Regulation of Neighboring Gene Expression by the Herpes Simplex Virus Type 1 Thymidine Kinase Gene

Regulation of Neighboring Gene Expression by the Herpes Simplex Virus Type 1 Thymidine Kinase Gene

VIROLOGY 218, 193–203 (1996) 0179 ARTICLE NO. Regulation of Neighboring Gene Expression by the Herpes Simplex Virus Type 1 Thymidine Kinase Gene ¨ ...

200KB Sizes 0 Downloads 57 Views

VIROLOGY

218, 193–203 (1996) 0179

ARTICLE NO.

Regulation of Neighboring Gene Expression by the Herpes Simplex Virus Type 1 Thymidine Kinase Gene ¨ RG BO ¨ NI,2 and DONALD M. COEN3 W. JAMES COOK, KRISTIN K. WOBBE,1 JU Department of Biological Chemistry and Molecular Pharmacology, Harvard Medical School, Boston, Massachusetts 02115 Received November 14, 1995; accepted February 13, 1996 The herpes simplex virus type 1 thymidine kinase (tk) gene (UL23) lies upstream of the gH (UL22) gene with its 3* end overlapping the gH promoter, and it overlaps the UL24 gene’s regulatory and coding sequences at its 5* end in a head-to-head orientation. Thus, tk expression could affect gH expression by promoter occlusion and UL24 expression by transcriptional or posttranscriptional mechanisms. To investigate these possibilities, we analyzed the effects of tk promoter mutations that reduce tk expression on gH and UL24 expression. For gH, tk promoter mutations did not increase the accumulation of gH mRNA or the rate of gH transcription. Thus, tk expression does not appear to down-regulate gH expression. In contrast, we found that decreased tk expression correlated with increased accumulation of UL24 mRNA, particularly a 1.4-kb transcript, at early times postinfection during peak expression of tk, but not at late times when tk mRNA levels have fallen. Results from viral co-infection experiments indicated that down-regulation of UL24 mRNA accumulation requires tk expression in cis. Nuclear run-off experiments did not detect differences in UL24 transcription rates in the mutant viruses. Although we cannot completely exclude a transcriptional mechanism for this down-regulation, these results can be explained by an antisense RNA mechanism acting preferentially in cis. q 1996 Academic Press, Inc.

antisense RNA mechanism (reviewed in Nellen and Lichenstein, 1993; and in Takayama and Inouye, 1990). The HSV thymidine kinase (tk) gene (UL23) overlaps other genes on both its 3* and its 5* ends (Fig. 1). The 3* untranslated region of tk overlaps the TATA box of the downstream gH gene (Gompels and Minson, 1986; Sharp et al., 1983). Thus, transcription of tk could affect gH expression by promoter occlusion. As tk belongs to the early kinetic class (Harris-Hamilton and Bachenheimer, 1985; Sharp et al., 1983; Zhang and Wagner, 1987), its expression peaks at about 4 to 6 hr postinfection (p.i.) before rapidly declining. gH belongs to the strict late kinetic class since the peak of its mRNA accumulation occurs after 6 hr p.i. and is maintained at peak levels through late times, and its expression is critically dependent on viral DNA replication (Holland et al., 1984; Steffy and Weir, 1991; Weinheimer and McKnight, 1987). Indeed, because gH mRNA accumulation peaks after the level of tk transcripts declines, it was hypothesized that tk transcription decreases and/or delays gH transcription by a promoter occlusion mechanism (Sharp et al., 1983). tk overlaps the UL24 gene’s regulatory and coding sequences at its 5* end in a head-to-head orientation (Fig. 1). Because the major UL24 promoter lies within the transcribed region of tk (Cook and Coen, 1996 and references therein), tk transcription could interfere with UL24 transcription. In addition, because UL24 is transcribed in the opposite direction of tk and tk mRNA is in substantial excess over UL24 mRNA at early times during infection (Holland et al., 1984), tk transcripts could regulate UL24

INTRODUCTION Expression of the 70 to 80 protein coding genes of herpes simplex virus type 1 (HSV-1) is regulated primarily at the level of transcription (Godowski and Knipe, 1986; Wagner, 1994; Weinheimer and McKnight, 1987) and depends on the combined activity of cellular transcription factors and viral regulatory proteins working in trans. However, control mechanisms other than trans induction of transcription might also function to regulate HSV gene expression (Smith et al., 1992; Weinheimer and McKnight, 1987; Wobbe et al., 1993). For example, because the transcribed regions of many HSV genes overlap the promoters and transcribed regions of neighboring genes (McGeoch et al., 1988), it can be envisioned that one gene’s mRNA expression could directly interfere with another gene’s expression. If a gene’s transcribed region overlaps a downstream gene’s promoter, promoter occlusion could result (Adhya and Gottesman, 1982; Vales and Darnell, 1989). And if two overlapping genes are transcribed in opposite directions, transcriptional interference at the level of elongation might result, or the complementary mRNAs might control one another by an

1 Present address: Department of Chemistry, Worchester Polytechnic Institute, Worchester, Massachusetts 01609. 2 Present address: Swiss National Center for Retroviruses, University of Zurich, Zurich, Switzerland. 3 To whom correspondence and reprint requests should be addressed. E-mail: [email protected]

0042-6822/96 $18.00

193

AID

VY 7841

/

6a14$$$$81

03-18-96 10:50:57

Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.

vira

AP: Virology

194

COOK ET AL.

FIG. 1. A map showing the region of the HSV-1 genome which contains the tk, UL24, and gH genes, and reported transcripts derived from this region. The top line is a schematic representation of the HSV-1 genome in the prototype arrangement, with the location of tk at coordinate 0.3 indicated. Black rectangles represent major repeat sequences. The long line below the genome represents the DNA fragment containing the tk, UL24, and gH genes. Transcripts are shown above (UL24) and below (tk and gH) the DNA fragment, with the thickness of transcript lines representing approximate relative abundance and arrowheads denoting 3* ends. For UL24 transcripts, the upstream arrowheads represent termination at the UL24 poly(A) site with sizes of these transcripts shown. Run-on through this poly(A) site to the downstream site results in the longer transcripts indicated by sizes to the right (Cook and Coen, 1996). The tk transcript overlaps the 5* end of the 1.4- and 5.6-kb UL24 mRNAs by 504 nt (Cook and Coen, 1996; McGeoch et al., 1988) The tk, gH, and tk readthrough transcripts have been described (Holland et al., 1984; Sharp et al., 1983). The 3* end of the tk transcript was mapped to position /24 relative to the gH transcriptional start site (Gompels and Minson, 1986; Sharp et al., 1983). The tk and UL24 open reading frames (ORFs) are shown at the bottom; these ORFs overlap by 22 codons (McGeoch et al., 1988).

