Positive control of the herpes simplex virus thymidine kinase gene requires upstream DNA sequences

A simple procedure for expression and purification of selected non-structural (alpha and beta) herpes simplex virus 1 (HSV-1) proteins

The expression and isolation of herpes simplex virus 1 (HSV-1) immediate early (alpha) IE63 (ICP27) and of the early (beta) thymidine kinase (Tk) polypeptides in Escherichia coli JM 109 cells transformed with the PinPoint Xa-1 (Promega) plasmid construct carrying either the HSV-1 UL54 or UL23 genes are described. The resulting biotinylated fusion protein(s) could be easily induced and were purified in appropriate amounts by means of a monomeric avidin-conjugated resin (SoftLink Soft Release Avidin Resin, Promega) provided that: (1) the exponential growth of the selected transformed cells was monitored carefully; (2) the post-induction harvest interval was properly chosen; and (3) the period for adsorption to the avidin resin suitably adjusted. The isolated protein(s), although partially digested in the case of the IE63 polypeptide, were suitable antigen(s) for immunization of various animal species. Co-purification of trace amounts of endogenous biotinylated protein(s) produced in E. coli was eliminated by shortening the duration of adsorption to the avidin resin.

Activation of gene expression by herpes simplex virus type 1 ICP0 occurs at the level of mRNA synthesis

ICP0 is a nuclear phosphoprotein involved in the activation of herpes simplex virus type 1 (HSV-1) gene expression during lytic infection and reactivation from viral latency. Although available evidence suggests that ICP0 acts at the level of transcription, definitive studies specifically addressing this issue have not been reported. In the present study we measured the ability of ICP0 to activate gene expression (i) from promoters representing the major kinetic classes of viral genes in transient expression assays and (ii) from the same promoters during viral infection at multiplicities of infection ranging from 0.1 to 5.0 PFU/cell. The levels of synthesis and steady-state accumulation of mRNA, mRNA stability, and levels of protein synthesis were compared in cells transfected with a reporter plasmid in the presence and absence of ICP0 and in cells infected with wild-type HSV-1 or an ICP0 null mutant, n212. In transient expression assays and during viral infection at all multiplicities tested, the levels of steady-state mRNA and protein were significantly lower in the absence of ICP0, indicating that ICP0 activates gene expression at the level of mRNA accumulation. In transient expression assays and during infection at low multiplicities (< 1 PFU/cell) in the presence or absence of ICP0, marked increases in the levels of viral mRNAs accompanied by proportional increases in the levels of protein synthesis were observed with increasing multiplicity. At a high multiplicity (5 PFU/cell) in the presence or absence of ICP0, mRNA levels did not increase as a function of multiplicity and changes in the levels of protein were no longer related to changes in the levels of mRNA. Collectively, these tests indicate that transcription of viral genes is rate limiting at low multiplicities and that translation is rate limiting at high multiplicities, independent of ICP0. Consistent with the lower levels of mRNA detected in the absence of ICP0, the rates of transcription initiation measured by nuclear run-on assays were uniformly lower in cells infected with the ICP0 null mutant at all multiplicities tested, implying that ICP0 enhances transcription at or before initiation or both. No evidence was found of posttranscriptional effects of ICP0 (i.e., effects on the stability of mRNA, nuclear-cytoplasmic distribution, polyribosomal mRNA distribution, or rates of protein synthesis). Taken together, these results suggest that ICP0 activates gene expression prior to or at the level of initiation of mRNA synthesis in transient expression assays and during viral infection. Based on these findings; we hypothesize that the exaggerated multiplicity-dependent growth phenotype characteristic of ICP0 null mutants reflects the requirement for ICP0 under conditions where the steady-state level of mRNA is rate limiting, such as during low-multiplicity infection and reactivation from latency.

Herpes simplex type 1:lacZ recombinant viruses. I. Characterization and application to defining the mechanisms of action of known antiherpes agents

Recombinant viruses with the lacZ gene placed under the control of the HSV-1 ICP4, TK and gD regulatory regions were constructed by recombination into the TK locus of HSV-1. Difficulty in isolating ICP4 and gD recombinant viruses with high level, regulated expression of beta-galactosidase was overcome by the use of HSV-1 translational initiation sequences of these genes in place of vector-derived sequences. beta-Galactosidase expression displayed the kinetics particular to each viral class. The maximal expression of beta-galactosidase from the recombinant viruses within a 22-h period (m.o.i. 5) (relative to the ICP4 virus) was gD(3) > gC(2) > ICP4(1) > TK(0.5). The ICP4 virus produces easily quantifiable levels of beta-galactosidase activity for multiplicities of infection from 5 x 10(-4) through 5 over 48 h postinfection. At multiplicities of infection between 2 and 5, ICP4-driven activity was measurable within 2 h postinfection from a monolayer of 3 x 10(4) Vero cells in microtiter wells. Mechanisms of inhibition of several antivirals were probed by using the regulated expression of beta-galactosidase from the ICP4 virus as a marker for viral growth. An experimental antiviral (E3925, IC50 1 microgram/ml) and a neutralizing gD MAb (DUP55306, IC50 0.6 microgram/ml) acted prior to immediate early synthesis, consistent with inhibition of viral entry or uncoating. IFN-gamma inhibited expression of immediate-early synthesis, while having no effect on viral entry. IC50 values for E3925 obtained using either the ICP4 or gD viruses at m.o.i. 0.005, were in good agreement with those obtained by standard plaque assays, but were determined in only 1 day, using a microtiter plate format. Thus, these reporter viruses are useful tools for defining the mechanisms of action of antiherpes agents, while quantitatively reproducing the results for IC50 determinations from standard plaque assays within 24 h in a microtiter plate format.

