 2), HveC (PRR1, nectin 1), HveD (PVR, CD55), and 3-O-sulfated heparan sulfate (reviewed in references 275, 377, and 378). Human HveB, HveC,and HevD, but not HveA or 3-O-sulfated heparan sulfate, could mediate entry of PRV into CHO cells, with HveC being the most effective (306). HveC encodes nectin 1, a protein important for cell adhesion, and the mammalian HveC homologs are found to be highly conserved: the porcine homolog shares 96% amino acid identity with the human homolog (284). Both human and porcine HveC can mediate the entry of HSV-1, HSV-2, PRV, and BHV-1 (88, 146, 284). HveC was also found as the primary receptor for HSV-1 infection in rat and mouse sensory neurons (337).
Regulation of viral glycoprotein-induced membrane fusion. Transfected cells expressing PRV gB, gH, and gL can induce cell fusion in the absence of gD (221). In contrast, all four proteins are required for HSV1 glycoprotein-induced cell fusion (410). HSV-1 and PRV gD differ in glycosylation (PRV gD lacks the N-glycosylation sites found in HSV-1 and is instead only O-glycosylated [90, 321] and their role in cell-to-cell spread: gD is required for PRV penetration (Fig. 3-2) but not for cell-to-cell spread (316), while gD is required for both processes in HSV-1. The transfection-based assay serves as a model for membrane protein involvement in cell-cell spread. A number of PRV membrane proteins inhibit fusion when cotransfected with fusogenic glycoproteins: gM, UL43, or the combination of gK and UL20, all significantly reduce cell fusion (212, 221).
The fusion inhibition function of gM seems to be conserved among Herpesviridae. Like gM homologs in other herpesviruses, PRV gM forms a disulfide bond with the product of the UL49.5 gene, gN (2, 195, 231, 251, 238, 444). Yet, PRV gM is capable of inhibiting fusion in the absence of gN while the HSV-1, EHV-1, ILTV, BHV-1 and HHV8 homologs inhibit membrane fusion only when coexpressed with their UL49.5 homologs (94, 221, 223, 231). PRV gM inhibits fusion by internalizing viral glycoproteins from the plasma membrane and redirecting them to the Golgi apparatus (94). Likewise, the amount of cell surface viral glycoproteins is reduced when HSV-1 gK and UL20 are expressed together (11). PRV UL20 is required for the maturation of gK, and both proteins are required to inhibit fusion in the transfection based assay (111, 212). Inhibition of membrane fusion between infected and uninfected cells may promote efficient cell-cell spread of virus, while the internalization of fusogenic glycoproteins may direct them to sites of secondary envelopment.
Earlier studies in rabbit kidney cells have described the general appearance kinetics of PRV-encoded transcripts and proteins (29). The length of time required to complete the PRV growth cycle varies according to cell types, and in the most commonly used cell types, viral progeny can be detected within 4 to 5 h. The immediate-early IE180 transcript is synthesized within 40 min of infection and the IE180 protein is synthesized up until 2.5 h postinfection. Early transcripts appear around 1 hour postinfection, with transcript levels peaking around 3 to 4 hours postinfection (early mRNAs) or later (early-late mRNAs). Early proteins are synthesized most abundantly between 1 and 4 hours postinfection, prior to and during the early stages of DNA replication. The products of early-late mRNAs appear around 1.5 hours postinfection but their synthesis peaks at later times (between 4 and 9 hours postinfection). Finally, late mRNAs are detected as early as 2.5 hours postinfection (when viral DNA synthesis starts), and their protein products are detectable around 3 hours postinfection, progressively accumulating to high levels thereafter.