expression through an antisense RNA mechanism. Because the most abundant UL24 transcript accumulates with leaky–late kinetics (Cook and Coen, 1996, and references therein), tk expression could also affect the kinetics of UL24 expression as hypothesized for gH. To investigate the effects of tk expression on gH and UL24 expression, we analyzed how tk promoter mutations that reduce tk expression affect gH and UL24 transcription and mRNA accumulation. We found that reducing tk expression more than 10-fold did not significantly increase either gH mRNA accumulation or gH transcription rates, and thus no evidence of promoter occlusion was observed. In contrast, reduced tk expression resulted in corresponding increases in UL24 mRNA accumulation. UL24 transcription rates were not affected by reduced tk transcription, suggesting that down-regulation of UL24 mRNA accumulation involves a posttranscriptional mechanism. We discuss how these results suggest the hypothesis of an antisense RNA control mechanism.

resistance mutation and a temperature-sensitive tk mutation, and viruses that contain the same mutations as PKG7 plus linker scanning (LS) mutations in the tk promoter region (McKnight and Kingsbury, 1982) including LS-16/-6 (Coen et al., 1986), LS-29/-18 (Coen et al., 1986), LS/5//15 (Coen et al., 1986), LS-111/-101//-56/-46 (Bo¨ni and Coen, 1989), and LS/16//36/10 which contains a deletion of tk bp /16 to /36 spanned by a 10-bp BamHI linker (McKnight and Kingsbury, 1982; Cook et al., 1995b). The D1 virus was described previously (Halpern and Smiley, 1984). These viruses were propagated in Vero cells as described (Coen et al., 1985). The mutants n12 tk TATA (Cook et al., 1995a), n12 SV40 TATA (Cook et al., 1995a), and n12 LS-29/-18 (Imbalzano et al., 1991) were derived from the ICP4-deficient n12 virus (DeLuca and Schaffer, 1987) and were propagated as described previously in E5 cells which are a Vero cell line that expresses complementing levels of ICP4 upon HSV-1 infection (DeLuca et al., 1985). Primer extension analysis

MATERIALS AND METHODS

Viruses used were HSV type 1 KOS and the following mutants derived from it: PKG7 (Irmiere et al., 1989) which is a control virus containing the phosphonoacetic acid

Vero cells (107 cells) were infected at a multiplicity of 5 PFU per cell, and isolation of cytoplasmic RNA and primer extensions were performed as described previously (Wobbe et al., 1993). Primers specific for tk mRNA (Wobbe et al., 1993), ICP8 mRNA (Wobbe et al., 1993),

AID

vira

Viruses and cells

VY 7841

/

6a14$$$$81

03-18-96 10:50:57

AP: Virology

DOWN-REGULATION OF UL24 BY HSV-1 tk

gH mRNA (Jacobson et al., 1993), UL24 mRNA (Jacobson et al., 1993), and gB mRNA (Pederson et al., 1992) were described previously. The primer specific for gC mRNA was prepared by digesting a plasmid containing gC transcribed sequences with NheI, dephosphorylating with alkaline phosphatase, radiolabeling with [g-32P]ATP (New England Nuclear) and T4 polynucleotide kinase (New England Biolabs), and digesting with Bsp1286 to liberate a 26-bp fragment encoding /86 to /112 of the gC noncoding strand (Frink et al., 1983) that was gel purified (Wobbe et al., 1993). Relevant bands were quantified in dried gels using a PhosphorImager (Molecular Dynamics) or a b-Scope (Betagen).

195

on Northern blots were linear with the amount of RNA added (not shown). RESULTS Effects of tk expression on gH expression

Vero or E5 cells were infected at a multiplicity of 10 PFU per cell, and isolation of total RNA and Northern analysis were performed as described (Cook et al., 1995a). Preparation of radiolabeled, double-stranded DNA probes specific for tk , UL24, and gB mRNA were as described in Cook and Coen (1996), and preparation of the ICP8 probe was described previously (Cook et al., 1995a). A probe that overlaps tk and UL24 to approximately similar extents was prepared by radiolabeling a 0.63-kb SacI–EcoRI fragment isolated from pKG3.6. The ratio of tk mRNA to UL24 mRNA in virus-infected cells was determined by first probing with the UL24-specific probe and quantifying the 1.4-kb UL24 transcript relative to the 5.6-kb UL24 species on Northern blots and then stripping blots and reprobing with the 0.63-kb SacI– EcoRI probe. Levels of the major 1.4-kb tk mRNA and the 5.6-kb UL24 mRNAs were quantified and adjusted for levels of the 1.4-kb UL24 transcript. Transcript signals were measured on Northern blots using a PhosphorImager (Molecular Dynamics), or on autoradiographs of Northern blots using a Microtek scanner. Under the conditions used, the intensities of UL24 hybridization signals

The 3* untranslated region of tk overlaps the promoter region of gH (Gompels and Minson, 1986; Sharp et al., 1983; Steffy and Weir, 1991), and thus transcription of tk could down-regulate gH expression by promoter occlusion. To investigate whether reducing levels of tk mRNA affects the accumulation of gH mRNA, we measured tk and gH mRNA levels in cells infected with herpes simplex type 1 viruses that contain mutations in the tk promoter. Control viruses included LS-16/-6, which expresses identical amounts of tk mRNA as strain PKG7 (Coen et al., 1986), which contains the wild-type tk promoter, and LS/16//36/10, which contains an LS mutation that had little effect on tk expression in transient transfection assays (Eisenberg et al., 1985) or microinjected oocytes (McKnight and Kingsbury, 1982). Viruses that exhibit reduced tk expression included LS-111/-101//-56/-46, which contains LS disruptions of both Sp1 sites in tk (Bo¨ni and Coen, 1989); LS-29/-18, which contains an LS disruption of the tk TATA box (Coen et al., 1986); and LS/5//15, which contains an LS disruption just downstream of the tk mRNA start site (Coen et al., 1986). Vero cells were infected with the recombinant viruses, and cytoplasmic RNA isolated at either 5 or 10 hr p.i. was analyzed for tk mRNA relative to the early ICP8 gene mRNA and gH mRNA relative to the strict–late gC gene mRNA by primer extension analysis (Wobbe et al., 1993) (Fig. 2). Results indicated that while LS/16//36/10, LS/5//15, LS-111/-101//-56/-46, and LS-29/-18 expressed tk mRNA levels that were 80, 60, 20, and õ10%, respectively, of that expressed by LS-16/-6 at both 5 and 10 hr p.i., gH mRNA levels in these viruses showed only minor variations relative to LS-16/-6 at both time points (Fig. 2 and Table 1). Importantly, gH mRNA levels at 5 hr p.i., when promoter occlusion would be most likely to occur, were essentially unchanged in all viruses tested. These results indicate that reducing tk mRNA has no meaningful effect on gH mRNA levels. To determine if reducing tk transcription rate affects the rate of gH transcription, we performed nuclear runoff transcription assays on the tk promoter mutants. Vero cells were infected with the recombinant viruses, and infected-cell nuclei were isolated at 4.5 hr p.i. and subjected to run-off transcription assays. Although gH is classified as a strict–late gene (Holland et al., 1984; Steffy and Weir, 1991; Weinheimer and McKnight, 1987), substantial amounts of properly initiated gH mRNA are expressed at early times (i.e., 4 to 6 hr p.i.) during infection (Fig. 2) and thus we could easily detect gH transcription at 4.5 hr p.i. tk transcription was normalized to that of the early pol gene, and gH transcription was normal-