A truncated herpes simplex virus thymidine kinase phosphorylates thymidine and nucleoside analogs and does not cause sterility in transgenic mice

Dividing eukaryotic cells expressing the herpes simplex virus type 1 thymidine kinase (TK) gene are sensitive to the cytotoxic effect of nucleoside analogs such as acyclovir or ganciclovir (GCV). Transgenic mice with cell-targeted expression of this conditional toxin have been used to create animals with temporally controlled cell-specific ablation. In these animal models, which allow the study of the physiological importance of a cell type, males are sterile. In this study, we showed that this phenomenon is due to testis-specific high-level expression of short TK transcripts initiated mainly upstream of the second internal ATG of the TK gene. This expression is DNA methylation independent. To obtain a suicide gene that does not cause male infertility, we generated and analyzed the properties of a truncated TK (delta TK) lacking the sequences upstream of the second ATG. We showed that when expressed at sufficient levels, the functional properties of delta TK are similar to those of TK in terms of thymidine or GCV phosphorylation. This translated into a similar GCV-dependent toxicity for delta TK- or TK-expressing cells, both in vitro and in transgenic mice. However, delta TK behaved differently from TK in two ways. First, it did not cause sterility in delta TK transgenic males. Second, low-level delta TK RNA expression did not confer sensitivity to GCV. The uses of delta TK in cell-specific ablation in transgenic mice and in gene therapy are discussed.

Transient expression of deletion mutants of the herpes simplex virus thymidine kinase-encoding gene in mouse fibroblast cells

Previous studies have shown that at least three polypeptides of 43, 39 and 38 kDa are translated from separate AUG codons of the thymidine kinase (TK) encoding mRNA of herpes simplex virus type 1. In addition, small tk-specific transcripts initiated within the tk coding region were observed. However, functional activity of these three proteins and their role in establishing of the TK+ cell phenotype is not yet clear. In order to locate the 5' boundary of the gene encoding functionally active TK, we constructed a set of deletion mutants with truncated 5' ends and examined their ability to provide a TK+ phenotype after microinjection into nuclei of LTK- cells. The results demonstrate that nucleotide sequences upstream from the second ATG codon can be removed without affecting the TK+ phenotype. Deletion of the second start codon and its downstream region inactivates the TK function. Those deletion mutants which contain only the third ATG codon are TK-. Thus, the 38-kDa polypeptide that initiates at the third start codon is not endowed with the TK+ activity. Constructs containing deletions up to nt +210 and lacking all 5'-end canonical and aberrant transcription control regions, as well as first start codon, can provide the TK+ function.

Delineation of regulatory domains of early (beta) and late (gamma 2) genes by construction of chimeric genes expressed in herpes simplex virus 1 genomes

The expression of the gamma 2 class of viral genes in cells infected with herpes simplex virus 1 requires viral DNA synthesis and functional viral products made earlier in infection. To identify the sequences required for gamma 2 gene expression, we constructed recombinant viruses in which regions of the thymidine kinase gene (tk), a beta gene normally expressed early in infection, were replaced by specific domains of a gamma 2 gene. The phenotypic attributes examined were (i) sensitivity or resistance of expression in cells exposed to sufficient phosphonoacetate to block viral DNA synthesis, properties of gamma 2 and beta genes, respectively; (ii) expression early in infection, a property of beta genes; and (iii) expression late in infection, a property of gamma 2 genes. We report that replacement of nucleotides -200 to +51 of tk with nucleotides -77 to +104 of the gamma 2 gene conferred upon tk all of the tested attributes of gamma 2 genes. The tk sequence in the 5' transcribed noncoding domain downstream of nucleotide +51 played no apparent role in the expression of the chimeric genes. Similarly, tk sequence downstream of -16 and gamma 2 sequence upstream of -12, when juxtaposed in correct orientations, yielded a chimeric gene that was poorly expressed. In contrast, the chimeric gene consisting of tk sequence upstream of -16 fused to gamma 2 sequence downstream of -12 had the attributes of both beta and gamma 2 genes in that it was expressed both early and late in infection and was partially resistant to phosphonoacetate. The capacity for expression late in infection encoded in the gamma 2 5' transcribed noncoding domain was observed in cells infected with a recombinant virus in which gamma 2 nucleotides +17 to +104 were inserted into the 5' transcribed noncoding domain of the tk gene. We conclude that whereas in the beta genes exemplified by the tk gene the regulatory domains are mainly upstream from nucleotide +51, the sequence(s) that confers gamma 2 regulation is downstream from the TATAA box.