IE180. PRV encodes only one genuine immediate-early gene, IE180 (ICP4 homolog). In contrast, most herpesvirus genomes express several immediate-early proteins. For example, HSV-1 encodes at least five: RL2 (ICP0), UL54 (ICP27), RS1 (ICP4), US1 (ICP22), and US12 (ICP47). PRV does not encode an ICP47 (US12) homolog, because the PRV genome lacks the DNA corresponding to HSV-1 US10, US11, and US12. Though the PRV genome lacks the IRL and TRL repeats found in HSV-1, an ICP0 homolog called EP0 is located in the PRV UL region. Both EP0 and the ICP27 homolog encoded by UL54 are expressed with early kinetics in PRV (25, 76, 182). PRV US1 mRNA accumulates in the presence of cycloheximide and may be an immediate-early gene but no other analysis has been done (132). The related VZV contains three immediate-early proteins, ORF4 protein (ORF4), IE62 (ORF62) and IE63 (ORF63), homologs of ICP27 (UL54), ICP4 (RS1), and ICP22 (US1), respectively (91). VZV IE62 serves as the major immediate-early transactivator of viral genes during lytic infection, stimulating the transcription of all VZV promoters tested (91).
The IE180 transcript is the predominant PRV transcript detected in cycloheximide-treated rabbit kidney cells, appearing as early as 30 min postinfection (127, 185). As expected, the IE180 promoter drives expression of a reporter gene in the absence of any viral protein synthesis or infection (58, 232). Unlike some of the HSV-1 immediate-early genes (ICP0, ICP22, and ICP47), the IE180 transcript is not spliced in infected bovine kidney cells (79). The IE180 protein is 1,460 amino acids long and contains two ICP4-like domains (amino acids 493 to 669 and amino acids 1052 to 1366, respectively) (78). Though predicted to have a molecular mass of 153 kDa, the product of the IE180 gene migrates as a 180-kDa protein during sodium dodecyl sulfate-polyacrylamide gel electrophoresis, is reportedly phosphorylated (86), and accumulates in the nuclei of infected cells (391, 446). A nuclear localization signal was mapped to the positively charged region (amino acids 930 to 935) RRKRR (391).
Like HSV-1 ICP4, the PRV IE180 gene is present in two copies in the genome, located in the IRS and TRS repeats. The gene is essential for viral replication in tissue culture, as it is required for the efficient transcription of early (and possibly late) viral genes (reviewed in reference 29). When subjected to the nonpermissive temperature, the temperature-sensitive IE180 mutant tsG1 arrests the infection at the immediate-early stage, expressing only IE180 RNA and IE180 protein (185). Recombinant PRV deleted for both copies of IE180 fails to synthesize viral products (445). Since most of the IE180 gene overlaps with the oppositely transcribed large latency transcript (LLT) (76), deletions in IE180 delete a portion of LLT as well. Studies using a recombinant PRV with altered IE180 promoters determined that infection initiation in cultured cells depends on the induction of IE180 (152). Finally, cells expressing a dominant negative form of IE180 that strongly represses the IE180 promoter support PRV replication very poorly, but allow normal replication of HSV-1 (309).
The role of IE180 as a potent transcriptional activator has been well established. Like most typical cellular activators, IE180 contains a separate domain for DNA-binding and another for trans-activation (258, 441). A strong acidic activation domain maps to the N terminus (amino acids 1 to 34) of IE180 (258). In vivo, IE180 has been shown to activate gene expression from the following PRV promoters: US4 (gG), UL12 (AN), UL22 (gH), UL23 (thymidine kinase), and UL41 (vhs) (73, 312, 389). IE180 has a dose-dependent effect on the UL41 promoter, activating gene expression at low levels, but inhibiting it at higher doses (73). It is not known whether this mechanism reflects IE180 negative autoregulation, where expression of IE180 decreases gene expression from its own promoter (420). IE180 can also activate transcription from cellular promoters such as human beta-globin and topoisomerase I (157, 439) and viral promoters such as adenovirus 2 early genes (187, 442), simian virus 40 early genes (157), and human immunodeficiency virus long terminal repeat (449). Partially purified IE180 protein can activate transcription initiation in vitro from human promoters beta-globin and hsp70 (157, 439), and IE180 may aid the formation of a stable transcription preinitiation complex by enhancing TFIID binding (1).
