A deletion mutant of the latency-associated transcript of herpes simplex virus type 1 reactivates from the latent state with reduced frequency

Evidence for a bidirectional element located downstream from the herpes simplex virus type 1 latency-associated promoter that increases its activity during latency

Herpes simplex virus type 1 (HSV-1) latent infection in vivo is characterized by the constitutive expression of the latency-associated transcripts (LAT), which originate from the LAT promoter (LAP). In an attempt to determine the functional parts of LAP, we previously demonstrated that viruses harboring a DNA fragment 3' of the LAT promoter itself were able to maintain detectable promoter expression throughout latency whereas viruses not containing this element could not (J. R. Lokensgard, H. Berthomme, and L. T. Feldman, J. Virol. 71:6714-6719, 1997). This element was therefore called a long-term expression element (LTE). To further study the role of the LTE, we constructed plasmids containing a DNA fragment encompassing the LTE inserted into a synthetic intron between the reporter lacZ gene and either the LAT or the HSV-1 thymidine kinase promoter. Transient-expression experiments with both neuronal and nonneuronal cell lines showed that the LTE locus has an enhancer activity that does not activate the cytomegalovirus enhancer but does activate the promoters such as the LAT promoter and the thymidine kinase promoter. The enhancement of these two promoters occurs in both neuronal and nonneuronal cell lines. Recombinant viruses containing enhancer constructs were constructed, and these demonstrated that the enhancer functioned when present in the context of the viral DNA, both for in vitro infections of cells in culture and for in vivo infections of neurons in mouse dorsal root ganglia. In the infections of mouse dorsal root ganglia, there was a very high level of promoter activity in neurons infected with viruses bearing the LAT promoter-enhancer, but this decreased after the first 2 or 3 weeks. By 18 days postinfection, neurons harboring latent virus without the enhancer showed no beta-galactosidase (beta-gal) staining whereas those harboring latent virus containing the enhancer continued to show beta-gal staining for long periods, extending to at least 6 months postinfection, the longest time examined.

The latency-associated transcript gene enhances establishment of herpes simplex virus type 1 latency in rabbits

The latency-associated transcript (LAT) gene the only herpes simplex virus type 1 (HSV-1) gene abundantly transcribed during neuronal latency, is essential for efficient in vivo reactivation. Whether LAT increases reactivation by a direct effect on the reactivation process or whether it does so by increasing the establishment of latency, thereby making more latently infected neurons available for reactivation, is unclear. In mice, LAT-negative mutants appear to establish latency in fewer neurons than does wild-type HSV-1. However, this has not been confirmed in the rabbit, and the role of LAT in the establishment of latency remains controversial. To pursue this question, we inserted the gene for the enhanced green fluorescent protein (EGFP) under control of the LAT promoter in a LAT-negative virus (DeltaLAT-EGFP) and in a LAT-positive virus (LAT-EGFP). Sixty days after ocular infection, trigeminal ganglia (TG) were removed from the latently infected rabbits, sectioned, and examined by fluorescence microscopy. EGFP was detected in significantly more LAT-EGFP-infected neurons than DeltaLAT-EGFP-infected neurons (4.9% versus 2%, P < 0.0001). The percentages of EGFP-positive neurons per TG ranged from 0 to 4.6 for DeltaLAT-EGFP and from 2.5 to 11.1 for LAT-EGFP (P = 0.003). Thus, LAT appeared to increase neuronal latency in rabbit TG by an average of two- to threefold. These results suggest that LAT enhances the establishment of latency in rabbits and that this may be one of the mechanisms by which LAT enhances spontaneous reactivation. These results do not rule out additional LAT functions that may be involved in maintenance of latency and/or reactivation from latency.

Identification of a novel 0.7-kb polyadenylated transcript in the LAT promoter region of HSV-1 that is strain specific and may contribute to virulence

Herpes Simplex virus expresses latency-associated transcripts (LATs) the function of which remains obscure despite increasing knowledge of their structure and expression. Upstream of the LAT coding region is a region of the genome that is poorly characterized although it lies in an area that is responsible for modulation of reactivation efficiency in two different animal models. Transcript mapping with strains 17, McKrae, KOS, and F has revealed strain differences in this region of the viral genome. Strain 17 and McKrae expressed a novel polyadenylated 0.7-kb transcript that is absent from KOS and F. This transcript is expressed in the LAT direction and has the kinetics of a true late gene during the lytic cycle of infection. A deletion mutant, 17DeltaBsa, which does not express the 0.7-kb RNA, is less virulent than the parental strain 17. A rescuant with F sequence (17DeltaBsa/RF) shows virulence similar to F, whereas a rescuant with strain 17 sequence (17DeltaBsa/R17) is similar to strain 17. Virulence is altered by deletion or substitution in the region encoding the 0.7-kb transcript (BsaI-BsaI); however, reactivation in the mouse explant cocultivation assay or the adrenergically induced rabbit reactivation model remained unchanged. The importance of this region for virulence is discussed.

Herpes simplex virus type 1 serum neutralizing antibody titers increase during latency in rabbits latently infected with latency-associated transcript (LAT)-positive but not LAT-negative viruses

The herpes simplex virus type 1 (HSV-1) latency-associated transcript (LAT) gene is essential for efficient spontaneous reactivation in the rabbit ocular model of HSV-1 latency and reactivation. LAT is also the only viral gene abundantly expressed during latency. Rabbits were ocularly infected with the wild-type HSV-1 strain McKrae or the McKrae-derived LAT null mutant dLAT2903. Serum neutralizing antibody titers were determined at various times during acute and latent infection. The neutralizing antibody titers induced by both viruses increased and were similar throughout the first 45 days after infection (P > 0.05). However, by day 59 postinfection (approximately 31 to 45 days after latency had been established), the neutralizing antibody titers induced by wild-type virus and dLAT2903 diverged significantly (P = 0.0005). The dLAT2903-induced neutralizing antibody titers decreased, while the wild-type virus-induced neutralizing antibody titers continued to increase. A rescuant of dLAT2903, in which spontaneous reactivation was fully restored, induced wild-type neutralizing antibody levels on day 59 postinfection. A second LAT mutant with impaired spontaneous reactivation had neutralizing antibody levels comparable to those of dLAT2903. In contrast to the results obtained in rabbits, in mice, neutralizing antibody titers did not increase over time during latency with any of the viruses. Since LAT is expressed in both rabbits and mice during latency, the difference in neutralizing antibody titers between these animals is unlikely to be due to expression of a LAT protein during latency. In contrast, LAT-positive (LAT(+)), but not LAT-negative (LAT(-)), viruses undergo efficient spontaneous reactivation in rabbits, while neither LAT(+) nor LAT(-) viruses undergo efficient spontaneous reactivation in mice. Thus, the increase in neutralizing antibody titers in rabbits latently infected with LAT(+) viruses may have been due to continued restimulation of the immune system by spontaneously reactivating virus.

LAT expression during an acute HSV infection in the mouse

Previous studies using cell culture systems to evaluate LAT expression demonstrated that the LAT promoter expresses at much higher levels in neuroblastoma cell lines than fibroblast lines. The high level of LAT expression in neuronal-derived cell lines correlates with the high level of LAT accumulation observed in sensory ganglia neurons during a latent infection. We have found that using LAT promoters to express reporter genes from recombinant viruses in vivo produces high levels of LAT promoter activity in the epithelium of the mouse foot. An analysis of LAT promoter activity during an acute infection in the mouse clearly demonstrates that in contrast to studies performed with selected cell lines, the LAT promoter expresses similar levels of reporter gene product in peripheral cells and in neurons. In addition, the amount of reporter gene product is higher when the LAT promoter is located within the R(L) as compared to the U(L) region, and when expression is adjusted for copy number of the reporter construct, expression is roughly the same. These results suggest the activity of the LAT promoter varies greatly according to cell type and that high levels of expression is not limited solely to neurons, especially in the in vivo setting.

[Progressive multifocal leukoencephalopathy]

PATHOGENESIS

Progressive multifocal leukoencephalopathy is a demyelinating disease of the central nervous system caused by infection and reactivation of JC-virus. About 5% of all HIV-infected patients develop this fatal disease. Although pathogenesis is not completely understood, progressive multifocal leukoencephalopathy is thought to be a persistent infection. The kidneys, bone marrow, peripheral blood lymphocytes and the brain itself are candidates for latency sites of JC-virus. Loss of T-helper-cells in the course of HIV-infection or other immunosuppressive states result in reactivation of JC-virus.