Effect of an amber mutation in the herpes simplex virus thymidine kinase gene on polypeptide synthesis and stability

KG111 is a mutant of herpes simplex virus (HSV), strain KOS, that exhibits temperature-dependent drug resistance. For example, it is almost as resistant as a thymidine kinase (tk)-deficient virus at 39 degrees, but is relatively sensitive to acyclovir at 34 degrees, Using marker transfer techniques, we have mapped the mutation conferring temperature-dependent drug resistance in KG111 to the 5' portion of the tk gene. Sequencing of this region revealed an amber mutation at codon 44, which lies between the first and second methionine codons of the tk polypeptide. This mutation is identical to that found in TK4, an HSV mutant derived from Cl 101 (L. Haarr et al., 1985, J. Virol. 56, 512-519). Analyses of immunoprecipitated tk proteins from KG111- and TK4-infected cells showed that KG111 and TK4 do not synthesize full-length tk polypeptides, but instead produce a truncated form of the protein. Small amounts of a similar truncated tk polypeptide are also produced in wild-type-infected cells and are thought to arise from initiation at a downstream AUG. The relative amounts and size of the mutant tk proteins compared with those of the wild-type are consistent with the amber mutation eliminating translation of full-length polypeptide and causing a four- to fivefold increase in the utilization of downstream AUG codons for initiation. The truncated polypeptides specified by KG111 and TK4 are less stable than the full-length polypeptide at 39 degrees, which may contribute to the conditional drug-resistant phenotype. On the other hand, the truncated polypeptides normally expressed by wild-type virus at low levels and the more highly expressed truncated tk polypeptides from a deletion mutant are relatively stable at 39 degrees. These results suggest that stability of the truncated tk polypeptide is influenced by the amount of tk present.

The regulation of SV40 early gene expression in embryonal carcinoma stem cells--faithful transcriptional regulation in vitro

We have derived from F9 murine embryonal carcinoma stem cells, and from differentiated derivatives induced with retinoic acid and cAMP, whole cell extracts which efficiently and accurately transcribe a variety of supercoiled DNA templates in vitro. These extracts and control elements from viral genomes have been used to study changes in transcriptional activity which accompany differentiation. The SV40 enhancer is inefficiently utilized in stem cells but is activated upon differentiation to parietal endoderm and the in vitro systems mimick this regulation. Mixing experiments demonstrate that the differentiated cell phenotype is dominant, suggesting that stem cells contain limiting amounts of factors required for enhancer activity. Our results may explain the developmental regulation of cellular genes with enhancer motifs in their control sequences.

Regulated expression of stably transfected herpes simplex virus thymidine kinase genes in continuous cell lines expressing a temperature-sensitive mutant form of the immediate-early protein ICP4

We stably transfected the herpes simplex virus type 1 thymidine kinase gene into a continuous cell line expressing a temperature-sensitive form of the viral immediate-early protein ICP4. In these cells, expression of the thymidine kinase gene was regulated in a temperature-sensitive manner, partially reproducing the controls that operate during a viral infection.

A specific 15-bp TATA box promoter element is required for expression of a herpes simplex virus type 1 late gene

The herpes simplex virus type 1 (HSV-1) glycoprotein C (gC) gene is a true late or gamma 2 gene in that its expression shows a strict requirement for viral DNA replication. Elements required for regulated expression of this gene were previously shown to consist of the gC TATA box, transcription start site and a large portion of the leader sequence of the gC gene. In this paper we show that transcription of the gC gene requires a 15-bp sequence, GGGTATAAATTCCGG, which contains the gC TATA box. This sequence contains specific promoter elements because replacement of this sequence with either the TATA box of the HSV-1 early thymidine kinase (tk) gene or two random TATA-like elements results in a transcriptionally inactive gC gene. In addition, we show that temporal expression of HSV beta and gamma genes at early and late times during infection are controlled by separate and distinct regulatory elements; regulatory signals distal to the TATA box are needed for early expression, whereas a gC-like TATA box is needed for late expression. These signals were identified by construction of a chimeric HSV gene that contained the distal control signals of the beta tk gene fused upstream of the TATA sequence of the gamma 2 gC gene. When RNA was isolated at various times postinfection from cells infected with a virus whose genome contained this chimeric tk-gC gene, synthesis of gC mRNA showed both early and late kinetics.

Expression of a cellular gene cloned in herpes simplex virus: rabbit beta-globin is regulated as an early viral gene in infected fibroblasts

We constructed nondefective herpes simplex virus type 1 recombinants bearing the intact rabbit beta-globin gene inserted into the viral gene for thymidine kinase to study the expression of a cellular gene when it is present in the viral genome during lytic viral infections. The globin promoter was activated to high levels during productive infection of Vero cells, giving rise to properly spliced and processed cytoplasmic globin transcripts. Expression of globin RNA occurred with early kinetics, was not affected by blocking viral DNA replication, and was strongly inhibited by preventing viral immediate-early protein synthesis with cycloheximide. These results support the hypothesis that temporal control of herpes simplex virus early gene expression is accomplished by mechanisms that are not restricted to viral promoters. In addition, these data show that a cellular transcript can be correctly processed and can accumulate to high levels during viral infection; this indicates that the mechanisms of virally induced shutoff of host RNA accumulation and degradation of host mRNAs do not depend on sequence-specific differentiation between host and viral RNAs. These findings also suggest that herpesviruses have considerable potential as high-capacity gene transfer vectors for a variety of applications.