Like its homolog in VZV (ORF62) and HSV-1 (ICP4), the IE180 DNA-binding activity is located in the first ICP4-like domain (441). IE180 has been shown to bind both single-stranded (86) and double-stranded DNA. Partially purified IE180 exhibits specific DNA-binding activity at the promoters of the adenovirus major late gene, human hsp70, PRV US4 (gG), and at the PRV LAP1 promoter (93, 312, 313). In DNA protection assays, IE180 protected sites located near the transcription initiation site as well as sites upstream of the core promoter. The high affinity binding sites (25 to 28 nt long) protected in herpesvirus promoters share only a 5'-ATCGT-3' sequence (441), while the affinity sites in the adenovirus major late and human hsp70 promoters contained a near-identical 5'-CATCG-3'. The direct IE180 binding site at the PRV US4 promoter maps to a different sequence, the TEF-1 (transcription-enhancing factor 1) element, 5'-TGGAATGTG-3' (312). However, we note that consensus TEF-1 elements are found only once in the PRV genome. It is clear that IE180, like ICP4, recognizes highly degenerate or nonconsensus DNA sequences.
As the first viral gene to be transcribed during infection, the IE180 promoter can direct expression of a reporter gene in the absence of any viral protein synthesis or infection (58, 232). The upstream region of the promoter contains numerous binding sites for cellular transcription factors: nine imperfect direct repeats (approximately 80 bp each) containing Oct-1 and NF-µE1 binding sites (232, 420). The core promoter itself contains a TATA, as well as Sp1 and CCAAT motifs (420). IFN-
treatment of Vero cells reduces gene expression from the IE180 promoter, and a negative regulation element was mapped to be within 90 bp upstream of the transcription initiation site (407). The negative autoregulation of IE180 transcription is probably direct, since the IE180 DNA-binding domain can bind the IE180 promoter (441). Mapping of the IE180 protein regions critical for negative autoregulation suggests that the DNA-binding domain is required (389). We notice two CATCGT elements flanking the IE180 transcription initiation, and suggest them as likely targets for IE180 binding.
Despite the effect of IE180 on cellular gene expression, stable IE180-expressing lines were established in both Vero and PK15 cells to complement IE180-deleted PRV, though the authors also noted that IE180 expression seemed to enhance or reactivate the production of endogenous retroviruses in PK15 (445). The effect of IE180 on global cellular gene transcription awaits further study, as its effects are unlikely to be confined to the beta-globin and hsp70 genes.
EP0. As the name implies, the Early Protein 0 (EP0) gene is transcribed with early kinetics (76). The protein can be detected within 2 h after infection, and an additional slower migrating form appears later (310). The PRV EP0 has been detected within virions and shares the characteristics of a promiscuous transactivator (310). Recombinant EP0 can activate transcription initiation from synthetic TATA-based promoters in nuclear extracts (176). Expression of EP0 in vivo activates gene expression from PRV promoters, such as IE180, UL23 (thymidine kinase), and US4 (gG), as well as other viral promoters, such as VZV ORF29, HSV-1 UL23 (thymidine kinase), simian virus 40 early gene, and human immunodeficiency virus type 1 long terminal repeat (288, 424). However, EP0 has the opposite effect on the UL41 (vhs) promoter, reducing gene expression (73). Whether EP0 acts directly or indirectly to modulate transcription is not yet established. EP0 localizes to the nucleus following infection or transfection. Although no typical nuclear localization signal can be found in the EP0 protein sequence, deletion analysis suggests the existence of multiple nuclear localization signals within EP0 (423, 424).