DIAGNOSIS

Progressive multifocal leukoencephalopathy can be diagnosed by focal neurological symptoms, radiographic signs in magnetic resonance imaging and detection of JC-virus in brain tissue or cerebrospinal fluid.

TREATMENT

A specific therapy is not yet available or established. Highly active antiretroviral therapy (HAART) and cidofovir are promising and may prove useful in the near future.

Human corneal cells and other fibroblasts can stimulate the appearance of herpes simplex virus from quiescently infected PC12 cells

A two-cell system for the stimulation of herpes simplex virus type 1 (HSV-1) from an in vitro model of long-term (quiescent) infection is described. Rat pheochromocytoma (PC12) cells differentiated with nerve growth factor were infected with HSV-1 strain 17. Little, if any, cytotoxicity was observed, and a quiescent infection was established. The long-term infection was characterized by the absence of all detectable virus in the culture medium and little, if any, detectable early or late viral-gene expression as determined by reverse transcriptase PCR analysis. The presence of HSV-1 DNA was determined by PCR analysis. This showed that approximately 180 viral genomes were present in limiting dilutions where as few as 16 cells were examined. The viral DNA was infectious, since cocultivation with human corneal fibroblasts (HCF) or human corneal epithelial cells (HCE) resulted in recovery of virus from most, if not all, clusters of PC12 cells. Following cocultivation, viral antigens appeared first on PC12 cells and then on neighboring inducing cells, as determined by immunofluorescent staining, demonstrating that de novo viral protein synthesis first occurred in the long-term-infected PC12 cells. Interestingly, the ability to induce HSV varied among the cell lines tested. For example, monkey kidney CV-1 cells and human hepatoblastoma HepG2 cells, but not mouse neuroblastoma cells or undifferentiated PC12 cells, mediated stimulation. This work thus shows that (i) quiescent HSV infections can be maintained in PC12 cells in vitro, (ii) HSV can be induced from cells which do not accumulate significant levels of latency-associated transcripts, and (iii) the activation of HSV gene expression can be induced via neighboring cells. The ability of adjacent cells to stimulate HSV gene expression in neuron-like cells represents a novel area of study. The mechanism(s) whereby HCF, HCE, and HepG2 and CV-1 cells communicate with PC12 cells and stimulate viral replication, as well as how this system compares with other in vitro models of long-term infection, is discussed.

A herpes simplex virus type 1 latency-associated transcript mutant with increased virulence and reduced spontaneous reactivation

The herpes simplex virus type 1 (HSV-1) latency-associated transcript (LAT) gene is essential for efficient spontaneous reactivation of HSV-1 from latency. We previously reported that insertion of the LAT promoter and just the first 1.5 kb of the 8. 3-kb LAT gene into an ectopic location in the virus restored wild-type spontaneous reactivation to a LAT null mutant. This mutant, LAT3.3A (previously designated LAT1.5a), thus showed that the expression of just the first 1.5 kb of LAT is sufficient for wild-type spontaneous reactivation. We also showed that in the context of the entire LAT gene, deletion of LAT nucleotides 76 to 447 (LAT mutant dLAT371) had no effect on spontaneous reactivation or virulence. We report here on a LAT mutant designated LAT2.9A. This mutant is similar to LAT3.3A, except that the ectopic LAT insert contains the same 371-nucleotide deletion found in dLAT371. We found that LAT2.9A had a significantly reduced rate of spontaneous reactivation compared to marker-rescued and wild-type viruses. This was unexpected, since the combined results of dLAT371 and LAT3.3A predicted that spontaneous reactivation of LAT2.9A would be wild type. We also found that LAT2.9A was more virulent than wild-type or marker-rescued viruses after ocular infection of rabbits. This was unexpected, since LAT null mutants and LAT3.3A have wild-type virulence. These results suggest for the first time (i) that regions past the first 1.5 kb of LAT can compensate for deletions in the first 1.5kb of LAT and may therefore play a role in LAT dependent spontaneous reactivation and (ii) that regions of LAT affect viral virulence.

Identical 371-base-pair deletion mutations in the LAT genes of herpes simplex virus type 1 McKrae and 17syn+ result in different in vivo reactivation phenotypes

The herpes simplex virus type 1 (HSV-1) LAT gene is the only viral gene abundantly transcribed during latency. LAT null mutants created with strains McKrae and 17syn+ are impaired for both in vivo spontaneous and in vivo-induced reactivation. Thus, LAT is essential for efficient in vivo-induced and spontaneous reactivation. Different investigators have studied two LAT mutants containing a StyI-StyI region deletion corresponding to LAT nucleotides 76 to 447. One mutant, dLAT371 (parent strain, McKrae), had parental high frequencies of spontaneous reactivation. In vivo-induced reactivation was not examined. The other mutant, 17DeltaSty (parent strain, 17syn+), had parental frequencies of in vitro reactivation following cocultivation of explanted ganglia but reduced frequencies of in vivo-induced reactivation. Spontaneous reactivation frequency was not reported for 17DeltaSty. These combined results suggested the possibility that in vivo spontaneous reactivation and in vivo-induced reactivation may map to different regions within the LAT domain. We now report that dLAT371 has in vivo-induced reactivation frequencies of the parent and that 17DeltaSty has reduced frequencies of in vivo spontaneous reactivation. Thus, dLAT371 demonstrated the parental phenotype for both in vivo spontaneous and -induced reactivation while the apparently identical 17DeltaSty was impaired for both in vivo spontaneous and -induced reactivation. These results suggest that one or more differences between the genetic backgrounds of McKrae and 17syn+ result in different in vivo reactivation phenotypes of otherwise identical deletion mutations and that McKrae may have compensating sequences sufficient to overcome the loss of the StyI-StyI region of the LAT transcript.

ATF/CREB elements in the herpes simplex virus type 1 latency-associated transcript promoter interact with members of the ATF/CREB and AP-1 transcription factor families

The herpes simplex virus type 1 (HSV-1) latency-associated transcript (LAT) promoter 1 (LP1) is an inducible and cell type-specific promoter involved in regulating the production of an 8.3-kb primary LAT transcript during acute and latent infection of peripheral sensory neurons and during subsequent virus reactivation. A number of cis-acting regulatory elements have been identified in LP1, including two cyclic-AMP (cAMP) response element (CRE)-like sequences, designated CRE-1 and CRE-2. CRE-1 has previously been shown to confer cAMP responsiveness to LP1 and to regulate reactivation of HSV-1 from latency in vivo. A role for CRE-2 in modulating inducible activity is not yet as clear; however, it has been shown to support basal expression in neuronal cells in vitro. Electrophoretic mobility shift (EMS) analyses demonstrate that the LP1 CRE-like elements interact with distinct subsets of neuronal ATF/CREB and Jun/Fos proteins including CREB-1, CREB-2, ATF-1, and JunD. The factor-binding properties of each LP1 CRE element distinguish them from each other and from a highly related canonical CRE binding site and the TPA response element (TRE). LP1 CRE-1 shares binding characteristics of both a canonical CRE and a TRE. LP1 CRE-2 is more unusual in that it shares more features of a canonical CRE site than a TRE with two notable exceptions: it does not bind CREB-1 very well and it binds CREB-2 better than the canonical CRE. Interestingly, a substantial proportion of the C1300 neuroblastoma factors that bind to CRE-1 and CRE-2 have been shown to be immunologically related to JunD, suggesting that the AP-1 family of transcription factors may be important in regulating CRE-dependent LP1 transcriptional activity. In addition, we have demonstrated the two HSV-1 LP1 CRE sites to be unique with respect to their ability to bind neuronal AP1-related factors that are regulated by cAMP. These studies suggest that both factor binding and activation of bound factors may be involved in cAMP regulation of HSV-1 LP1 through the CRE elements, and indicate the necessity of investigating the expression and posttranslational modification of a variety of ATF/CREB and AP-1 factors during latency and reactivation.

Generation and use of recombinant reporter viruses for study of herpes simplex virus infections in vivo

It has become increasingly clear that the fate of herpes simplex virus (HSV) infections involves complex interactions between the virus and the specific cell types that comprise the tissues of the animal host. No reliable cell culture system for studying the establishment of latency and reactivation exists, and therefore these studies must be performed within animal models. One difficulty in elucidating the molecular regulation of these events is in determining the transcriptional activity of key viral genes during different stages of the infection in vivo. The heterogeneous cell types comprising infected tissues make PCR analysis of tissue homogenates difficult to interpret. The need to characterize expression of multiple transcriptional markers reliably and quantitatively at the level of individual cells is therefore key to determining the interplay between viral and cellular genes during latency and reactivation. Here we discuss the construction and evaluation of HSV reporter viruses that have been used in these analyses. HSV-1 recombinants have been engineered with representative viral promoters driving beta-galactosidase as a reporter. Methodology used to evaluate the levels of gene expression using (1) quantitative enzyme assays, (2) precipitatable substrate assays to localize the positive cells, and (3) immunohistochemistry and fluorescence assays to look at colocalization of markers during in vivo infection is presented. In addition to studying the molecular pathogenesis of HSV, the application of similar reporter viruses to evaluate promoters for targeting various differentiated tissues will be useful in developing these viruses as potential vectors for gene therapy.