AID

vira

Nuclear run-off transcription assays Nuclear run-off assays were performed as described (Bo¨ni and Coen, 1989; Coen et al., 1986) using nuclei prepared at 4.5 hr postinfection from Vero cells infected at a multiplicity of 2.5 PFU per cell. Single-stranded DNA probes were prepared from bacteriophage M13 clones that contained transcribed regions of tk, pol, gH, gC, ICP5, and the negative control chicken tk as described (Coen et al., 1986; Weinheimer and McKnight, 1987). The UL24 probe was prepared by cloning a 1.7-kb SmaI– SmaI fragment derived from pKG3.6 (Coen et al., 1986) into SmaI-digested M13mp11. Transcription signals were measured by scanning autoradiographs on a LKB laser scanner. Transcription rates were determined by linear regression analysis of signals hybridizing to different probe dilutions (315, 100, 31.5, and 10 ng) (Bo¨ni and Coen, 1989; Coen et al., 1986). Northern blot analysis

VY 7841

/

6a14$$$$81

03-18-96 10:50:57

AP: Virology

196

COOK ET AL.

FIG. 2. Accumulation of gH mRNA is not affected by tk expression. Vero cells were mock-infected or infected with the indicated tk promoter mutant viruses, and cytoplasmic RNA was harvested at either 5 or 10 hr postinfection (p.i.) and analyzed by primer extension as described (Wobbe et al., 1993). (Top) Primer extension products obtained with the tk primer or the control ICP8 primer. The tk product produced in the LS/16//36/10 virus is 10 bp shorter than the other tk products due to a deletion as described under Materials and Methods. (Bottom) Primer extension products obtained with the gH primer or the control gC primer. The gH primer detects a cellular transcript (M) as previously described (Jacobson et al., 1993). Results are tabulated in Table 1. This and subsequent figures were generated by using the programs Adobe Photoshop and Canvas on a Macintosh computer following scanning of autoradiograms with a Microtek scanner.

ized to that of the strict–late gC gene. In agreement with previous results (Bo¨ni and Coen, 1989; Coen et al., 1986), LS-111/-101//-56/-46 and LS-29/-18 exhibited 5- and 10fold lower rates of tk transcription, respectively, relative

FIG. 3. tk and gH transcription by the tk promoter mutants. Nuclear run-off transcription assays were performed as described (Bo¨ni and Coen, 1989; Coen et al., 1986) with nuclei isolated from Vero cells at 4.5 hr following infection with the indicated viruses. The radiolabeled transcription products were hybridized to four different dilutions of single-stranded DNA from the tk gene and the gH gene, and the control genes pol and gC. Results are tabulated in Table 1.

to LS-16/-6 (Fig. 3 and Table 1). gH transcription at this early time point was 60% lower in LS-111/-101//-56/-46 and 70% lower in LS-29/-18 than in LS-16/-6. Thus, mutations that decreased tk transcription did not increase gH transcription as would be predicted if promoter occlusion were occurring, but rather gH transcription diminished. We attribute this effect to a decrease in readthrough transcription across the gH gene that initiates from the tk promoter (Sharp et al., 1983). Based on results with LS-29/-18, it appears that at 4.5 hr p.i. approximately 70% of the transcription across gH came from the tk promoter. Taken together, results from primer extension and tran-

TABLE 1 Relative Levels of tk and gH mRNA Accumulation and Transcription in tk Promoter Mutants Primer extensiona tk mRNA

LS016/06 LS/16//36/10 LS0111/0101//056/046 LS029/018

Transcriptionb

gH mRNA

5 hr

10 hr

5 hr

10 hr

tk

gH

1.0 0.8 0.2 õ0.1

1.0 0.7 0.2 õ0.1

1.0 1.3 0.7 1.2

1.0 1.1 1.1 1.8

1.0 0.8 0.2 0.1

1.0 0.7 0.4 0.3

a Values are tabulated from PhosphorImaging and densitometry of primer extension data (for example, Fig. 2). tk values were normalized as the ratio of tk mRNA levels to ICP8 mRNA levels, and gH values were normalized as the ratio of gH mRNA levels to gC mRNA levels. Each value was then compiled relative to the value obtained from LS016/06 which is given a value of 1.0. 5 hr and 10 hr refer to hours postinfection. b Values are tabulated from densitometry of nuclear run-off data (for example, Fig. 3). tk transcription values were normalized to pol transcription values, and gH transcription was normalized to gC transcription.