Most of the EP0 gene overlaps with the oppositely transcribed large latency transcript (LLT) (76), so that deletions in EP0 inevitably delete part of the LLT as well. EP0 is dispensable for viral growth in cultured cells, but in its absence, viral titers and plaque size are reduced (18, 38). EP0-negative PRV mutants are attenuated in mice, swine and neonatal piglets (38, 81, 428). The lack of EP0 does not impair PRV in reaching and persisting in the trigeminal ganglia of swine after intranasal inoculation (80), but the amount of viral DNA harbored in trigeminal ganglia tissue is found to be reduced and dexamethasone is not effective in inducing the reactivation of infectious mutant virus.
There are two possible sources of the reactivation defect observed in EP0-negative infected animals: as the only gene known to be transcribed during latency, LLT seems the likely gene to be involved, but alternatively, the impaired reactivation could be due to the reduced replication of the mutant virus, a defect ascribed to the loss of EP0 function. Despite the reduced virulence and apparent defect in reactivation, an EP0 deletion mutant was able to elicit complete protective immunity in very young piglets (428), making the gene a potential target in PRV vaccine engineering.
EP0 is functionally homologous to the immediate-early ICP0 (RL2) gene of HSV-1, and they share little homology beyond a conserved C3HC4 RING finger domain, a zinc-binding motif thought to mediate protein-protein interactions (76). The PRV EP0 transactivation domains have been mapped to the N terminus and to the RING finger domain, and an intact RING domain is required for enhanced expression from PRV TK and IE180 promoters (423).
Expression of PRV EP0 or one of its homologs in related alphaherpesviruses (HSV-1 ICP0, VZV Vg61, BHV-1 BICP0, and EHV-1 Eg63) causes changes to ND10 structures and induces the colocalization of normally diffuse conjugated ubiquitin (314, 315). ND10 structures are repositories of transactivating factors, but their exact function remains uncertain (reviewed in reference 262). The growth defect of a PRV EP0 deletion mutant in cultured cells could be complemented by expression of the HSV-1 or VZV homolog (288). The conserved functions between alphaherpesvirus ICP0 homologs are likely to derive from the presence of the conserved RING finger domain. The RING finger region of HSV-1 ICP0 is essential for its regulation of gene expression, stimulation of lytic infection, enhancement of reactivation from quiescence, disruption of ND10 structures, induction of proteasome-dependent degradation of cellular proteins, and interaction with cyclin D3 (reviewed in reference 162). ICP0 also plays a role in blocking the antiviral effects of interferon in mice. It has recently been proposed that the multitude of functions demonstrated by HSV-1 ICP0 indirectly derive from its ability to serve as a component of the ubiquitin proteasome pathway (162). In this model, ICP0 does not regulate gene expression at the level of transcription, but rather at the level of protein stability (162).
Other regulators of gene expression: UL54, UL41, and UL48. PRV UL54 and UL41 are likely to encode potent regulators of both viral and cellular gene expression. In HSV-1, UL54 encodes ICP27, a multifunctional RNA-binding protein that stimulates or inhibits transcription in a gene-specific manner, inhibits pre-mRNA splicing, modulates pre-mRNA polyadenylation and stability, and exports viral mRNAs into the cytoplasm (83, 166, 167, 223, 244, 267, 271, 336, 358). While HSV-1 UL54 (ICP27) is expressed as an immediate-early gene, PRV UL54 is expressed with early kinetics (182).
Like HSV-1, the PRV UL54 protein resides in the nucleus of infected cells and avidly binds poly(G) RNA (182). The predicted protein sequence shows a zinc-finger like motif of unknown importance at the C terminus and an N-terminal arginine-glycine rich stretch (amino acids 45 to 54) that resembles the RGG RNA-binding motif found in HSV-1 ICP27 (273). PRV deleted for UL54 shows reduced cell-cell spread, and this defect can be complemented by expression of the homologous proteins of HSV-1 (ICP27) or VZV (ORF4) (364). Furthermore, the absence of UL54 reduces viral replication and alters viral gene expression: gC (UL44) amounts were reduced, gK (UL53) was absent, and the levels of gB (UL27), gE (US8) and US9 were increased. Whether these changes can all be attributed to the loss UL54 is unclear: UL54, UL53, and UL52 are transcribed as 3'coterminal transcripts and a UL54 deletion is expected to alter the 3' untranslated region of the UL53 and UL52 mRNAs (Fig. 2) (219). Indeed, deletion of UL54 reduces the mRNA levels of UL53 and UL52, encoding gK and a component of the viral replication machinery, respectively (364).