Herpes simplex virus

Herpes simplex virus (HSV) infection is prevalent worldwide. Herpes labialis, caused predominantly by HSV-1, and herpes vulvovaginitis, caused predominantly by HSV-2, may result in significant morbidity and mortality for infected neonates exposed during delivery. The diagnosis of HSV infection is made by serological testing, viral culture, or polymerase chain reaction. Women with primary herpes vulvovaginitis exhibit a painful vesicular rash which is self-limited but may be followed by multiple recurrences. Women at greatest risk to transmit HSV to their neonates are those who experience their first episode of HSV during the latter stage of pregnancy. If infected, their neonates may have localized skin, eye and mucosal lesions, invasive central nervous system infection, or disseminated disease. Because of the potentially devastating outcome for a baby infected with HSV, pregnant women with active HSV lesions at delivery should be offered a cesarean section. Still, many neonates who are infected with HSV are born to women with asymptomatic HSV shedding. Therefore, prevention of HSV during pregnancy is exceedingly important.

Herpes simplex virus latency and the immune response

Following infection, herpes simplex virus establishes latency in the nervous system and recurrences of lytic replication occur periodically. Molecular events which may determine how virus enters latency, how it is maintained and what occurs during reactivation have been investigated. The role of the immune response in limiting infection of the nervous system, influencing the latent state and removing virus from peripheral sites following reactivation has also been studied.

The probability of in vivo reactivation of herpes simplex virus type 1 increases with the number of latently infected neurons in the ganglia

The purpose of this study was to define the relationship between herpes simplex virus (HSV) latency and in vivo ganglionic reactivation. Groups of mice with numbers of latently infected neurons ranging from 1.9 to 24% were generated by varying the input titer of wild-type HSV type 1 strain 17syn+. Reactivation of the virus in mice from each group was induced by hyperthermic stress. The number of animals that exhibited virus reactivation was positively correlated with the number of latently infected neurons in the ganglia over the entire range examined (r = 0.9852, P < 0. 0001 [Pearson correlation]).

Herpes simplex type 1 induction of persistent NF-kappa B nuclear translocation increases the efficiency of virus replication

The latent form of the dimeric transcription factor NF-kappa B is sequestered in the cytoplasm by proteins containing ankyrin repeats, such as 1 kappa B alpha and beta, or by the p105 precursor form of the NF-kappa B p50 subunit. Tumor necrosis factor alpha or virus infection can cause targeted destruction of 1 kappa B and nuclear translocation of NF-kappa B. Following translocation, NF-kappa B mediates immune, inflammatory, or anti-apoptotic responses. Here we present evidence that beginning at around 6 h postinfection, herpes simplex virus (HSV) induces a persistent translocation of NF-kappa B into the nucleus of C33 cells, coincident with loss of both 1 kappa B alpha and 1 kappa B beta. Translocation failed to occur when infecting virus was preincubated with neutralizing antibody to viral envelope glycoproteins gD or gH, thus preventing entry, or when cells infected with viruses expressing mutated forms of immediate-early regulatory proteins lCP4 or lCP27. Surprisingly, no increase in the trans-activation function of NF-kappa B, as assayed by transient expression of CAT, was detected following HSV infection. The significance of NF-kappa B nuclear translocation for virus replication was demonstrated by an 80-90% reduction in virus yield following infection of C33 cells expressing a constitutive repressor form of 1 kappa B alpha. Models that reconcile nuclear translocation of NF-kappa B with the inability to detect NF-kappa B-dependent gene expression are discussed.

Herpes simplex virus type 1 latency-associated transcripts suppress viral replication and reduce immediate-early gene mRNA levels in a neuronal cell line

During herpes simplex virus type 1 (HSV-1) latent infection in human dorsal root ganglia, limited viral transcription, which has been linked to HSV-1 reactivation ability, takes place. To study the involvement of this transcription in HSV-1 replication in neuronal cells and consequently in viral latency, we constructed stably transfected neuronal cell lines containing (i) the entire HSV-1 latency transcriptionally active DNA fragment, (ii) the same DNA sequence with deletions of the latency-associated transcript (LAT) promoters, or (iii) the DNA coding sequence of the LAT domain. Replication of HSV-1 or a LAT-negative mutant was markedly repressed in the LAT-expressing cells, a phenomenon mediated by the LATs. To study the mechanism responsible for this effect, we examined LAT influence upon expression of HSV-1 immediate-early (IE) genes ICP0, ICP4, and ICP27, by Northern blot analysis. Following infection of a LAT-expressing neuronal cell line with a LAT-negative mutant, the steady-state levels of all three IE mRNAs were reduced compared to those for control cells. Transient transfections into a neuronal cell line indicated that the LAT suppressive effect upon ICP0 mRNA was mediated directly and was not due to the LAT effect upon the ICP0 promoter. We therefore propose that the LATs may repress viral replication in neuronal cells by reducing IE gene mRNA levels and thus facilitate the establishment of HSV-1 latency in nervous tissue.

A virus with a mutation in the ICP4-binding site in the L/ST promoter of herpes simplex virus type 1, but not a virus with a mutation in open reading frame P, exhibits cell-type-specific expression of gamma(1)34.5 transcripts and latency-associated transcripts

The herpes simplex virus type 1 L/S junction-spanning transcripts (L/STs) are a family of multisized transcripts expressed at high levels in cells infected with mutant viruses that (i) do not express ICP4, (ii) specify forms of ICP4 unable to bind to the consensus ICP4 binding site, or (iii) contain mutations in the ICP4 binding site located at the transcriptional start site of the L/STs. By extension, the failure to detect the L/STs in wild-type virus-infected cells is due to the repressive effect of ICP4 bound to its cognate binding site upstream of the L/ST transcription initiation site. ORF-P, the first and largest open reading frame (ORF) encoded by the L/STs, overlaps >90% of the ORF encoding ORF-34.5, a putative neurovirulence factor, which is transcribed from the opposite DNA strand. Viruses with mutations in the overlapping region of ORF-P and ICP34.5 exhibit premature shutoff of infected-cell protein synthesis and are highly attenuated following intracranial inoculation of juvenile mice. To determine whether the premature protein shutoff and neuroattenuated phenotypes of ORF-P ORF-34.5 double mutants are a consequence of alterations in ORF-P, ORF-34.5, or both, viruses containing mutations only in ORF-P or only in the ICP4 binding site in the L/ST promoter were isolated and characterized. Mutant virus L/ST-n38 contains a single-base-pair transition mutation in ORF-P codon 38, resulting in translational termination of the ORF-P protein (OPP). This mutation does not alter the amino acid sequence of ICP34.5. Expression of a truncated form of OPP by mutant virus L/ST-n38 did not result in premature shutoff of infected-cell protein synthesis and produced no other observable phenotype relative to wild-type virus in in vitro tests. Moreover, the 50% lethal dose (LD50) of L/ST-n38 was comparable to that of wild-type virus following intracranial inoculation of 3-week-old mice, as were the latency and reactivation phenotypes of the virus. These properties of L/ST-n38 indicate that the attenuated phenotype of ORF-P ORF-34.5 double mutants is a consequence of mutations that affect the function of ICP34.5 and not the function of OPP. Mutant virus LST-4BS contains four single-base-pair substitutions in the ICP4 binding site in the L/ST promoter that abrogate the binding of ICP4 to this site, leading to high-level expression of the L/STs and OPP. LST-4BS induced premature shutoff of viral and cellular protein synthesis and was slightly growth restricted in cells of neural lineage (SK-N-SH human neuroblastoma cells) but was wild type for these two parameters in cells of nonneural lineage (immortalized primate Vero cells). Of particular interest was the observation that L/ST-4BS exhibited cell-type-specific expression of both the gamma(1)34.5 transcripts and the latency-associated transcripts (LATs). Thus, expression of these transcripts was barely detectable in cells of neural lineage (NB41A3 mouse neuroblastoma cells) but was wild type in Vero cells. In vivo, L/ST-4BS was reactivated from mouse trigeminal ganglia with reduced efficiency and delayed kinetics relative to wild-type virus. L/ST-4BS was completely attenuated for neurovirulence (LD50 > 10(6) PFU) relative to wild-type virus (LD50 < 900 PFU), although the four single-base-pair substitutions lie outside the coding region for the neurovirulence factor, ICP34.5. Collectively, the complex in vitro and in vivo phenotypes of L/ST-4BS can be attributed to (i) disruptions of the ICP4 binding site in the L/ST promoter and subsequent overexpression of the L/STs and OPP; (ii) alterations in ORF-O, which is also mutated in L/ST-4BS; or (iii) alterations in other cryptic genes or cis-acting elements.