Regulation of the herpes simplex virus type 1 late (gamma 2) glycoprotein C gene: sequences between base pairs -34 to +29 control transient expression and responsiveness to transactivation by the products of the immediate early (alpha) 4 and 0 genes

The glycoprotein C (gC) gene of herpes simplex virus type 1 is a true late gene, in that its expression occurs late in infection with a strict requirement for viral DNA replication. Recently, we reported on gC expression during infection with mutant viruses carrying deletions in the gC gene promoter. Analysis of RNA extracted from cells infected with individual mutants showed that the DNA sequences required for regulated expression of this late gene lie within bases -34 to +124 relative to the 5' end of the mRNA. In the present study, the deleted gC promoter sequences were fused to the bacterial chlorampheniol acetyltransferase (CAT) gene and expression was measured in short-term transfection assays after transactivation by infection with HSV or cotransfection with a second plasmid carrying the immediate early genes of HSV-1. The 63 base pair sequence located between -34 to +29 on the gC promoter was sufficient to give induction of CAT activity following infection and on cotransfection with plasmids which code for the immediate early gene products ICP4 and ICPO. This 63 base pair region contains the TATA homology and the transcriptional start site of the gC gene, and apparently contains specific promoter elements not found in a similar region of the HSV TK promoter. This was shown by removing the distal upstream region of the TK promoter, 5' to -37, and found that the TK gene was no longer activated by infection or cotransfection with an alpha 4-alpha 0 gene containing plasmid.

Transcriptional control signals of a herpes simplex virus type 1 late (gamma 2) gene lie within bases -34 to +124 relative to the 5' terminus of the mRNA

The cis-acting DNA sequences required for regulated expression of a herpes simplex virus type 1 (HSV-1) late (gamma 2) gene were studied by using viruses containing specific deletions in the 5' transcribed noncoding and upstream regions of the HSV-1 glycoprotein C (gC) gene, a model gamma 2 gene. Nine mutant viruses which had variable 5' and 3' deletions within bases -569 to +124 relative to the 5' terminus of the gC mRNA were isolated. The mutants were isolated by a simple in situ hybridization screening procedure not requiring any prior selective pressure for or against expression of the gC gene. Analysis of RNA extracted from cells infected with individual mutants showed that the DNA sequences required for regulated expression of this gamma 2 gene lay within bases -34 to +124. This 158-base-pair fragment was sufficient to confer accurate and quantitative expression of gC mRNA and to maintain the stringent requirement on viral DNA replication for expression of this gene. Moreover, it was found that sequences located between -34 and +14 contained signals essential for expression of gC. To determine whether the -34 to +124 sequences would function as a gamma 2 promoter when moved to another region of the HSV-1 genome, the 158-base-pair fragment was substituted for the normal thymidine kinase promoter-regulatory sequences in the thymidine-kinase gene locus. Transcription of this chimeric gene was regulated as a gamma 2 gene in that its expression in infected cells was dependent on viral DNA synthesis. The only recognizable consensus sequence upstream of the transcription initiation site for this gene was the TATAAA sequence at -30.

Transcriptional control signals of a herpes simplex virus type 1 late (gamma 2) gene lie within bases -34 to +124 relative to the 5' terminus of the mRNA

The cis-acting DNA sequences required for regulated expression of a herpes simplex virus type 1 (HSV-1) late (gamma 2) gene were studied by using viruses containing specific deletions in the 5' transcribed noncoding and upstream regions of the HSV-1 glycoprotein C (gC) gene, a model gamma 2 gene. Nine mutant viruses which had variable 5' and 3' deletions within bases -569 to +124 relative to the 5' terminus of the gC mRNA were isolated. The mutants were isolated by a simple in situ hybridization screening procedure not requiring any prior selective pressure for or against expression of the gC gene. Analysis of RNA extracted from cells infected with individual mutants showed that the DNA sequences required for regulated expression of this gamma 2 gene lay within bases -34 to +124. This 158-base-pair fragment was sufficient to confer accurate and quantitative expression of gC mRNA and to maintain the stringent requirement on viral DNA replication for expression of this gene. Moreover, it was found that sequences located between -34 and +14 contained signals essential for expression of gC. To determine whether the -34 to +124 sequences would function as a gamma 2 promoter when moved to another region of the HSV-1 genome, the 158-base-pair fragment was substituted for the normal thymidine kinase promoter-regulatory sequences in the thymidine-kinase gene locus. Transcription of this chimeric gene was regulated as a gamma 2 gene in that its expression in infected cells was dependent on viral DNA synthesis. The only recognizable consensus sequence upstream of the transcription initiation site for this gene was the TATAAA sequence at -30.