AID

VY 7841

/

6a14$$$$81

03-18-96 10:50:57

vira

AP: Virology

AID

VY 7841

/

6a14$$$$81

03-18-96 10:50:57

1.0 1.5 2.3 1.0 0.9 3.0

1.0 1.0 1.2 1.5 1.0 1.3 1.0 1.7

1.0 3.2 6.1 1.0 0.1 õ0.1

1.0 0.9 2.4 6.2 1.0 0.9 0.2 0.2

5 hr 10 hr

1.0 0.2 õ0.1 ND ND ND ND ND ND NDe ND ND n12tk TATA n12SV40 TATA n12LS029/018

ND ND ND

1.0 1.0 0.2 õ0.1 1.0 0.9 1.2 1.2 1.0 1.3 2.2 2.8 1.0 0.7 0.2 õ0.1 1.0 0.8 0.2 õ0.1

10 hr 5 hr

tk mRNAc

LS016/06 LS/16//36/10 LS0111/0101//056/046 LS029/018

5 hr

10 hr

5 hr

tk mRNA UL24 mRNA Primer extension

vira

AP: Virology

a Values are tabulated from PhosphorImaging of primer extension data. UL24 mRNA represents the extension products obtained with a primer specific for mRNA originating from the most 5* UL24 start site at nt 47402 and, therefore, values represent accumulation of the 1.4- and 5.6-kb transcripts. UL24 mRNA was normalized to gB mRNA. Each value is shown relative to the value obtained for LS016/06 which is given a value of 1.0. 5 hr and 10 hr refer to hours postinfection. b Values are tabulated from PhosphorImaging and densitomitry of Northern blot hybridization data (for example, Fig. 4). tk mRNA values were normalized to ICP8 mRNA, and UL24 mRNA values were normalized to gB mRNA. 1.4 and 5.6 refer to the 1.4-kb and 5.6-kb UL24 transcripts, respectively. c Values taken from Table 1. d Values are tabulated from densitomitry of nuclear run-off data (for example, Fig. 7). UL24 transcription values were normalized to UL19 transcription values. e ND, not done.

ND ND ND

1.0 0.6 0.9 0.9 1.0 0.8 0.2 0.1

ND ND ND 1.0 1.1 1.0

10 hr 10 hr

5 hr

UL24 (5.6) UL24 (1.4)

Northernb

Relative Levels of tk and UL24 mRNA and Transcription

a

TABLE 2

We next determined whether reducing tk expression affects expression of the UL24 gene which overlaps tk in a head-to-head orientation (Fig. 1). For this we analyzed production of the most abundant UL24 mRNAs, the 1.4and 5.6-kb transcripts (Cook and Coen, 1996). The 1.4and 5.6-kb UL24 transcripts are 5* coterminal but contain different 3* ends (Fig. 1). Using a probe in Northern analysis that overlaps tk and the 1.4- and 5.6-kb UL24 mRNAs to roughly similar extents (as described under Materials and Methods), we determined that at 5 hr p.i. the major 1.4-kb tk mRNA was approximately 30-fold more abundant than the 1.4- and 5.6-kb UL24 transcripts together, which agrees with previous observations (Holland et al., 1984). Thus the large excess of tk mRNA over that of UL24 suggests that tk expression might down-regulate UL24 expression by transcriptional or posttranscriptional mechanisms. We tested this by analyzing how the tk promoter mutations affected UL24 expression. A primer extension assay (Jacobson et al., 1993) was used to measure UL24 mRNA that originated from the UL24 start site at HSV-1 nt 47402 and, therefore, detected the combined accumulation of the 1.4- and 5.6-kb UL24 messages. UL24 mRNA was normalized to that of the leaky–late gB gene (Homa et al., 1988; Pederson et al., 1992). At 5 hr p.i., UL24 expression in LS-111/-101//-56/-46 and LS-29/-18 was higher than that of LS-16/-6 (2.2- and 2.8-fold higher, respectively, Table 2). Thus, the net accumulation of UL24 transcripts originating from this mRNA start site was increased in these viruses at early times. At 10 hr p.i., no meaningful differences in relative UL24 transcript levels were observed (Table 2). These results suggest that reducing tk mRNA accumulation results in increased accumulation of the 5.6- and 1.4-kb UL24 transcripts at early times postinfection when tk expression is normally high, but has no effect at late times when tk expression is low. To substantiate these findings and to determine whether each of the UL24 transcripts is affected equally by reduced tk expression, we analyzed tk and UL24 mRNA levels in the tk promoter mutants using Northern analysis (Fig. 4). As was done with the primer extension analysis, tk mRNA was normalized to ICP8 mRNA and UL24 mRNA was normalized to gB mRNA. In agreement with previous results (Bo¨ni and Coen, 1989; Coen et al., 1986) and consistent with primer extension measurements (Table 2), LS-111/101//-56/-46 and LS-29/-18 expressed tk mRNA levels at 5 hr p.i. that were 5-fold and õ10-fold lower, respectively, than that of LS-16/-6 (Fig. 4 and Table 2). At 10 hr p.i., tk mRNA accumulation in both LS-111/-101//-56/-46 and LS29/-18 was 5-fold lower than that of LS-16/-6.

tkc

tk mRNA accumulation down-regulates UL24 mRNA accumulation

1.0 0.9 0.7 1.2

Transcriptiond

scription assays suggest that tk expression does not down-regulate gH expression; no evidence for promoter occlusion was observed.

197

UL24

DOWN-REGULATION OF UL24 BY HSV-1 tk

198

COOK ET AL.

n12 SV40 TATA (Cook et al., 1995a) and n12 LS-29/-18 (Imbalzano et al., 1991) relative to n12 tk TATA (Cook et al., 1995a) which contains the wild-type tk promoter. n12 SV40 TATA and n12 LS-29/-18 accumulated 5- and 10fold less tk mRNA, respectively, than n12 tk TATA at 5 hr p.i., and both viruses accumulated approximately 10fold less tk mRNA at 10 hr p.i. (Table 2). At 5 hr p.i., levels of the 1.4-kb UL24 transcript in n12 SV40 TATA and n12 LS-29/-18 were 3.2- and 6.1-fold higher, respectively, than in n12 tkTATA. The 5.6-kb transcript levels at 5 hr p.i. were also higher for n12 SV40 TATA (1.5-fold) and n12 LS-29/-18 (2.3-fold) relative to n12 tk TATA, but not nearly as much as for the 1.4-kb species. At 10 hr p.i., the 1.4kb transcript was elevated in n12 LS-29/-18 (3-fold) but not in n12 SV40 TATA. No meaningful variations in levels of the 5.6-kb transcript were measured at 10 hr p.i.. Thus, except for slight variations in levels of the 1.4-kb transcript at 10 hr p.i., similar patterns of UL24 mRNA accumulation were measured in the two different groups of viruses. These results indicate that reduced tk mRNA accumulation results in increased levels of UL24 mRNA, especially of the 1.4-kb species as shown by Northern analysis. This effect was most prominent at 5 hr p.i., which corresponds to the peak of tk expression.