The quantity of latent viral DNA correlates with the relative rates at which herpes simplex virus types 1 and 2 cause recurrent genital herpes outbreaks

Herpes simplex virus types 1 and 2 (HSV-1 and HSV-2) have evolved specific anatomic tropisms and site-dependent rates of reactivation. To determine whether reactivation rates depend on distinct abilities of HSV-1 and -2 to establish latency and to express latency-associated transcripts (LATs), virulent strains of each virus were studied in the guinea pig genital model. Following infection with equivalent titers of virus, the quantities of latent HSV-2 genomes and LATs were higher in lumbosacral ganglia, and HSV-2 infections recurred more frequently and lasted longer than HSV-1 infections. In contrast, if the inoculum of HSV-1 was 10 times that of HSV-2, the quantity of HSV-1 DNA and LATs increased correspondingly and HSV-1 infections were as likely to recur as those with HSV-2. The quantity of latent virus DNA correlates with and may be a major determinant of the site-specific patterns and rates of reactivation of HSV-1 and -2.

The region of the herpes simplex virus type 1 LAT gene involved in spontaneous reactivation does not encode a functional protein

We previously showed that the LAT function required for efficient spontaneous reactivation of herpes simplex virus type 1 (HSV-1) from neuronal latency in the rabbit maps within the first 1.5 kb of the 8.3-kb primary LAT transcript. This demonstrated that LAT does not function via an antisense mechanism, since the first 1.5 kb of LAT does not overlap any other known HSV-1 gene. Furthermore, if LAT encodes a protein essential for efficient spontaneous reactivation, it must map within the functional first 1.5 kb of LAT. Thus, the absence of a well-conserved LAT open reading frame in this region among all HSV-1 LAT genes capable of supporting high levels of spontaneous reactivation would demonstrate that LAT does not encode a protein essential for efficient spontaneous reactivation. In this report, we sequenced the first 1.5 kb of LAT from HSV-1 McKrae, a strain with a very high spontaneous reactivation rate. Of the HSV-1 LAT sequences available for comparison (17syn+, KOS, and F), only strain 17syn+ has a high spontaneous reactivation rate. However, as shown in this report, a chimeric virus containing the KOS LAT gene on an HSV-1 McKrae genetic background had a spontaneous reactivation rate indistinguishable from McKrae (15 versus 13.6%; P > 0.05). Thus, the spontaneous reactivation competency of the LAT gene from HSV-1 KOS was similar to that of the McKrae LAT gene. Comparative sequence analysis of the LAT genes from McKrae, 17syn+, and KOS revealed that none of the eight potential McKrae LAT ORFs were well conserved. Additional types of sequence analyses further confirmed that none of the potential ORFs were likely to encode a functional LAT protein. These results strongly support the notion that the LAT function involved in spontaneous reactivation is mediated by a direct DNA or RNA mechanism rather than a protein.

A 2.9-kilobase noncoding nuclear RNA functions in the establishment of persistent Hz-1 viral infection

Differential viral gene expression during both productive and persistent infections of Hz-1 virus in insect cells was elucidated. Despite more than 100 viral transcripts being expressed during productive viral infection, massive viral gene shutoff was observed during viral persistency, leaving the 2.9-kb persistence-associated transcript 1 (PAT1) as the only detectable viral RNA. Persistence-associated gene 1 (pag1), which encodes PAT1, was cloned and found to contain no significant open reading frames. PAT1 is not associated with the cellular translation machinery and is located exclusively in the nucleus. Further experiments showed that PAT1 is functional in the establishment of persistent Hz-1 viral infection in the cells. All the evidence collectively indicates that PAT1 is a novel nuclear transcript of viral origin. Our results showed that although PAT1 and XIST RNA, a mammalian X-inactive specific transcript, are transcribed by different genes, they have interesting similarities.

Mutations in the 5' end of the herpes simplex virus type 2 latency-associated transcript (LAT) promoter affect LAT expression in vivo but not the rate of spontaneous reactivation of genital herpes

The primary herpes simplex virus type 2 (HSV-2) latency-associated transcript (LAT) promoter influences LAT expression and rates of virus reactivation. We explored the biological importance of particular neuronally responsive regions within the promoter by creating new recombinant viruses bearing a targeted deletion (246 bp [strain 524]) or a point mutation (2 bp [strain 167]) in this region. These recombinant viruses grew efficiently in vitro and in vivo, caused acute genital disease in guinea pigs, and, as measured by quantitative-competitive (QC) DNA PCR, established latency, all as well as did the wild-type parental HSV-2 strain 333, the rescuant strain 524R, and the previously described 624-bp LAT- promoter deletion mutant. By QC-reverse transcriptase PCR of RNA from latently infected ganglia, mutant 167 expressed wild-type levels of LAT and the deletion mutant 524 expressed 9- to 15-fold less LAT than normal, while the LAT expression of the LAT- mutant was undetectable or at least 5 log units less than that of the wild type. The rates of recurrence of genital lesions were normal for recombinant viruses 524 and 167 but reduced (as expected) for the LAT- mutant. Alteration of a subset of LAT promoter elements reduced LAT expression by 1 log unit but did not influence the rate of spontaneous disease reactivation in vivo. Far greater reductions in LAT expression are necessary before reactivation rates are noticeably changed.

The product of ORF O located within the domain of herpes simplex virus 1 genome transcribed during latent infection binds to and inhibits in vitro binding of infected cell protein 4 to its cognate DNA site

The partially overlapping ORF P and ORF O are located within the domains of the herpes simplex virus 1 genome transcribed during latency. Earlier studies have shown that ORF P is repressed by infected cell protein 4 (ICP4), the major viral regulatory protein, binding to its cognate site at the transcription initiation site of ORF P. The ORF P protein binds to p32, a component of the ASF/SF2 alternate splicing factors; in cells infected with a recombinant virus in which ORF P was derepressed there was a significant decrease in the expression of products of key regulatory genes containing introns. We report that (i) the expression of ORF O is repressed during productive infection by the same mechanism as that determining the expression of ORF P; (ii) in cells infected at the nonpermissive temperature for ICP4, ORF O protein is made in significantly lower amounts than the ORF P protein; (iii) the results of insertion of a sequence encoding 20 amino acids between the putative initiator methionine codons of ORF O and ORF P suggest that ORF O initiates at the methionine codon of ORF P and that the synthesis of ORF O results from frameshift or editing of its RNA; and (iv) glutathione S-transferase-ORF O fusion protein bound specifically ICP4 and precluded its binding to its cognate site on DNA in vitro. These and earlier results indicate that ORF P and ORF O together have the capacity to reduce the synthesis or block the expression of regulatory proteins essential for viral replication in productive infection.

Mutagenesis of a cAMP response element within the latency-associated transcript promoter of HSV-1 reduces adrenergic reactivation

Mutagenesis of a cyclic AMP response element (CRE) within the LAT promoter of HSV-1 reduces the ability of LAT expression to be induced in transient assays, but has only a minimal impact on reactivation of the virus in in vitro systems. Here we show that a CRE mutation results in a significant reduction of adrenergically induced reactivation in vivo in the rabbit eye model. Spontaneous reactivation frequencies were also reduced. In addition, we demonstrate that this mutation has no effect on the amount of LAT expressed during latency when compared with the parent, 17syn+, and the rescuant. These results indicate a greater effect of CRE on induced reactivation in vivo than in in vitro systems, but also suggest that the CRE in the LAT promoter is not autonomous in conducting the reactivation signal.