Promoter domains required for expression of plasmid-borne copies of the herpes simplex virus thymidine kinase gene in virus-infected mouse fibroblasts and microinjected frog oocytes

A transient expression assay was used to measure the relative template activities of mutated tk genes in mouse L cells induced in trans by herpes simplex virus (HSV). In this assay, expression of the wild-type HSV type 1 tk gene is induced at least 200-fold by the superinfecting virus. Genetic lesions that were assayed include 5' deletions, clustered base substitutions, single base substitutions, intrapromoter inversions, and intrapromoter recombinants with the HSV type 2 tk gene. Roughly half of the mutations that were tested were found to weaken tk expression efficiency, and the remaining mutations did not alter expression. The spatial distribution of mutations that reduce expression efficiency in trans-induced mouse fibroblasts facilitated the construction of a map of promoter domains. The most gene-proximal promoter domain is located between 16 and 32 base pairs (bp) upstream of the tk mRNA cap site and contains a TATA homology. Two more distally located promoter domains were mapped to discrete locations upstream from the TATA homology. One of these distal domains is located between 47 and 79 bp upstream from the mRNA cap site, and the other is located between 84 and 105 bp upstream from the tk gene. The boundaries of these three promoter domains, with one exception, coincided with the set of domains delineated previously in a frog oocyte microinjection assay. The concordant behavior of tk promoter mutants in microinjected frog oocytes and trans-induced mouse fibroblasts leads us to propose that recognition and activation of the HSV tk promoter is mediated by cellular transcription factors that are common to frogs and mice.

A genetic approach to promoter recognition during trans induction of viral gene expression

Viral infection of mammalian cells entails the regulated induction of viral gene expression. The induction of many viral genes, including the herpes simplex virus gene encoding thymidine kinase (tk), depends on viral regulatory proteins that act in trans. Because recognition of the tk promoter by cellular transcription factors is well understood, its trans induction by viral regulatory proteins may serve as a useful model for the regulation of eukaryotic gene expression. A comprehensive set of mutations was therefore introduced into the chromosome of herpes simplex virus at the tk promoter to directly analyze the effects of promoter mutations on tk transcription. The promoter domains required for efficient tk expression under conditions of trans induction corresponded to those important for recognition by cellular transcription factors. Thus, trans induction of tk expression may be catalyzed initially by the interaction of viral regulatory proteins with cellular transcription factors.

A cis-acting element within the 5' leader of a cytomegalovirus beta transcript determines kinetic class

During cytomegalovirus (CMV) infection, gene expression is regulated by transcriptional and posttranscriptional events. Although recent studies have established that posttranscriptional controls are important determinants of gene expression in several eukaryotic systems, the precise signals and mechanisms have not been clearly identified. We present evidence for a cis-acting signal, contained within the 5' leader region of a CMV beta (or early) gene, that acts posttranscriptionally in the regulation of gene expression. Addition of this signal to an alpha (or immediate early) gene construct converted expression of the indicator protein to the beta class, even though the gene remained under alpha transcriptional control. Deletion of a portion of the beta gene leader sequence (nucleotides +62 to +142) reverted expression to the alpha class. This cis-dominant signal appears to act by blocking expression posttranscriptionally until a viral function activates full gene expression at the appropriate time in infection.

Regulation of cytomegalovirus late gene expression: gamma genes are controlled by posttranscriptional events

We investigated the control of human cytomegalovirus (CMV) late (gamma)-gene expression in human fibroblast cells. Transcriptional activity of two gamma genes, encoding ICP27, a structural component (matrix or tegument) of virions, and ICP36, a major DNA-binding protein family, was followed by analysis of steady-state RNA levels during viral infection. Synthesis of the protein products of these genes was analyzed with specific monoclonal antibodies in conjunction with sensitive immunoblot or immunoprecipitation analysis. Although accumulation of ICP27 and ICP36 was not abundant until late times, both late genes were as transcriptionally active at early times (4 h postinfection) as at late times (48 h postinfection). Reduced amounts (less than 5% of late levels) of the protein products were detected at early times, demonstrating that a small proportion of the ICP27 and ICP36 RNA made at this time was translated. These observations establish that expression of at least two CMV gamma genes is regulated through posttranscriptional events. The very early transcriptional activation of late genes and the relative importance of posttranscriptional regulation to late-gene expression distinguishes CMV from other well-studied herpesviruses and does not appear analogous to late-gene regulation in any other DNA animal virus.