FIG. 4. Decreased tk expression primarily affects levels of the 1.4kb UL24 transcript. Northern analysis was performed on total RNA isolated at either 5 or 10 hr following infection of Vero cells with the indicated tk promoter mutants at an m.o.i. of 10 PFU per cell as described under Materials and Methods. The Northern blot was subjected to multiple cycles of hybridization, stripping, and rehybridization using the following probes: a radiolabeled 0.66-kb SacI–SmaI fragment that spans tk coding sequences (Coen et al., 1986), a 1.4-kb PstI fragment that spans ICP8 coding sequences (Gao et al., 1988), a 0.45-kb EcoRI– HindIII fragment from pLS/ts/5//15 (Coen et al., 1986) that spans UL24 coding sequences, and a 0.9-kb PstI fragment that spans gB coding sequences (Rafield and Knipe, 1984). Results are tabulated in Table 2.

Kinetics of accumulation of UL24 mRNA in LS-29/-18 In Cook and Coen (1996), we showed that maximal accumulation of the 1.4-kb UL24 transcript was at early times during infection, while the 5.6-kb species accumulated at peak levels throughout early and late times. We considered the possibility that the kinetics of UL24 mRNA accumulation are affected by tk expression and, therefore, we investigated whether decreased tk expression in LS-29/-18 results in an earlier onset of UL24 mRNA expression. This was accomplished by using Northern analysis to measure UL24 mRNA expression in Vero cells infected with LS-29/-18 at different times postinfection (Fig. 5A). Values obtained for LS-29/-18 were plotted with values obtained for LS-16/-6 that were taken from Cook and Coen (1996). Results indicated that the time courses of expression of the UL24 transcripts in LS-29/-18 were essentially the same as those in the control LS-16/-6 virus (Fig. 5B). While the 1.4-kb transcript in LS-29/-18 accumulated to a sixfold higher level at the peak time (6 hr p.i.) relative to LS-16/-6, this mRNA displayed early kinetics of expression in both viruses. The 5.6-kb message, when normalized to gB mRNA, was only slightly elevated in LS-29/-18 relative to LS-16/-6 at early times (6 and 8 hr p.i.). Thus, decreased tk expression in LS-29/18 did not significantly alter the kinetics of UL24 mRNA accumulation.

At 5 hr p.i., LS-111/-101//-56/-46 and LS-29/-18 expressed substantially higher levels of the 1.4-kb UL24 mRNA relative to LS-16/-6 (2.4- and 6.2-fold higher, respectively; Fig. 4 and Table 2). Accumulation of the 5.6-kb UL24 transcript at 5 hr p.i. was also elevated in LS-111/-101//-56/-46 and LS-29/-18, but only slightly (1.2and 1.5-fold higher, respectively). Thus the down-regulation was greater for levels of the 1.4-kb than for the 5.6kb UL24 transcript. At 10 hr p.i., we detected only a small increase in the 1.4-kb transcript in LS-29/-18 (1.7-fold), while LS-111/-101//-56/-46 showed no difference. No meaningful differences in the accumulation of the 5.6-kb transcript were measured at 10 hr p.i. To determine if we could reproduce these effects in other viruses, we analyzed tk and UL24 mRNA levels in

Down-regulation of UL24 mRNA accumulation requires tk expression in cis To determine if we could affect the increased accumulation of the UL24 1.4-kb transcript in LS-29/-18 by introducing

AID

vira

VY 7841

/

6a14$$$$81

03-18-96 10:50:57

AP: Virology

DOWN-REGULATION OF UL24 BY HSV-1 tk

199

should be the average of that found in each single infection (Ç2-fold lower than in LS-29/-18-infected cells). Results showed that in cells doubly infected with LS29/-18 and LS-16/-6 (Fig. 6, lane d), levels of the 1.4-kb transcript were 3-fold higher than those in infection with LS-16/-6 alone (lane e), and 2-fold lower than those in infection with LS-29/-18 alone (lane c). Therefore, levels during co-infection were the average of levels measured during infection with each virus alone, consistent with a requirement for tk expression in cis. When we increased the amount of LS-16/-6 virus up to 3-fold while keeping the amount of LS-29/-18 the same, levels of the 1.4-kb transcript remained the average of levels measured with each virus alone (not shown). Thus increasing tk mRNA up to a level that was Ç45-fold over that of UL24 mRNA had no trans-inhibitory effect on accumulation of the 1.4kb UL24 transcript.

FIG. 5. The temporal expression of UL24 transcripts in LS-29/-18infected cells. (A) Northern analysis was conducted as described in Fig. 4, except that total RNA was isolated from Vero cells at different times following infection with LS-29/-18. The Northern blot was first hybridized with the UL24 probe (top) and then stripped and reprobed for gB mRNA (bottom). (B) Signals from the 1.4- and 5.6-kb UL24 transcripts in the Northern blot in (A) were quantified using a PhosphorImager and, following normalizing to gB mRNA, plotted as a function of time postinfection with values obtained for LS-16/-6 taken from Cook and Coen (1996). Values are relative to that of the 5.6-kb UL24 transcript in LS-16/-6 at 6 hr p.i. which is given a value of 1.0.

tk mRNA in trans, we conducted co-infection experiments in which Vero cells were doubly infected with LS-29/-18 and either of two viruses that express high levels of tk mRNA, LS-16/-6 and D1 (Halpern and Smiley, 1984). For co-infection with LS-29/-18 and LS-16/-6, the 1.4kb UL24 transcripts synthesized from each genome are of the same size. Therefore, if expression of tk downregulates UL24 mRNA accumulation in trans, we expected that the amount of the 1.4-kb UL24 transcript in cells doubly infected with LS-29/-18 and LS-16/-6 would be similar to that found in cells singly infected with LS16/-6, i.e., 6-fold lower than that in cells singly infected with LS-29/-18. If this down-regulation requires tk expression in cis, then the amount of the 1.4-kb UL24 transcript

FIG. 6. Down-regulation of UL24 mRNA accumulation requires tk expression in cis. Northern analysis was conducted as described in Fig. 4 using total RNA that was isolated from Vero cells at 5 hr following infection with 10 PFU per cell of D1 (lane a), 5 PFU of D1 plus 5 PFU of LS-29/-18 (lane b), 10 PFU of LS-29/-18 (lane c), 5 PFU of LS-29/-18 plus 5 PFU of LS-16/-6 (lane d), and 10 PFU of LS-16/-6 (lane e). The Northern blot was first probed for UL24 mRNA (middle) and then successively stripped and reprobed for tk (top) and gB mRNA (bottom). The sizes of the full-length UL24 transcripts and the deleted 1.2-kb UL24 transcript from D1 are indicated. Although the truncated 1.2-kb UL24 mRNA in D1 is only faintly visualized in this figure, this signal was easily detected upon longer exposures of the autoradiograph and was readily quantified using a PhosphorImager. Levels of the longer UL24 mRNAs were reduced in D1 relative to the KOS-derived strains, but were the same relative to the parental strain of D1, CL101 (Wobbe and Coen, unpublished).