A LAT-associated function reduces productive-cycle gene expression during acute infection of murine sensory neurons with herpes simplex virus type 1

Herpes simplex virus (HSV) persists in the human population by establishing long-term latent infections followed by periodic reactivation and transmission. Latent infection of sensory neurons is characterized by repression of viral productive-cycle gene expression, with abundant transcription limited to a single locus that encodes the latency-associated transcripts (LATs). We have observed that LAT- deletion mutant viruses express viral productive-cycle genes in greater numbers of murine trigeminal ganglion neurons than LAT+ HSV type 1 at early times during acute infection but show reduced reactivation from latent infection. Thus, a viral function associated with the LAT region exerts an effect at an early stage of neuronal infection to reduce productive-cycle viral gene expression. These results provide the first evidence that the virus plays an active role in down-regulating productive infection during acute infection of sensory neurons. The effect of down-regulation of productive-cycle gene expression during acute infection may contribute to viral evasion from the host immune responses and to reduced cytopathic effects, thereby facilitating neuronal survival and the establishment of latency.

Selection of a nonconsensus branch point is influenced by an RNA stem-loop structure and is important to confer stability to the herpes simplex virus 2-kilobase latency-associated transcript

Herpes simplex virus type 1 latent infection in sensory neurons is characterized by the highly restricted transcription of viral genes. The latency-associated transcripts (LAT) family members are the only transcripts that can be identified in large amounts in latently infected cells. The most abundant LAT species is a 2-kb RNA that results from splicing of a rare primary transcript. Analysis of a LAT mutant virus (TB1) in cell culture revealed an aberrant splicing pattern and production of a stable small (0.95-kb) LAT intron. A panel of deletion constructs expressing truncated LAT in transiently transfected cells mapped the region influencing stability to the 3' end of the LAT intron. This region encompasses the branch point and a putative stable stem-loop hairpin structure immediately upstream of the splice acceptor consensus polypyrimidine tract. Mutagenic analysis of the sequence in this region confirmed our hypothesis that the stem-loop structure is important for efficient splicing by influencing the selection of a nonconsensus branch point. Changes in this structure correlate with changes in branch point selection and production of an unstable 2-kb LAT.

A viral function represses accumulation of transcripts from productive-cycle genes in mouse ganglia latently infected with herpes simplex virus

Latent infections of neurons by herpes simplex virus form reservoirs of recurrent viral infections that resist cure. In latently infected neurons, viral gene expression is severely repressed; only the latency-associated transcripts (LATs) are expressed abundantly. Using sensitive reverse transcriptase PCR assays, we analyzed the effects of a deletion mutation in the LAT locus on viral gene expression in latently infected mouse trigeminal ganglia. The deletion mutation, which reduced expression of the major LATs 10(5)-fold, resulted in a approximately 5-fold increase in accumulation of transcripts from the immediate-early gene encoding ICP4, an essential transactivator of viral gene expression. The LAT deletion also resulted in a >10-fold increase in the accumulation of transcripts from the early gene encoding thymidine kinase, whose expression during productive infection stringently depends on ICP4, and positively affected the correlation of the levels of these transcripts with the levels of ICP4 transcripts. We also detected transcripts antisense to ICP4 RNA, which were in substantial excess to ICP4 transcripts in ganglia latently infected with wild-type virus. In contrast to its effects on productive-cycle transcripts, the LAT deletion reduced the accumulation of these antisense transcripts approximately 15-fold. Thus, a viral function associated with the LAT locus represses the accumulation of transcripts from at least two productive-cycle genes in latently infected mouse ganglia. We discuss possible mechanisms and consequences of this repression.

Experimental investigation of herpes simplex virus latency

The clinical manifestations of herpes simplex virus infection generally involve a mild and localized primary infection followed by asymptomatic (latent) infection interrupted sporadically by periods of recrudescence (reactivation) where virus replication and associated cytopathologic findings are manifest at the site of initial infection. During the latent phase of infection, viral genomes, but not infectious virus itself, can be detected in sensory and autonomic neurons. The process of latent infection and reactivation has been subject to continuing investigation in animal models and, more recently, in cultured cells. The initiation and maintenance of latent infection in neurons are apparently passive phenomena in that no virus gene products need be expressed or are required. Despite this, a single latency-associated transcript (LAT) encoded by DNA encompassing about 6% of the viral genome is expressed during latent infection in a minority of neurons containing viral DNA. This transcript is spliced, and the intron derived from this splicing is stably maintained in the nucleus of neurons expressing it. Reactivation, which can be induced by stress and assayed in several animal models, is facilitated by the expression of LAT. Although the mechanism of action of LAT-mediated facilitation of reactivation is not clear, all available evidence argues against its involving the expression of a protein. Rather, the most consistent models of action involve LAT expression playing a cis-acting role in a very early stage of the reactivation process.

Comparative analyses of the latency-associated transcript promoters from herpes simplex virus type 1 strains H129, +GC and KOS-63

We have analyzed the activity of a specific portion of the latency-associated transcript (LAT) promoter of three strains of herpes simplex virus type 1 (HSV-1) in neuronal and non-neuronal cell types. Restriction fragments containing the LAT promoter sequences and the 5'-end of the LATs were isolated from HSV-1 strains H129, +GC and KOS-63, sequenced and cloned into a chloramphenicol transferase (CAT) plasmid vector. These vectors were separately assayed for CAT production in human (SknSH) and mouse (C-1300) neuroblastoma cell lines and a human continuous cell line (HeLa). Strain KOS-63 contained a C to T base substitution within the LAT promoter binding factor element upstream of the cAMP response element binding sequence. In replicate experiments, in which the construct DNA was used for transfection, the CAT constructs from strains H129 and +GC functioned equally well in all three cell lines. In contrast, the strain KOS-63 CAT construct functioned significantly better in HeLa cells than in neuroblastoma cell lines and better than the identical CAT constructs from strains H129 and +GC. In addition, the construct from strain KOS-63 functioned less well in the human neuroblastoma cell line than in HeLa or C-1300 neuroblastoma cells. When LAT expression was examined directly in vivo by in situ hybridization, strain KOS-63 produced slightly less LAT RNA than strain H129 within trigeminal ganglionic neurons of latently infected rabbits. However, utilizing competitive gel-shift assays, DNA fragments containing the LAT promoter binding element from all three strains bound equivalent amounts of HeLa cell nuclear proteins. Together, these results suggest that the activity expressed by the strain KOS-63 LAT promoter in vivo and in vitro may relate to positive or negative effects of DNA binding proteins on LAT transcription, and that these effects are cell-type dependent.

The virion host shutoff function of herpes simplex virus type 1 plays a role in corneal invasion and functions independently of the cell cycle

A significant restriction was demonstrated in the ability of herpes simplex virus type 1 virion host shutoff (vhs) mutant viruses to invade the corneal epithelium. Viral replication and invasion was confined to the areas of the cornea which were scarified prior to infection. Differences between wild-type and vhs mutant replication in corneas in vivo were 100- to 1000-fold at all timepoints postinfection. Smaller but still significant growth restrictions were observed in cultured corneal cells. This difference between in vitro and in vivo is not likely to be due to differences in cell cycle status since vhs-induced RNA degradation can occur in both cycling and noncycling cells in vitro. The vhs function is therefore important for invasion of the cornea and secondarily the nervous system and is thereby required for efficient establishment of latency.

The abundant latency-associated transcripts of herpes simplex virus type 1 are bound to polyribosomes in cultured neuronal cells and during latent infection in mouse trigeminal ganglia

During herpes simplex virus type 1 (HSV-1) latency, limited viral transcription takes place. This transcription has been linked to the ability of the HSV-1 genome to reactivate and consists of abundant 2.0- and 1.5-kb collinear latency-associated transcripts (LATs), spanned by minor hybridizing RNA (mLAT). The 1.5-kb LAT is derived from the 2.0-kb LAT by splicing, and both transcripts contain two large overlapping open reading frames. The molecular action mechanisms of the latency-associated gene expression are unknown, and no HSV-1 latency-encoded proteins have been convincingly demonstrated. We have cloned the entire latency-associated transcriptionally active HSV-1 DNA fragment (10.4 kb) under control of a constitutive promoter and generated a neuronal cell line (NA4) stably transfected with the viral LAT's region. NA4 cells produced the 2.0- and the 1.5-kb LATs. Northern blotting and reverse transcription-PCR analysis of RNA from NA4 cells and from trigeminal ganglia of mice latently infected with HSV-1 revealed that the two abundant LAT species were present in the polyribosomal RNA fractions. After addition of EDTA, which causes dissociation of mRNA-ribosome complexes, both LATs could be detected only in subpolyribosomal, but not in polyribosomal fractions. These results show that (i) HSV-1 LATs are bound to polyribosomes during latency in vivo, as well as in neuronal cells in vitro, and therefore might be translated, and that (ii) the NA4 cell line is a suitable tool with which to look for HSV-1 latency-encoded gene products.