Virus-induced modification of the host cell is required for expression of the bacterial chloramphenicol acetyltransferase gene controlled by a late herpes simplex virus promoter (VP5)

The requirements for expression of genes under the control of early (alkaline exonuclease) and late (VP5) herpes simplex virus type 1 (HSV-1) gene promoters were examined in a transient expression assay, using the bacterial chloramphenicol acetyltransferase gene as an expression marker. Both promoters were induced, resulting in the production of high levels of the enzyme upon low-multiplicity infection by HSV-1. S1 nuclease analysis of hybrids between RNA isolated from infected cells containing HSV-1 promoter constructs and marker gene DNA demonstrated normal transcriptional initiation of the marker gene directed by the viral promoters. Viral DNA sequences no more than 125 bases 5' of the putative transcriptional cap site were sufficient for maximum activity of the late promoter. In contrast to expression controlled by the early gene, the late promoter was not active at a measurable level in uninfected cells until DNA sequences between 75 and 125 bases 5' of the transcriptional cap site were deleted. Cotransfection of cells with the expression marker controlled by HSV promoters and a cosmid containing HSV alpha (immediate-early) genes indicated that full expression of both early and late promoters requires the same virus-induced host cell modifications. Inhibition of viral DNA synthesis results in an increased rate of transient expression of marker genes under control of either early or late promoters in contrast to the situation in normal virus infection. These data provide evidence that the normal course of expression of late HSV genes involves negative modulation of potentially active promoters in the infected cell.

Regulation of cytomegalovirus gene expression: alpha and beta promoters are trans activated by viral functions in permissive human fibroblasts

We have fused immediate (alpha) and delayed (beta) early promoter-regulatory sequences taken from the cytomegalovirus (CMV) genome to Escherichia coli lacZ (beta-galactosidase) as an indicator gene to study regulated expression of these promoters. After transfection of human fibroblast cells with plasmid constructs carrying beta-galactosidase fusions, and subsequent infection with CMV, we have demonstrated that viral trans-acting functions up-regulate the expression of these genes in a temporally authentic manner. The alpha promoter is activated even when de novo protein synthesis is blocked and when UV-inactivated virus is used, suggesting that, as for herpes simplex virus type 1 (HSV-1), a virion structural protein is responsible for its up-regulation. We have found that HSV-1, as well as CMV, is capable of trans activating the CMV alpha promoter. The beta promoter is activated by CMV but is completely unresponsive to HSV-1 infection. The temporal synthesis of the alpha and beta promoters in the transient expression system conforms with their natural regulation during viral replication. The beta-galactosidase fusions we describe provide a most exquisitely sensitive indicator system for the study of cis- and trans-acting viral regulatory functions.

Promoter domains required for expression of plasmid-borne copies of the herpes simplex virus thymidine kinase gene in virus-infected mouse fibroblasts and microinjected frog oocytes

A transient expression assay was used to measure the relative template activities of mutated tk genes in mouse L cells induced in trans by herpes simplex virus (HSV). In this assay, expression of the wild-type HSV type 1 tk gene is induced at least 200-fold by the superinfecting virus. Genetic lesions that were assayed include 5' deletions, clustered base substitutions, single base substitutions, intrapromoter inversions, and intrapromoter recombinants with the HSV type 2 tk gene. Roughly half of the mutations that were tested were found to weaken tk expression efficiency, and the remaining mutations did not alter expression. The spatial distribution of mutations that reduce expression efficiency in trans-induced mouse fibroblasts facilitated the construction of a map of promoter domains. The most gene-proximal promoter domain is located between 16 and 32 base pairs (bp) upstream of the tk mRNA cap site and contains a TATA homology. Two more distally located promoter domains were mapped to discrete locations upstream from the TATA homology. One of these distal domains is located between 47 and 79 bp upstream from the mRNA cap site, and the other is located between 84 and 105 bp upstream from the tk gene. The boundaries of these three promoter domains, with one exception, coincided with the set of domains delineated previously in a frog oocyte microinjection assay. The concordant behavior of tk promoter mutants in microinjected frog oocytes and trans-induced mouse fibroblasts leads us to propose that recognition and activation of the HSV tk promoter is mediated by cellular transcription factors that are common to frogs and mice.

Identification of immediate early genes from herpes simplex virus that transactivate the virus thymidine kinase gene

A HeLa cell transient-expression assay system was used to determine if isolated immediate early (alpha) genes from herpes simplex virus (HSV) could transcriptionally activate (transactivate) the type 1 (HSV-1) thymidine kinase (TK) gene [an early (beta) gene]. Cells transfected with the TK gene alone transcribed very low levels of TK RNA. Cells cotransfected with plasmids bearing the sequences that encode the alpha-gene product infected cell protein 0 or 4 (ICP0 or ICP4) and the TK gene faithfully transcribed high levels of TK RNA. The plasmid containing the sequences encoding ICP0 was a more potent transactivator than the plasmid containing the sequences for ICP4.