AID

vira

VY 7841

/

6a14$$$$81

03-18-96 10:50:57

AP: Virology

200

COOK ET AL.

We repeated the co-infection experiment using LS-29/18 and D1 which contains a deletion (tk bp 011 to /189) that shortens the tk and UL24 transcripts by 200 nt and, therefore, allowed us to differentiate tk and UL24 mRNA transcribed from the LS-29/-18 and D1 genomes. This deletion does not reduce tk mRNA relative to that found in wild-type controls (Halpern and Smiley, 1984; Fig. 6). If expression of tk down-regulates UL24 mRNA in trans, we expected that in cells doubly infected with LS-29/-18 and D1, the amount of the full-length 1.4-kb UL24 transcript synthesized from the LS-29/-18 genome would be 6-fold lower than that found in cells singly infected with LS-29/-18. In this case the amount of the shortened 1.2kb transcript synthesized from the D1 genome would be slightly reduced in doubly infected cells compared to that of cells singly infected with D1 due to 2-fold less D1 used in co-infections (Fig. 6). If this down-regulation requires tk expression in cis, then the amount of the 1.4kb UL24 transcript in doubly infected cells should be reduced less than 2-fold (due to less input virus) relative to cells singly infected with LS-29/-18, and levels of the shortened 1.2-kb UL24 transcript should also be only slightly reduced. Results shown in Fig. 6 indicated that in cells doubly infected with LS-29/-18 and D1 (lane b), normalized levels of the 1.4-kb UL24 transcript (from the LS-29/-18 genome) were 1.6-fold lower than those in infection with LS-29/-18 alone (lane c), while levels of the shortened 1.2-kb transcript were similar in co-infected cells (lane b) and cells singly infected with D1 (lane a). Thus the high levels of tk mRNA synthesized from the D1 genome had little if any trans-inhibitory effect on UL24 mRNA expressed from the LS-29/-18 genome. These results indicate that tk expression affects UL24 mRNA accumulation preferentially in cis.

FIG. 7. The rates of UL24 transcription are similar in the tk promoter mutants. Nuclear run-off transcription assays were performed exactly as described in Fig. 3, except that the radiolabeled transcription products were hybridized to DNA from the UL24 gene and the control UL19 gene. Results are tabulated in Table 2.

that the observed increases in UL24 mRNA accumulation are due to a posttranscriptional mechanism such as antisense RNA control. DISCUSSION tk expression does not down-regulate gH expression

To determine if reduced tk expression affects UL24 mRNA accumulation at the level of transcription, a nuclear run-off transcription assay was used to measure UL24 transcription rates in the tk promoter mutants. Infected-cell nuclei were isolated at 4.5 hr p.i. and transcription products were analyzed as described above for tk and gH, with UL24 transcription normalized to transcription of the leaky–late UL19 (ICP5) gene (Coen et al., 1986; Weinheimer and McKnight, 1987). We found that the rates of UL24 transcription were not elevated in LS-111/-101//-56/-46 and LS-29/-18 relative to LS-16/-6 (Fig. 7 and Table 2). Reduced tk expression in other tk promoter mutants also had no effect on UL24 transcription rates (not shown). Thus, the increased accumulation of UL24 mRNA in the tk promoter mutants is not due to increased transcription rates as would be expected if promoter occlusion or interference of transcriptional elongation were occurring. Instead, this result suggests

tk transcribed sequences overlap the gH promoter region and thus high levels of tk expression at early times during infection could interfere with gH transcription by a promoter occlusion mechanism. We report here, however, that reduced tk expression had no effect on gH mRNA accumulation and did not increase gH transcription rate (Figs. 2 and 3, Table 1). Thus, we found no evidence for promoter occlusion of gH. Instead, inefficient termination of tk transcription appears to lead to increased transcription across gH at early times. Because reduced tk transcription does not affect accumulation of properly initiated gH mRNA (Fig. 2), tk transcription does not appear to activate gH transcription as was reported for activation of adenovirus E1b gene expression by E1a transcription (Falck-Pedersen et al., 1985; Parks and Spector, 1990). Thus, tk expression neither inhibits nor activates gH transcription. Previous reports based on studies with adenovirus proposed that transcriptional activation elements such as CCAAT boxes and Sp1-binding sites are important for both promoter occlusion and activation of downstream promoters by upstream transcription (Connelly and Manley, 1989; Parks and Spector, 1990). gH lacks CCAAT and Sp1 sites; the only cis elements important for gH expression are the TATA box and sites located 3* to the TATA box that do not match these activator elements (Steffy and Weir, 1991). Therefore, the lack of both promoter occlusion and promoter activation of gH by tk transcription may be due at least in part to the absence of activator binding sites in the gH promoter.

AID

vira

The rate of UL24 transcription is not affected by reduced tk transcription

VY 7841

/

6a14$$$$81

03-18-96 10:50:57

AP: Virology

DOWN-REGULATION OF UL24 BY HSV-1 tk

tk expression down-regulates UL24 mRNA accumulation

201

viewed in Takayama and Inouye, 1990). In our co-infection experiments, we determined that tk mRNA was in 15- to 45-fold excesses of UL24 mRNA. Thus, the levels of tk mRNA in these experiments may have been insufficient to produce effective inhibition in trans. It is also possible that, at least under some conditions, antisense RNA control might operate in cis. This could be due to there being higher local concentrations of antisense transcripts in the vicinity of transcribed genes, or to compartmentalization of cellular proteins that promote RNA hybrid formation within the nucleus (reviewed in Nellen and Lichenstein, 1993). Naturally occurring antisense control has been described in a limited number of eukaryotic systems (Hildebrandt and Nellen, 1992; Kimelman and Kirschner, 1989; Krystal et al., 1990; Lee et al., 1993; Liu et al., 1994; Wightman et al., 1993). In HSV, an antisense role for the major latency-associated transcript in controlling expression of the immediate–early ICP0 gene has been proposed (Stevens et al., 1987), and transcripts expressed during latency that are antisense to the immediate–early ICP4 mRNA have been discovered (Kramer, Chen, and Coen, unpublished). Various mechanisms that operate on RNA–RNA duplexes have been suggested (reviewed in Nellen and Liechtenstein, 1993; and in Takayama and Inouye, 1990). These include (i) an enzymatic activity that unwinds RNA hybrids and in the process modifies adenosines to inosines, which is thought to alter the stability and/or translatability of RNA; (ii) a double-stranded RNase that selectively degrades RNA hybrids; and (iii) in the absence of enzymatic activity, disruption of mRNA transport out of the nucleus or interference with translation resulting from hybrid formation. One or more of these potential mechanisms might explain why the 1.4-kb UL24 transcript, which is most affected by tk expression, was not detected in previous Northern analyses that used polysomal RNA (Holland et al., 1984). We are presently investigating the details of this regulatory mechanism.