Analysis of the 2-kilobase latency-associated transcript expressed in PC12 cells productively infected with herpes simplex virus type 1: evidence for a stable, nonlinear structure

The major latency-associated transcript (LAT) expressed in PC12 cells productively infected with herpes simplex virus type 1 is a 2-kb, nonpolyadenylated RNA molecule that accumulates in the nuclei of infected cells. In actinomycin D-treated cells, the 2-kb LAT gene transcript has a half-life considerably greater than 12 h. After polyacrylamide gel electrophoresis, two species of the transcript were observed, a major species that was retarded in the gel and a minor species that migrated as a 1.96-kb RNA molecule. RNase H digestion after hybridization of the RNA with an oligonucleotide complementary to positions -80 to -101 relative to the 3' end of the 2-kb LAT gene transcript changed the mobility of the retarded species into that of the rapidly migrating species. Our data indicate that the 2-kb LAT gene transcript expressed in productively infected PC12 cells is present in a stable, nonlinear form.

The herpes simplex virus type 1 latency-associated transcript promoter is activated through Ras and Raf by nerve growth factor and sodium butyrate in PC12 cells

Herpes simplex virus establishes latent infections in the nuclei of sensory neurons. These infections are characterized by the abundant expression of a series of 5' coterminal transcripts termed the latency-associated transcripts (LATs). Available evidence indicates that LAT expression is specifically regulated in latently infected neurons. Although previous studies have examined the regulation of LAT expression in neuronal and nonneuronal cells, the mechanism of regulation of LAT expression in neuronal cells in response to external factors has not been investigated. To address this question, we characterized the activity of LAT promoter fusion constructs in PC12 cells following treatment with nerve growth factor (NGF) and/or sodium butyrate (NaB), agents that affect expression of cell cycle-associated genes. Expression from the LAT promoter was induced 8- to 12-fold by either NGF or NaB alone and 40- to 60-fold when the two agents were added simultaneously. Fibroblast growth factor also induced expression from the LAT promoter but to a lesser extent than NGF. Treatment with factors such as epidermal growth factor, phorbol myristate acetate, cyclic AMP, or KCI had no significant effect on LAT promoter activity. Notably, promoter-reporter constructs containing immediate-early (ICP0 and ICP4), early (ICP8 and UL9), and late (UL10 and UL30) viral promoters were induced only two- to fourfold by NGF, suggesting that the LAT promoter may be unusual among herpes simplex virus genes in the magnitude of its response to this factor. To identify pathways leading to LAT activation in vitro, we characterized the response of the LAT promoter to NGF and/or NaB in PC12-derived cell lines containing mutations in specific signal transduction pathways. We found that activation of the LAT promoter requires Ras activation and that activation of the serine/threonine kinase, Raf, is sufficient to induce LAT expression. Together, these results indicate that the LAT promoter is regulated via the Ras/Raf signal transduction pathway in response to external factors such as NGF and NaB and that LAT expression may be regulated by NGF in latently infected neurons.

Identification of cis-acting sequences in the promoter of the herpes simplex virus type 1 latency-associated transcripts required for activation by nerve growth factor and sodium butyrate in PC12 cells

In the absence of detectable viral proteins, expression of the latency-associated transcripts (LATs) is likely regulated by cellular factors during latent infection of neurons with herpes simplex virus type 1. The amounts and activation states of these factors may in turn be regulated by extracellular regulatory factors. Consistent with this hypothesis, we have recently demonstrated that LAT expression is significantly enhanced by nerve growth factor (NGF) and sodium butyrate (NaB) in neurally derived PC12 cells. With the ultimate goal of identifying trans-acting cellular factors involved in regulating LAT expression during latency, we have attempted to identify the cis-acting elements to which these putative cellular factors bind by characterizing the LAT promoter and a series of 5' promoter deletion mutants in PC12 cells following treatment with the LAT-enhancing agents NGF and NaB. Transient expression assays demonstrated that distinct cis-acting sequences mediate basal and induced LAT promoter expression. Basal activity in PC12 cells is mediated by two elements: a negative regulatory element between -435 and -270 and a positive element between -240 and -204. The positive element contains binding sites for the transactivator Sp-1, whereas the negative element bears some resemblance to known neuron-specific silencer elements. In contrast to basal expression, maximum induction of the LAT promoter by NGF and NaB requires sequences between -159 and -81. Using gel mobility shift assays, we have identified three sets of protein-DNA complexes that bind to this 78-bp region and shown by competition analysis that binding is specific. The abundance and mobility of these complexes were altered by treatment with NGF or NaB. The nucleotide sequences to which these complexes bind were fine mapped by competition analysis with oligonucleotide probes containing substitution mutations. The target sequences identified exhibit no homology to binding sites of known transcription factors. These regions were critical for complex formation in vitro and for maximum induction of the LAT promoter by NGF and NaB in transient expression assays. The protein complexes that form with target sequences likely participate in the regulation of LAT expression in response to physiological stimuli in neurons in vivo.

Herpes simplex: evolving concepts

A large body of molecular biologic research has begun to clarify some basic aspects of viral latency and reactivation. The clinical definition of herpes simplex virus infection is expanding, with the recognition that the disease is largely asymptomatic and that most transmission occurs during periods of asymptomatic viral shedding. With this awareness, serologic diagnosis has become increasingly important. New treatment modalities are now available, and other promising treatments are in development.

Human herpes viruses latent infection in the nervous system

The neurotropic herpes viruses, HSV-1, HSV-2 and VZV, colonize and establish latent infection in human peripheral sensory ganglia. Recurrent diseases due to reactivation of these viral pathogens can take place despite an effective immune response. Molecular, cellular, physiological and immune mechanisms work in concert to enable the establishment of latency, the maintenance of the latent state for the entire life of the host, and the reactivation infection. Although all three viruses belong to the same family and establish latent infection in the same tissue, the clinical pattern of their reactivation is quite different. This review covers current knowledge of the basis of these infections, and offers a theory explaining the basis of HSV-1 latent infection and the differences of the disorders caused by HSV-1 and VZV reactivation in humans.

The characteristic site-specific reactivation phenotypes of HSV-1 and HSV-2 depend upon the latency-associated transcript region

After replication at sites of initial inoculation, herpes simplex virus type 1 and 2 (HSV-1 and HSV-2) establish lifelong latent infections of the sensory and autonomic neurons of the ganglia serving those sites. Periodically, the virus reactivates from these neurons, and travels centripetally along the neuronal axon to cause recurrent epithelial infection. The major clinically observed difference between infections with herpes simplex virus type 1 and type 2 is the anatomic site specificity of recurrence. HSV-1 reactivates most efficiently and frequently from trigeminal ganglia, causing recurrent ocular and oral-facial lesions, while HSV-2 reactivates primarily from sacral ganglia causing recurrent genital lesions. An intertypic recombinant virus was constructed and evaluated in animal models of recurrent ocular and genital herpes. Substitution of a 2.8-kbp region from the HSV-1 latency-associated transcript (LAT) for native HSV-2 sequences caused HSV-2 to reactivate with an HSV-1 phenotype in both animal models. The HSV-2 phenotype was restored by replacing the mutated sequences with wild-type HSV-2 LAT-region sequences. These sequences or their products must act specifically in the cellular environments of trigeminal and sacral neurons to promote the reactivation patterns characteristic of each virus.

A 371-nucleotide region between the herpes simplex virus type 1 (HSV-1) LAT promoter and the 2-kilobase LAT is not essential for efficient spontaneous reactivation of latent HSV-1

The herpes simplex virus type 1 (HSV-1) latency-associated transcript (LAT) gene is essential for efficient spontaneous reactivation of HSV-1 from latency. However, neither the mechanism by which LAT carries out this function nor the region of LAT responsible for this function in known. LAT is transcribed as an unstable 8.3-kb RNA that gives rise to a very stable 2-kb LAT RNA that is readily detected in latently infected sensory neurons. We show here that 371 of the 662 nucleotides located between the start of LAT transcription and the 5' end of the 2-kb LAT RNA do not appear to be essential for wild-type levels of spontaneous reactivation in the rabbit ocular model of HSV-1 latency. We deleted LAT nucleotides 76 to 447 from both copies of the LAT gene (one in each viral long repeat) to produce the mutant dLAT371. Rabbits were ocularly infected with dLAT371, and spontaneous reactivation was measured in comparison with the marker-rescued virus dLAT371R. Both dLAT371 and dLAT371R had spontaneous reactivation rates of approximately 13 to 14%. This was consistent with the parental McKrae wild-type virus (11.7%; P = 0.49) and significantly higher than the LAT transcription-negative mutant dLAT2903 (2.4%; P < 0.0001). Southern analysis confirmed that the spontaneously reactivated dLAT371 virus retained the deletion in both copies of LAT. Therefore, it appeared that the function of LAT involved in efficient spontaneous reactivation mapped outside the 371-nucleotide region deleted from the LAT gene of dLAT371.