Activation and inhibition of expression of the 72,000-Da early protein of adenovirus type 5 in mouse cells constitutively expressing an immediate early protein of herpes simplex virus type 1

It has been previously reported that immediate early proteins of pseudorabies and cytomegalo viruses can substitute for the products of the human adenovirus type 5 (Ad5) E1A gene in the activation of early Ad5 transcription. In the present report the effect of one of the herpes simplex virus type 1 (HSV-1) immediate early genes, ICP4, on Ad5 early gene expression has been examined using mouse cell lines that constitutively express ICP4. These lines as well as nonproducers were infected with wild-type (wt) Ad5 or with various Ad5 E1A mutants and the levels of expression of the Ad5 E2A 72K DNA binding protein were measured by immunoprecipitation with a monoclonal antibody specific for 72K. With dl 312, which lacks E1A, some 72K expression was seen in nonproducer lines but levels were considerably higher in the producer lines. A similar result was also obtained using dl 312-infected nonproducer cells that were superinfected with HSV-1 virions. These data suggest that HSV-1 ICP4 can substitute for E1A in the activation of expression of early Ad5 proteins. With wt Ad5, 72K was also expressed at high levels in nonproducer mouse cells, however, in the ICP4 producer cell lines, a marked inhibition of 72K expression was observed and this inhibition correlated with the amount of ICP4 present. Using the E1A mutants pm 975 and hr 1, this inhibition was found to be specific for the products of the 1.1-kb E1A mRNA. These data suggest that ICP4 and E1A proteins either directly inhibit each other, or more likely, operate independently and competitively on factors required for viral gene activation.

Cells that constitutively express the herpes simplex virus immediate-early protein ICP4 allow efficient activation of viral delayed-early genes in trans

To study the role of herpes simplex virus type 1 immediate-early proteins in the transcriptional activation of herpes simplex virus genes, we isolated stably transformed cells expressing herpes simplex virus type 1 ICP4, an immediate-early protein known from previous studies to be necessary for delayed-early and late transcription. These cells efficiently expressed six delayed-early herpes simplex virus genes introduced by viral superinfection, in the absence of de novo viral protein synthesis. In contrast, the delayed-early gene encoding alkaline exonuclease and the late gene encoding the capsid protein VP5 were expressed at much lower levels. Expression of a second late gene, that for glycoprotein C, was undetectable under the same experimental conditions. These results suggest that many, but not all, delayed-early genes are efficiently activated by ICP4; in addition, they demonstrate that although the late gene for VP5 is detectably activated by ICP4, its full expression requires additional factors.

"Transactivation" control signals in the promoter of the herpesvirus thymidine kinase gene

The herpes simplex virus thymidine kinase (TK) gene is transcriptionally activated in trans ("transactivated") by virus-encoded proteins during the infectious cycle. We show that TK plasmids introduced into a HeLa cell transient transcription assay system are also transactivated after infection with a TK- virus. Several aspects of this response are similar to regulation during the normal infectious cycle. Assay of TK promoter deletion and 5- to 10-base-pair substitution mutants in this system reveals that the transactivation response depends on the intactness of 109 base pairs of 5' gene flanking sequence. Differences between these results and analogous assays in the Xenopus oocyte system are discussed. A model for the putative binding of transactivator(s) to the promoter region is presented.

Expression of hepatitis B virus S gene by herpes simplex virus type 1 vectors carrying alpha- and beta-regulated gene chimeras

The domain of the hepatitis B virus (HBV) S gene specifying the HBV surface antigen (HBsAg) and comprising 25 base pairs of the 5'-transcribed noncoding region, the structural gene sequences, and the 3'-noncoding gene sequences including the polyadenylylation site was fused to the promoter-regulatory regions of the beta-thymidine kinase and of the alpha 4 gene of herpes simplex virus type 1 (HSV-1). The chimeric constructs were then inserted into the HSV-1 genome and specifically into the thymidine kinase gene by homologous recombination through flanking sequences. Cells infected with recombinants carrying the chimeric genes produced and excreted the HBsAg into the extracellular medium for at least 12 hr concurrently with the multiplication of the HSV-1 vector. The temporal patterns of expression and the observation that HBV S gene linked to the HSV-1 alpha promoter-regulatory region was regulated as an HSV-1 alpha gene indicate that the HBsAg gene chimeras inserted into the virus were regulated as viral genes. The HBsAg banded in isopycnic CsCl density gradients at a density of 1.17 g/cm3. Electron microscopic studies revealed that HBsAg harvested from the extracellular medium and banded in CsCl density gradients contained spherical particles 15-22 nm in diameter, characteristic of empty HBV envelopes. The results indicate that HSV-1 is a suitable vector for the expression of foreign genes placed under the control of HSV promoter-regulatory regions.

Transactivation of a late herpes simplex virus promoter

We have asked whether the promoter for the gene encoding the major capsid protein (VP5) of herpes simplex virus functions in uninfected mouse cells. Our experimental strategy was to first fuse the VP5 promoter to the herpes simplex virus thymidine kinase (TK) structural sequence and then to use the resulting hybrid gene to transform TK- cells to TK+. The recombinant gene transferred TK at an extremely low frequency by comparison with the wild-type TK gene, and the TK transcripts present within the resulting rare transformants initiated within the TK structural gene, rather than in the vicinity of the VP5 promoter. However, after infection with herpes simplex virus, large amounts of RNA driven from the VP5 promoter accumulated. We conclude that the VP5 promoter does not function in uninfected cells but is efficiently activated by virally coded factors, most likely one or more immediate-early proteins.