UL41 is conserved within alphaherpesviruses and encodes the vhs protein responsible for the virion host shut-off of cellular protein synthesis. Upon entry, the HSV-1 vhs protein present in the tegument induces the degradation of cellular (and viral) mRNAs (360, 235) by endoribonucleolytic cleavage of target RNAs (114). Subsequently, newly synthesized UL48 (VP16) binds to the vhs protein to inhibit its activity and allow the viral mRNAs to accumulate (239, 370). PRV UL41 lacks the VP16 binding site found in HSV-1, but has partially conserved the putative mRNA binding domain found in HSV-1 (6, 34). The PRV UL41 protein exhibits RNase activity, but is less active than HSV-1 vhs (114, 243, 359). As expected, deletion of PRV UL41 abrogates the degradation of host mRNAs and shows an early delay in viral growth (5). Unlike the early host protein shut-off observed with HSV-1, PRV infection results in a delayed shut-off similar to that seen in VZV, and requires de novo viral protein synthesis (5, 29, 30, 114, 185, 359). Like the VZV vhs homolog (ORF17), PRV UL41 is found in purified virion despite its lack of the VP16-binding site used for HSV-1 UL41 virion incorporation (243, 359). The PRV UL41 promoter can be transactivated by IE180 (73).
The tegument protein VP16 is encoded by UL48, a gene conserved among Alphaherpesvirinae (133). HSV-1 VP16 is known under many names (UL48, -TIF, Vmw65, or ICP25), and possesses multiple functions during induction of viral gene expression and viral egress (57, 292). The transactivation properties of HSV-1 VP16 have been extensively studied (347). Like its homologs in other alphaherpesviruses (24, 286, 289), PRV VP16 enhances expression of viral immediate-early genes in newly infected host cells (133). Virions lacking the UL48 gene and UL48 protein failed to produce the normally abundant IE180 transcript upon infection, leading to delayed onset of replication, reduced titers, and small plaque size in cultured cells (133). Like many other tegument proteins, PRV UL48 also functions in virion morphogenesis and egress: UL48-negative PRV mutant accumulates unenveloped cytoplasmic capsids (133). In vivo, the UL48-negative PRV mutant exhibits reduced virulence and neuroinvasion after intranasal inoculation of mice (210). The late apparition of clinical signs and extended time to death correlates with, and seems to be explained by, a delayed neuroinvasion of both first-order and second-order neurons.
PRV also modulates the host translation machinery, though little is known about the mechanistic details (reviewed in reference 29). In vitro translation of infected cell mRNAs in rabbit reticulocyte lysates find that a significant proportion of cellular mRNAs fail to be translated and that some early viral mRNAs are translated poorly or not at all. Furthermore, the polysomal mRNA species isolated from infected cells represent only a subset of cytoplasmic mRNA species.
Transcription from the PRV LAT region is active during lytic infection of cultured mammalian (PK15 and MDBK) cells although a different set of transcripts is expressed (191). Two LLT promoters have been identified. The first latency-active promoter (LAP1) has a TATA box located 34 nucleotides upstream from the initiation site of the LLT (76). The LAP2 TATA sequence is 143 bp downstream. The roles of the two latency-active promoters appear to be similar to those described for HSV-1 (75). LAP1 is thought to be a neuron-specific promoter but is not required for LAT transcription in cultured cells (183, 192). LAP2 is active in both neuronal and nonneuronal cells (82, 390). The PRV LAT promoter (LAP1, LAP2, and upstream region) is sufficient to direct transgene expression in the trigeminal ganglia and other neuronal tissues of transgenic mice (392).