Phenotypic properties of herpes simplex virus 1 containing a derepressed open reading frame P gene

Open reading frame P (ORF P) maps in the viral DNA sequences transcribed during latency and is located antisense to the gamma 1 34.5 gene. Earlier studies have shown that the expression of ORF P is repressed by an infected cell protein no. 4 binding site straddling the transcription initiation site. We have made monospecific polyclonal antibodies to the protein and constructed a virus, designated ORF P++, in which the infected cell protein no. 4 binding site has been mutagenized, thereby allowing full expression of an unmodified ORF P gene from its natural promoter. We report the following findings. (i) The native protein forms multiple bands on denaturing polyacrylamide gels suggestive of extensive processing and aggregation of the protein; (ii) the protein accumulates in the nucleus in rod-shaped structures perpendicular to the axis of attachment of the infected cell to the solid matrix; (iii) the virus was highly attenuated on inoculation into mice by the intracerebral or ocular route, and virus was not recovered upon explantation of trigeminal ganglia; (iv) although protein synthesis was not prematurely shut off in the human neuroblastoma cell line SK-N-SH, gamma 1 34.5 protein was not detected in immunoblasts. Analyses of electrophoretically separated denatured RNAs indicated that in cells infected with the ORF P++ virus, there was a large increase in the amount of ORF P RNA and a corresponding decrease in the amount of gamma 1 34.5 RNA. We conclude that either the overproduction of ORF P protein blocks the expression of some herpes simplex virus 1 genes or derepression of the transcription of ORF P has a negative effect on the transcription of the antisense gamma 1 34.5 RNA.

Downstream regulatory elements increase acute and latent herpes simplex virus type 2 latency-associated transcript expression but do not influence recurrence phenotype or establishment of latency

The role of putative promoter or activator sequences downstream of the herpes simplex virus type 2 latency-associated transcript (LAT) promoter and upstream of the LAT intron was investigated in vivo by constructing and evaluating mutant viruses with deletions in this region. The deletion of LAT promoter sequences upstream of the primary LAT transcript reduced levels of LAT expression during productive infections, compared with the LAT expression level of wild-type virus, and abolished LAT expression during latency. The deletion of the putative downstream regulatory elements reduced but did not eliminate LAT expression during productive and latent infections. The deletion of both regions almost completely eliminated acute LAT transcription, although additional acute LAT-region transcription directed by sequences upstream of either region was detected by reverse transcriptase PCR. The deletion of the downstream elements did not influence the ability of the virus to reactivate from latently infected guinea pigs relative to the ability of the wild-type virus to reactivate; thus, decreased LAT expression did not affect the frequency of recurrence. The deletion of both regions did not affect the ability of the virus to establish latency. We conclude that downstream regulatory elements are necessary for maximal acute LAT expression but do not constitute an independent promoter during latency and do not play an obvious role in the establishment of our reactivation from latency.

The spontaneous reactivation function of the herpes simplex virus type 1 LAT gene resides completely within the first 1.5 kilobases of the 8.3-kilobase primary transcript

The herpes simplex virus type 1 (HSV-1) latency-associated transcript (LAT) gene is essential for efficient spontaneous reactivation of HSV-1 from latency. We report here that although the LAT gene is 8.3 kb in length, the first 1.5 kb of the LAT gene alone is sufficient for wild-type levels of spontaneous reactivation. We began with a LAT deletion mutant of HSV-1 strain McKrae in which the LAT promoter and the first 1.6 kb of the 5' end of the LAT gene had been deleted from both copies of LAT (one in each viral long repeat). As we previously reported, this mutant (dLAT2903) was significantly impaired for spontaneous reactivation (G. C. Perng, E. C. Dunkel, P. A. Geary, S. M. Slanina, H. Ghiasi, R. Kaiwar, A. B. Nesburn, and S. L. Wechsler, J. Virol. 68:8045-8055, 1994). We then inserted the LAT promoter and the first 1.5 kb of the LAT gene into a location in the unique long region of dLAT2903 far removed from the normal location of LAT in the long repeats. This resulted in a virus (LAT15a) whose capacity for transcribing LAT RNA was limited to the first 1.5 kb of the 8.3-kb LAT primary transcript. Rabbits were ocularly infected with this mutant, and spontaneous reactivation was measured in comparison to those of the original LAT-negative mutant and its marker-rescued (wild-type) virus, dLAT2903R. LAT15a had an in vivo spontaneous reactivation rate of 12%, compared with a rate of 11% for the marker-rescued virus and 0% for the LAT-negative virus. Southern analysis confirmed that the spontaneously reactivated LAT15a virus retained the original deletions in both copies of LAT and the 1.5-kb LAT insertion in the unique long region. Thus, insertion of the first 1.5 kb of LAT (and its promoter) at a site distant from the normal LAT location appeared to completely restore in vivo spontaneous reactivation to wild-type levels, despite the remaining inability of the original LAT genes to transcribe any LAT RNA. The function of LAT involved in efficient spontaneous reactivation therefore appeared to map completely within the first 1.5 kb of the LAT gene.

The region of the herpes simplex virus type 1 LAT gene that is colinear with the ICP34.5 gene is not involved in spontaneous reactivation

The goal of this report was to determine if the region of the LAT gene that is colinear with ICP34.5 (kb 6.2 to 7.1 of LAT) is involved in spontaneous reactivation of herpes simplex virus type 1. We inserted one copy of the ICP34.5 gene into the unique long region of a herpes simplex virus type 1 (strain McKrae) mutant lacking both copies of ICP34.5 (one in each viral long repeat) and the corresponding 917-nucleotide colinear portion of LAT (kb 6.2 to 7.1). Rabbits were ocularly infected with this mutant, and spontaneous reactivation relative to that for the wild-type virus and the original mutant was measured. As we previously reported, the original ICP34.5-deleted virus (d34.5) was significantly impaired for spontaneous reactivation and virulence (G. C. Perng, R. L. Thompson, N. M. Sawtell, W. E. Taylor, S. M. Slanina, H. Ghiasi, R. Kaiwar, A. B. Nesburn, and S. L. Wechsler, J. Virol. 69:3033-3041, 1995). In contrast, we report here that restoration of one copy of ICP34.5 at a distant location completely restored the wild-type level of in vivo spontaneous reactivation, despite retention of the deletion in LAT (spontaneous reactivation rate = 0.3 to 1.4% for the ICP34.5 deletion mutant, 7.7 to 19.6% for the wild type, and 9 to 16.1% for virus with one copy of ICP34.5). Thus, the 917-nucleotide region of LAT from kb 6.2 to 7.1 was not involved in the LAT function required for wild-type spontaneous reactivation. We also found that restoration of a single ICP34.5 gene in a novel location did not restore wild-type virulence (rabbit death rate = 0% [0 of 15] for the original ICP34.5 deletion mutant, 8% [2 of 24] for the single-copy IPC34.5 virus, and 52% [14 of 27] for wild-type virus; P < 0.001 for one versus two copies of ICP34.5). It is likely that either two gene doses of ICP34.5 or its location in the long repeat is essential for full functionality of ICP34.5's virulence function. Furthermore, the ability of the single-copy ICP34.5 virus to reactivate at wild-type levels despite being significantly less virulent than wild-type virus separates the spontaneous reactivation phenotype from the virulence phenotype.

Two herpes simplex virus type 1 latency-active promoters differ in their contributions to latency-associated transcript expression during lytic and latent infections

Herpes simplex virus type 1 (HSV-1) establishes latency in human sensory ganglia, during which time the viral genome is transcriptionally silent with the exception of the latency-associated transcripts (LATs). The most abundant LAT is a 2-kb RNA whose biosynthesis is poorly characterized. The 2-kb LAT may be a primary transcript, or its synthesis may involve splicing and/or other forms of processing. Two potential RNA polymerase II promoters (LAP1 and LAP2) upstream of the 2-kb LAT 5' end have been identified. To investigate the role played by LAP1 and LAP2 in the synthesis of the 2-kb LAT under lytic and latent conditions, we analyzed HSV-1 mutants which contain deletions of one or both of these promoters. During lytic infection in cell culture, the cis elements critical for the normal accumulation of the 2-kb LAT were mapped to LAP2, while LAP1 sequences were largely dispensable. The 5' ends of the major 2-kb LATs produced by the wild-type and LAP deletion viruses were examined by primer extension analysis and were all found to be identical (+/- 2 bp). The accumulation of the 2-kb LAT during latent infections of murine trigeminal ganglia was examined by Northern (RNA) blot and by reverse transcription-PCR. In contrast to the results found in lytic infections, the critical cis elements needed for 2-kb LAT accumulation during latency were mapped to LAP1. Deletion of LAP1 resulted in a 500-fold reduction in 2-kb LAT accumulation, whereas deletion of LAP2 resulted in only a 2- to 3-fold reduction. Deletion of both LAP1 and LAP2 resulted in undetectable levels of the 2-kb LAT. Our results indicate that both LAP1 and LAP2 are critical for 2-kb LAT expression but under different conditions. LAP1 is essential for LAT expression during latency, while LAP2 is primarily responsible for LAT expression in lytic infections in cell culture. LAP1 and LAP2 may prove to be functionally independent promoter elements that control 2-kb LAT expression during different stages of HSV-1 infections.