In contrast to the lack of effect on gH expression, tk expression at early times down-regulates mRNA accumulation of the UL24 gene which overlaps tk in a head-tohead orientation. Although our results do not completely exclude a transcriptional mechanism, we favor the hypothesis that this regulation involves a posttranscriptional mechanism such as antisense RNA control. We discuss below the arguments for and against each of these mechanisms. The key result suggesting a transcriptional mechanism for down-regulation of UL24 mRNA accumulation is that the regulation appears to require tk expression in cis (Fig. 6). This is an indirect argument, and we will discuss below how antisense control might operate in cis. The most compelling evidence against a transcriptional mechanism is that reduction of tk transcription rate did not result in increased UL24 transcription (Fig. 7). We also showed (Figs. 4 and 5, Table 2) that tk expression caused greater reductions in levels of the 1.4-kb UL24 transcript than those of the 5.6-kb UL24 transcript. If a transcriptional mechanism such as interference with elongation or promoter occlusion was operating, we would expect that levels of these UL24 transcripts would be similarly down-regulated since they originate from the same promoter. For these reasons, we do not favor a transcriptional mechanism. In considering a posttranscriptional mechanism, antisense regulation is the most obvious candidate. The extensive complementarity between tk and the 1.4- and 5.6kb UL24 mRNAs (504 nt) and the 30-fold excess of tk mRNA over that of UL24 mRNA at early times during infection support the possibility of such a mechanism. While we are unable to explain completely why tk expression affects levels of the 1.4-kb UL24 transcript more than that of the 5.6-kb species, we note that the overlap relative to transcript length between tk mRNA and the 1.4-kb UL24 transcript is more than that for the 5.6-kb species (36 and 9% overlap, respectively). If an antisense mechanism were operating, we could envision that hybridization of tk and UL24 mRNAs and/or inhibition caused by this hybridization might be enhanced by there being more double-stranded RNA relative to its length for the 1.4-kb UL24 transcript than for the 5.6-kb species. Our finding (Fig. 6) that the down-regulation of UL24 mRNA accumulation requires tk expression in cis might seem to argue against antisense control because in other examples of antisense regulation, introduction of antisense RNA in trans inhibited target gene expression (Izant and Weintraub, 1985; Liu et al., 1994; reviewed in Takayama and Inouye, 1990). However, in most cases where antisense RNA inhibited gene expression in trans, a 50- to 1000-fold excess of antisense RNA was needed for effective inhibition (Izant and Weintraub, 1985; re-

The biological significance of down-regulation of UL24 expression by tk expression is unclear. This down-regulation does not appear to be important for controlling the kinetics of UL24 mRNA accumulation since reduced tk expression does not significantly alter UL24 kinetics (Fig. 5). Because the effect is greatest on the monocistronic 1.4-kb UL24 transcript, this down-regulation might provide a means to specifically modulate UL24 expression at early times during infection without greatly affecting expression of the polycistronic 5.6-kb UL24 transcript which also contains UL25, UL26, and UL26.5 sequences. Additional experiments are needed to test this hypothesis, including determinations of protein expression from the monocistronic and polycistronic messages.

AID

vira

VY 7841

/

6a14$$$$81

03-18-96 10:50:57

Potential biological relevance

AP: Virology

202

COOK ET AL.

Adhya, S., and Gottesman, M. (1982). Promoter occlusion: transcription through a promoter may inhibit its activity. Cell 29, 939–944. Bo¨ni, J., and Coen, D. M. (1989). Examination of the roles of transcription factor Sp1-binding sites and an octamer motif in trans induction of the herpes simplex virus thymidine kinase gene. J. Virol. 63, 4088– 4092. Coen, D. M., Fleming, H. E., Jr., Leslie, L. K., and Retondo, M. J. (1985). Sensitivity of arabinosyladenine-resistant mutants of herpes simplex virus to other antiviral drugs and mapping of drug hypersensitivity mutations to the DNA polymerase locus. J. Virol. 53, 477–488. Coen, D. M., Weinheimer, S. P., and McKnight, S. L. (1986). A genetic approach to promoter recognition during trans induction of viral gene expression. Science 234, 53–59. Connelly, S., and Manley, J. L. (1989). A CCAAT box sequence in the adenovirus major late promoter functions as part of an RNA polymerase II termination signal. Cell 57, 561–671. Cook, W. J., and Coen, D. M. (1996). Temporal regulation of herpes simplex virus type 1 UL24 mRNA expression via differential polyadenylation. Virology 218, 204–213. Cook, W. J., Gu, B., DeLuca, N. A., Moynihan, E. B., and Coen, D. M. (1995a). Induction of transcription by a viral regulatory protein depends on the relative strengths of functional TATA boxes. Mol. Cell. Biol. 15, 4998–5006. Cook, W. J., Lin, S. M., DeLuca, N. A., and Coen, D. M. (1995b). Initiator elements and regulated expression of the herpes simplex virus thymidine kinase gene. J. Virol. 69, 7291–7294. DeLuca, N. A., McCarthy, A., and Schaffer, P. A. (1985). Isolation and characterization of deletion mutants of herpes simplex virus type 1 in the gene encoding immediate–early regulatory protein ICP4. J. Virol. 56, 558–570. DeLuca, N. A., and Schaffer, P. A. (1987). Activities of herpes simplex virus type 1 (HSV-1) ICP4 genes specifying nonsense peptides. Nucleic Acids Res. 15, 4491–4511. Eisenberg, S. P., Coen, D. M., and McKnight, S. L. (1985). 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. Mol. Cell. Biol. 5, 1940– 1947. Falck-Pedersen, E., Logan, J., Shenk, T., and Darnell, J. E., Jr. (1985). Transcription termination within the E1a gene of adenovirus induced by insertion of the mouse b-major globin terminator element. Cell 40, 897–905. Frink, R. J., Eisenberg, R., Cohen, G., and Wagner, E. K. (1983). Detailed analysis of the portion of the herpes simplex virus type 1 genome encoding glycoprotein C. J. Virol. 45, 634–647. Gao, M., Bouchey, J., Curtin, K., and Knipe, D. M. (1988). Genetic identification of a portion of the herpes simplex virus ICP8 protein required for DNA binding. Virology 163, 319–329. Godowski, P., and Knipe, D. (1986). Transcriptional control of herpesvirus gene expression: Gene functions required for positive and negative regulation. Proc. Natl. Acad. Sci. USA 83, 256–260. Gompels, U., and Minson, A. (1986). The properties and sequence of glycoprotein H of herpes simplex virus type 1. Virology 153, 230– 247. Halpern, M. E., and Smiley, J. R. (1984). Effects of deletions on expression of the herpes simplex virus thymidine kinase gene from the