PRV IE180 is likely involved in the complicated, cell type-specific regulation of LAT transcription. IE180 binds oligonucleotide sequences corresponding to LAP1 and IE180 downregulates expression of a LAP1-driven reporter gene in mouse neuroblastoma Neuro-2a cells (313). However, IE180 downregulates transcription only in nonneuronal and not Neuro-2a cells when a larger region of the LAT promoter (LAP1, LAP2, and region upstream of LAP1) is used (390). The interaction of the immediate-early protein with the LAT promoter may be an initiating step in PRV reactivation from latency.
In contrast to the wealth of information regarding latent infection cycles of other alphaherpesviruses (193), relatively little is known about PRV gene expression during latency. Ongoing research, including the development of the mouse model of latent PRV infection and reactivation (discussed under Models for Reactivation from Latency) will undoubtedly identify the shared and unique aspects of the PRV latent infection cycle compared to other alphaherpesviruses.
Serial passage of alphaherpesviruses at high multiplicities of infection can result in the establishment of a parasitic subpopulation of defective altered viral genomes that can be replicated and packaged into virion-like particles, but only in the presence of helper virus. The virion-like particles containing these genomes are called defective interfering particles (DIPs), as they can compete and interfere with the functional viral genome during DNA replication and encapsidation. Indeed PRV DIP genomes are found enriched for origins of replication and packaging signals (26, 28, 131, 174, 342, 343, 382, 440).
Herpesviruses encode many of the enzymes required for viral DNA replication. Seven HSV-1 proteins are required for origin-dependent synthesis of plasmid DNA: UL52, UL42, UL30, UL29, UL9, UL8, and UL5 (reviewed in reference 241). All seven genes are found conserved in PRV and are presumed to function similarly (Table 1). UL52, UL8, and UL5 are essential core genes encoding the subunits of the heterotrimeric primase-helicase complex, having been well studied in HSV-1, but not in PRV (reviewed in reference 241). UL30 and UL42 are essential genes conserved within all Herpesviridae, and encode the catalytic subunit (Pol) and polymerase-associated protein (Pap) of the viral DNA-dependent DNA polymerase holoenzyme, respectively (348). PRV UL30 possesses a DNA polymerase activity that could be stimulated by the addition of PRV UL42 in vitro, similar to what is seen for HSV-1 (35). The stimulation by UL42 was abrogated in a UL30 mutant missing the C-terminal 30 amino acids. Because of its potential as a target for antiviral drugs, HSV-1 UL30 has been extensively studied; antiherpetic drugs targeting UL30 include phosphonoacetic acid, foscarnet, and acyclovir.
PRV UL29 contains a conserved zinc-binding motif and a conserved DNA-binding region (443) and plays an essential role in viral genome replication (31). The protein is thought to bind the single-stranded DNA in unwound DNA and replication forks. Recombinant UL29 protein binds single-stranded DNA in a nonspecific and cooperative manner (443). Furthermore, the recombinant protein also physically interacts with the UL12 alkaline nuclease to stimulate its DNase activity, suggesting a possible role in viral recombination as well (179). A motif of the helicase type II superfamily is conserved among UL9 homologs, but little else is known about the PRV UL9 protein. HSV-1 UL9 initiates viral DNA replication by binding and unwinding viral origins of replication (oriS and oriL) (347). A separate transcript, encoded by PRV UL8.5 is translated into a protein of unknown function that corresponds to the C-terminal 470 amino acids of the UL9 protein (112). The UL8.5 protein is conserved with 47% identity in HSV-1 but no homolog has been described for other alphaherpesviruses, aside from the larger UL9 gene. HSV-1 UL8.5 has been designated OBPC and is capable of binding the HSV-1 origins of replication in vitro (19).