An activity specified by the osteosarcoma line U2OS can substitute functionally for ICP0, a major regulatory protein of herpes simplex virus type 1

Among the five immediate-early regulatory proteins of herpes simplex virus (HSV) type 1, only ICP0 is capable of activating all kinetic classes of viral genes. Consistent with its broad transactivating activity, ICP0 plays a major role in enhancing the reactivation of HSV from latency both in vivo and in vitro. Although not essential for viral replication, ICP0 confers a significant growth advantage on the virus, especially at low multiplicities of infection. In this report we describe the expression of a novel activity by the osteosarcoma cell line U2OS that can substitute functionally for ICP0. Compared with Vero cells, both U2OS cells and cells of the ICP0-expressing line 0-28 significantly enhanced the plating efficiency of an ICP0 null mutant, 7134. In contrast, the plating efficiencies of the wild-type virus in all three cell types were similar. Single-step growth experiments demonstrated that the yield of 7134 in U2OS cells was severalfold higher than that in 0-28 cells and about 100-fold higher than that in Vero cells. In order to identify the viral genes whose expression is enhanced by the activity in U2OS cells, levels of expression of selected viral proteins in extracts of Vero and U2OS cells were compared by Western blot (immunoblot) analysis following low-multiplicity infection. At a multiplicity of 0.1 PFU per cell, the levels of expression of the immediate-early protein ICP4 and the early protein gD in 7134-infected U2OS cells were significantly higher than those in 7134-infected Vero cells. When infections were carried out at a multiplicity of 1 PFU per cell, however, no major differences in the levels of expression of these proteins in U2OS and Vero cells were observed. Cycloheximide reversal experiments demonstrated that the cellular activity expressed in U2OS cells that promotes high-level expression of ICP4 is not synthesized de novo but appears to exist as a preformed protein(s). To confirm this observation and to determine whether, like immediate-early genes, early, delayed-early, and late viral genes are also responsive to the cellular activity, transient-expression assays were performed. The results of these tests demonstrated that basal levels of expression from immediate-early, early, and delayed-early promoters, but not that from a late promoter, were significantly higher in U2OS cells than in Vero cells and that this enhancement occurred in the absence of viral proteins.(ABSTRACT TRUNCATED AT 400 WORDS)

Experimental herpes simplex virus type 1 (HSV-1) infection of the spinal cord and dorsal root ganglia

Host factors determining the outcome of herpes simplex virus type 1 (HSV-1) infection within neurons are poorly understood. This paper aims to identify regional differences in the behaviour of HSV-1 within the nervous system as an approach to investigating the role of the host environment in determining the outcome of infection. We describe a mouse model of HSV infection focused on motor neurons of the spinal cord, resulting from intramuscular injection (i.m.) and compare this with the behaviour of virus within sensory neurons following scarification of virus on to skin. Viral antigen was detectable immunohistochemically by 2 days in both models and disappeared by 9-11 days. The time course of acute infection was reflected in the i.m. group by quantitative plaque assay for virus. Inflammation and cell destruction occurred in both models, but clinical features and histological destruction were greater in the group infected via the intramuscular route. In the sensory ganglia, a latent state from which virus could be reactivated by explanation, was established with LATS expression detectable in many neurons at 35 days post-infection (p.i.), but not in non-neuronal cells. Expression of latency associated transcript (LATS) was detected in motor neutrons in spinal cords at 35 days p.i. providing evidence for establishment of a LATS-positive latent state at this site, and continued to be detectable up to 6 months post-infection. In addition, LATS was detected in white matter at late times, suggesting a non-neuronal site of latency. In contrast to the behaviour in sensory ganglia, induced reactivation from spinal cords, by explanation and nerve section, was a very rare event. We have shown that a LATS-positive latent state can be established within motor neurons of the CNS, but that there are regional differences in the biology and outcome of infection between the CNS and peripheral nervous system. We propose that this may be a useful model to study reproducibly, the behaviour of HSV-1 in a CNS environment and, by comparison with sensory ganglion infection, to explore host factors which may underlie these regional differences. The relevance of this model for using HSV-1 as a therapeutic vector for motor neurons is also discussed.

Analysis of the promoter and cis-acting elements regulating expression of herpes simplex virus type 2 latency-associated transcripts

In latently infected neurons, herpes simplex virus type 2 (HSV-2) expresses one abundant family of transcripts, the latency-associated transcripts (LATs). We demonstrate here that the sequence lying about 700 bp upstream of the 5' end of the HSV-2 major LAT acts as a very strong promoter in transient expression assays in both neuronal and nonneuronal cells. Transcription starts about 27 to 32 bp downstream of a functional TATA box. The proximal fragment from -102 to +34 includes the basal promoter and accounts for constitutive transcriptional activity in various cell lines. The distal region from -392 to -103 contributes to particularly strong promoter activity in neuronal cell lines and involves multiple cis-acting elements. A functional activating transcription factor/cyclic AMP (cAMP) response element binding protein motif lies just upstream of the TATA. By DNase I footprint and methylation protection assays, we identified several additional protein-binding sites upstream of the activating transcription factor/cAMP response element binding protein motif. A GC-rich element, termed LAT-3, was located between bases -128 to -102. A 2-bp substitution in LAT-3 markedly reduced promoter activity and abolished protein-binding ability in vitro. Gel retardation assay showed no competition for protein binding to LAT-3 by other GC-rich elements. LAT-3 appears to be a novel cis-acting element that may contribute to the neuronal responsiveness of the HSV-2 LAT promoter.

Quantification of transcripts from the ICP4 and thymidine kinase genes in mouse ganglia latently infected with herpes simplex virus

Herpes simplex virus establishes latency in nervous tissue in which it is maintained for the life of the mammalian host, with occasional reactivation leading to subsequent spread. Latency-associated transcripts are abundant during latency, but viral proteins and productive cycle RNAs have not been detected. Using sensitive, quantitative PCR assays, we have quantified certain viral RNAs specific to productive-cycle genes in mouse ganglia latently infected with herpes simplex virus type 1. Sense-strand RNA specific to the essential immediate-early gene, ICP4, was present in most ganglia in variable amounts relative to the amount of viral DNA, with one to seven molecules of RNA per viral genome in about 20% of ganglia. In contrast, the amount of latency-associated transcripts was much less variable, at an average of 4 x 10(4) molecules per viral genome. The amounts of ICP4-specific RNA were similar at 30 and 60 days postinfection, and at least some of these transcripts initiated within a region consistent with utilization of the ICP4 promoter. RNA specific to the thymidine kinase gene, whose transcription in productive infection is dependent on ICP4, was present in latently infected ganglia at a maximum level of 3.2 x 10(6) molecules per ganglion (500 molecules per viral genome). ICP4-specific and tk-specific RNAs measured from the same samples showed a positive correlation extending over 2 orders of magnitude. We conclude that ICP4-specific RNA is expressed in the absence of detectable reactivation and discuss possible implications of our findings for latent gene expression.

Expression of the herpes simplex virus type 2 latency-associated transcript enhances spontaneous reactivation of genital herpes in latently infected guinea pigs

The latency-associated transcript (LAT) is the only herpes simplex virus (HSV) gene product detectable in latently infected humans and animals. In this report, we show that a 624-bp deletion in the promoter of the HSV-2 LAT had no discernable effect on viral growth in tissue culture or in acute genital infection of guinea pigs, but impaired LAT accumulation and led to a marked decrease in spontaneous genital recurrences when compared with the behavior of wild-type and rescuant strains. Differences in the ability of the mutant to replicate, or in how readily it established or maintained latency did not account for this finding. Thus, HSV LAT expression facilitates the spontaneous reactivation of latent virus.