Nucleotide sequence of the marmoset herpesvirus thymidine kinase gene and predicted amino acid sequence of thymidine kinase polypeptide

The nucleotide sequence of a 2549-bp DNA fragment containing the entire coding region of the marmoset herpesvirus (MarHV) thymidine kinase gene (tk) and the flanking sequences was determined by the dideoxynucleotide chain termination method. The MarHV thymidine kinase polypeptide predicted from the nucleotide sequence contained 376 amino acids and had a molecular weight of 41,281. The sequencing data also reveal that the coding portion of another MarHV gene probably begins only 292 nucleotides downstream from the stop codon of the MarHV tk gene. There was relatively little nucleotide sequence homology between the MarHV tk gene and that of the herpes simplex virus (HSV) types 1 and 2 tk genes. Comparisons of the predicted amino acid sequences of the MarHV thymidine kinase polypeptide with that of the HSV-1 and HSV-2 thymidine kinase polypeptides, however, revealed clear, but interrupted, homology within several regions of the polypeptide chains. Amino acid sequence homology was particularly striking at residues 10 to 27 of the MarHV thymidine kinase polypeptide and residues 49 to 66 of the HSV-1 and HSV-2 thymidine kinase polypeptides. These same amino acid residues exhibit noticeable sequence homology to the mitochondrial beta subunit ATPase, oncogene p21 protein, adenylate kinase, and to other nucleotide-binding proteins. It has been proposed that the indicated regions of homology are elements of a nucleotide-binding pocket in ATPase, p21, and adenylate kinase, raising the possibility that amino acid residues 15 to 25 of the MarHV thymidine kinase and 54 to 64 of the HSV-1 and HSV-2 enzymes are likewise parts of nucleotide-binding sites.

Effects of deletions on expression of the herpes simplex virus thymidine kinase gene from the intact viral genome: the amino terminus of the enzyme is dispensable for catalytic activity

We have transferred two deletions affecting the 5' end of the herpes simplex virus thymidine kinase (TK) gene into the intact viral genome. One, extending from -12 to +189, had no effect on TK mRNA synthesis and only a small effect on TK activity, although the first 27 codons of the TK polypeptide were deleted. The other, extending from -85 to +85, severely impaired TK mRNA synthesis. We conclude that the amino terminus of the TK polypeptide is dispensable for catalytic activity, and that expression of TK in viral infections requires some of the same promoter elements used in uninfected cells.

A detailed analysis of an HSV-1 early promoter: sequences involved in trans-activation by viral immediate-early gene products are not early-gene specific

This report describes a detailed analysis of the functional DNA sequences within the HSV-1 glycoprotein D gene promoter. The transcriptional activity of deletion and insertion promoter mutants was studied after both trans activation, mediated by viral products, and cis activation by a linked SV40 enhancer. Two G-rich areas (upstream of a TATA signal) were identified as important regions of the promoter. These "upstream" signals were active in both orientations. A functional TATA-box region was detected. A second region, not homologous to the concensus TATA sequence, also appeared to have a role in the positioning of the RNA cap-sites, which included both purine and pyrimidine 5' ends. Deletion of the cap-site region resulted in a moderate reduction in transcription. All the promoter elements were important for both cis and trans activated transcription. No sequence specific for viral (trans) regulation was detected, implying that Early promoters are not distinguished by specific sequences. Since HSV-1 and some other animal viruses can activate transcription from unrelated promoters, this process is probably non-specific and applicable to many, particularly extra-chromosomal, genes. The possible mechanisms of this activation are discussed.

Control of Herpes simplex virus thymidine kinase gene expression in Saccharomyces cerevisiae by a yeast promoter sequence

This study presents the first evidence that the 5' promoter region of the Saccharomyces cerevisiae glyceraldehyde-3-phosphate dehydrogenase gene (G-3-PD) promoter will permit expression of an adjacent foreign gene. The S. cerevisiae G-3-PD promoter was linked to the herpes simplex virus--thymidine kinase (HSV-TK) gene in a shuttle plasmid capable of autonomous replication in both yeast and Escherichia coli. Since the HSV-TK gene promoter is not functional in yeast, yeast cells containing these plasmids will express the HSV-TK gene and synthesize thymidine kinase only if the yeast promoter fragment is fused to the HSV-TK gene in the proper orientation. The 5' flanking sequences necessary for the expression of heterologous eukaryotic genes in S. cerevisiae are discussed.

Transactivation of a late herpes simplex virus promoter

We have asked whether the promoter for the gene encoding the major capsid protein (VP5) of herpes simplex virus functions in uninfected mouse cells. Our experimental strategy was to first fuse the VP5 promoter to the herpes simplex virus thymidine kinase (TK) structural sequence and then to use the resulting hybrid gene to transform TK- cells to TK+. The recombinant gene transferred TK at an extremely low frequency by comparison with the wild-type TK gene, and the TK transcripts present within the resulting rare transformants initiated within the TK structural gene, rather than in the vicinity of the VP5 promoter. However, after infection with herpes simplex virus, large amounts of RNA driven from the VP5 promoter accumulated. We conclude that the VP5 promoter does not function in uninfected cells but is efficiently activated by virally coded factors, most likely one or more immediate-early proteins.