In addition to viral proteins, host proteins are likely required for PRV DNA replication. In HSV-1, cellular DNA polymerase alpha-primase, DNA ligase I, and topoisomerase II have all been proposed to participate in viral DNA synthesis (reviewed in reference 37). Host recombination proteins have also been suggested to play a role in HSV-1 DNA replication (reviewed in reference 434).
PRV UL50 encodes a bona fide dUTPase that is not incorporated into virions (198). Host and viral dUTPases catalyze the hydrolysis of dUTP into dUMP and pyrophosphate. Reducing the amount of dUTP is predicted to decrease misincorporation of dUTP into viral DNA, while the new product, dUMP, can serve as a precursor for dTMP and dTTP synthesis. PRV UL50 is dispensable for replication in cultured cells, its absence only slightly delaying viral growth kinetics (198). However, the same UL50-negative PRV strain is attenuated when young pigs are inoculated intranasally (196). Prior infection with the UL50-negative strain conferred protective immunity against Aujeszky's disease, making UL50 a good deletion target for safe and potent live vaccines.
UL39 and UL40 encode the small subunit (RR1) and large subunit (RR2) of the viral ribonucleotide reductase, respectively (201). UL39 is conserved within Herpesviridae and contains blocks of highly conserved sequences that are also found within the large subunit of cellular ribonucleotide reductases (201). Ribonucleotide reductase catalyzes the reduction of ribonucleotides into deoxyribonucleotides, the substrates for DNA synthesis. As opposed to most cellular ribonucleotide reductases, the PRV-encoded enzyme is resistant to dTTP product feedback inhibition (reviewed in reference 29). PRV strains mutated for either UL39 or UL40 are able to replicate in cultured cells but are severely attenuated in pigs and mice (103, 104).
UL23 is only found within the Alphaherpesvirinae and Gammaherpesvirinae and encodes the viral thymidine kinase. Cells also contain a thymidine kinase, and the phosphorylation of deoxythymidine is a critical step in the synthesis pathway of dTTP, a substrate for DNA synthesis. Herpesvirus thymidine kinases have broader substrate specificity than their host counterparts, allowing the development of nucleoside analogs, such as acyclovir, that can be phosphorylated into antiviral compounds (347). PRV thymidine kinase possesses thymidine kinase activity in vitro though its substrate spectrum is much more limited than that of HSV-1 thymidine kinase (254; reviewed in reference 29). While PRV UL23 is not essential for viral growth in most cultured cells, UL23-negative PRV mutants prove to be highly attenuated in mice, rabbits and pigs, and confer protective immunity against PRV challenge in pigs (208, 268). A thymidine kinase defect is responsible for the attenuation of the Tatarov vaccine strain (245).
PRV UL12 encodes the alkaline nuclease, an endo-exonuclease with catalytic properties similar to that of the bacterial recombination DNase RecBCD (180, 181). A UL12 PRV insertion mutant shows a strong reduction of virulence in mice (104). The HSV-1 homolog is a nuclease involved in the processing of replication intermediates of viral genomic DNA (324).
PRV UL2 is predicted to encode a uracil-DNA glycosylase, an enzyme conserved within prokaryotes and eukaryotes (105). Uracil DNA glycosylases (UDG or UNG) serve to remove the uracil bases that can occur following DNA damage (347). Removal of the uracil base then allows DNA repair to proceed. PRV UL2 has not been studied yet. HSV-1 UL2 is not essential for viral replication in cultured cells, but plays a role in viral pathogenicity and reactivation from latency (329).
Studies on the mechanism of DNA packaging in bacteriophages have provided considerable insight into the equivalent processes of the herpesviruses. It is likely that the basic mechanisms are conserved. For example, in most DNA viruses the two terminase subunits associate with the portal to form a powerful molecular motor to package DNA (70, 140). Herpesviruses encode a putative two-subunit terminase made up of UL15 and UL28 (3, 15, 16, 177), which can bind to the portal protein UL6 (432). In PRV, UL28 (ICP18.5) is required for DNA cleavage and encapsidation (282). In the absence of other viral proteins, PRV UL28 is distributed in the cytoplasm instead its normal nuclear location. Coexpression of HSV-1 UL15 and PRV UL28 is sufficient to target UL28 to the nucleus (229).
Other HSV-1 proteins involved in DNA cleavage and packaging include UL17, UL25, UL32, UL33, and the portal protein UL6 (347). PRV UL25 is found to be capsid-associated but has not yet been the subject of any mutagenesis study (200). The functions of the PRV genes UL6, UL32, and UL33 remain yet to be ascertained (229). Like HSV-1, PRV UL17 encodes an essential gene required for DNA cleavage and encapsidation (216, 357). PRV UL17 is a virion component and ultrastructural studies of infected cells strongly suggest that PRV UL17 is a located within the nucleocapsid, possibly associated with the viral DNA (216).
Nuclear egress and primary envelopment. The first step in egress is engagement of nucleocapsids with the inner nuclear membrane (Fig. 3). Primary envelopment occurs by budding of nucleocapsids through the inner nuclear membrane into the perinuclear space. This capsid with an envelope derived from the inner nuclear membrane is called the primary enveloped virion and the gene products of PRV UL31 and PRV UL34 contribute to its formation (138, 217). The UL34 protein is a type II, C-terminal anchored membrane protein, while the UL31 protein is a nuclear phosphoprotein. UL34 and UL31 colocalize in the nuclear envelope of infected cells and interact in a yeast two-hybrid assay (138). Nascent primary enveloped virions exit from the perinuclear space by fusion of the primary envelope with the outer nuclear membrane resulting in deenvelopment of the particles. Deenvelopment of the nascent virion results in entry of "naked" capsid structures into the cytoplasm, and requires the US3 protein kinase (215, 421).
In the absence of US3, a striking accumulation of primary enveloped virions can be observed by electron microscopy within invaginated portions of the inner nuclear membrane of the nuclear envelope (421). These structures extend into the perinuclear space. The US3 protein likely plays a role in fusion of the primary envelope of nascent virions with the outer nuclear membrane of the nuclear envelope, thus enabling deenvelopment of primary virions and release of "naked" virions into the cytoplasmic compartment. However, US3 is dispensable for viral replication in cultured epithelial cells and the absence of US3 only mildly reduces the titer of extracellular particles (215). Thus, deenvelopment can still occur in the absence of US3, albeit less efficiently. It is currently unknown whether the impairment in deenvelopment observed in the absence of US3 protein reflects an important structural role for this protein in primary enveloped virions or whether it is due to effects of its kinase function.
The US3 kinase influences the nuclear membrane association of a UL34, a primary virion component critical for nuclear egress (215, 217). UL34 can still localize to inner and outer leaflets of the nuclear membrane in absence of US3 protein, but does so less efficiently. Metabolic labeling experiments found no difference in UL34 phosphorylation in the absence of US3, indicating that the viral kinase UL13 or a cellular kinase phosphorylates UL34 (215). US3 was also found to induce actin stress fiber disassembly in swine epithelial cells, a step that could be important viral egress (414). The use of kinase-inactive US3 mutants will answer whether these functions require a catalytically active US3 kinase.
Detection of striking perinuclear accumulation of nascent virions in the absence US3 was recapitulated and further characterized in ultrastructural studies (155). In PRV US3-null mutant-infected rabbit kidney cells analyzed by immunoelectron microscopy, PRV US3 protein was detectable in both primary and mature virions. This correlates with previous findings on PRV US3 distribution (215, 250). The absence of US3 protein does not affect virion incorporation of several tegument proteins (UL11, UL37, UL46, UL47, UL48, and UL49) and envelope glycoprotein