intact virus genome: the amino terminus of the enzyme is dispensable for catalytic activity. J. Virol. 50, 733–738. Harris-Hamilton, E., and Bachenheimer, S. L. (1985). Accumulation of herpes simplex virus type 1 RNAs of different kinetic classes in the cytoplasm of infected cells. J. Virol. 53, 144–151. Hildebrandt, M., and Nellen, W. (1992). Differential antisense transcription from the Dictyostelium EB4 gene locus: Implications on antisense-mediated regulation of mRNA stability. Cell 69, 197–204. Holland, L. E., Sandri-Goldin, R. M., Goldin, A. L., Glorioso, J. C., and Levine, M. (1984). Transcriptional and genetic analyses of the herpes simplex type 1 genome: Coordinates 0.29 to 0.45. J. Virol. 49, 947– 959. Homa, F. L., Glorioso, J. C., and Levine, M. (1988). A specific 15-bp TATA box promoter element is required for expression of a herpes simplex virus type 1 late gene. Genes Dev. 2, 40–53. Imbalzano, A. N., Coen, D. M., and DeLuca, N. A. (1991). Herpes simplex viral transactivator ICP4 operationally substitutes for the cellular transcription factor Sp1 for efficient expression of the viral thymidine kinase gene. J. Virol. 65, 564–574. Irmiere, A. F., Manos, M. M., Jacobson, J. G., Gibbs, J. S., and Coen, D. M. (1989). Effect of an amber mutation in the herpes simplex virus thymidine kinase gene on polypeptide synthesis and stability. Virology 168, 210–220. Izant, J. G., and Weintraub, H. (1985). Constitutive and conditional suppression of exogenous and endogenous genes by anti-sense RNA. Science 229, 345–352. Jacobson, J. G., Ruffner, K. L., Kosz-Vnenchak, M., Hwang, C. B. C., Wobbe, K. K., Knipe, D. M., and Coen, D. M. (1993). Herpes simplex virus thymidine kinase and specific stages of latency in murine trigeminal ganglia. J. Virol. 67, 6903–6908. Kimelman, D., and Kirschner, M. W. (1989). An antisense mRNA directs the covalent modification of the transcript encoding fibroblast growth factor in Xenopus oocytes. Cell 59, 687–696. Krystal, G. W., Armstrong, B. C., and Battey, J. F. (1990). N-myc mRNA forms an RNA–RNA duplex with endogenous antisense transcripts. Mol. Cell. Biol. 10, 4180–4191. Lee, R. C., Feinbaum, R. L., and Ambrose, V. (1993). The C. elegans heterochronic gene lin-4 encodes small RNAs with antisense complementarity to lin-14. Cell 75, 843–854. Liu, Z., Batt, D. B., and Carmichael, G. G. (1994). Targeted nuclear antisense RNA mimics natural antisense-induced degradation of polyoma virus early RNA. Proc. Natl. Acad. Sci. USA 91, 4258–4262. McGeoch, D. J., Dalrymple, M. A., Davison, A. J., Dolan, A., Frame, M. C., McNab, D., Perry, L. J., Scott, J. E., and Taylor, P. (1988). The complete DNA sequence of the long unique region in the genome of herpes simplex virus type 1. J. Gen. Virol. 69, 1531–1574. McKnight, S. L., and Kingsbury, R. (1982). Transcriptional control signals of a eukaryotic protein coding gene. Science 217, 316–324. Nellen, W., and Lichtenstein, C. (1993). What makes an mRNA antisense-itive? Trends Biochem. Sci. 18, 419–423. Parks, C. L., and Spector, D. J. (1990). cis-dominant defect in activation of adenovirus type 5 E1b early RNA synthesis. J. Virol. 64, 2780– 2787. Pederson, N. E., Person, S., and Homa, F. L. (1992). Analysis of the gB promoter of herpes simplex virus type 1: High level expression requires both an 89-base-pair promoter fragment and a nontranslated leader sequence. J. Virol. 66, 6226–6232. Rafield, L. F., and Knipe, D. M. (1984). Characterization of the major mRNA transcribed from genes for glycoprotein B and the DNA-binding protein ICP8 of herpes simplex type 1. J. Virol. 49, 960–969. Sharp, J. A., Wagner, M. J., and Summers, W. C. (1983). Transcription of herpes simplex virus genes in vivo: Overlap of a late promoter with the 3* end of the early thymidine kinase gene. J. Virol. 45, 10–17. Smith, I. L., Hardwicke, M. A., and Sandri-Goldin, R. M. (1992). Evidence that the herpes simplex virus immediate early protein ICP27 acts post-transcriptionally during infection to regulate gene expression. Virology 186, 74–86.

AID

vira

ACKNOWLEDGMENTS We thank David Knipe for plasmids containing ICP8 and gB, Neal DeLuca for E5 cells and the n12 LS-29/-18 virus, James Smiley for the D1 and CL101 viruses, Fred Homa for communicating the sequence of the gB primer prior to publication, and Martin Dorf for use of the bScope. This work was supported by NIH Postdoctoral Fellowship AI08940 to W.J.C. and NIH Grant RO1 AI26126 to D.M.C.

REFERENCES

VY 7841

/

6a14$$$$81

03-18-96 10:50:57

AP: Virology

DOWN-REGULATION OF UL24 BY HSV-1 tk

203

Steffy, K. R., and Weir, J. P. (1991). Mutational analysis of two herpes simplex virus type 1 late promoters. J. Virol. 65, 6454–6460. Stevens, J. G., Wagner, E. K., Devi-Ro, G. B., Cook, M. L., and Feldman, L. T. (1987). RNA complementary to a herpesvirus alpha gene mRNA is prominent in latently infected neurons. Science 235, 1056–1059. Takayama, K. M., and Inouye, M. (1990). Antisense RNA. Crit. Rev. Biochem. 25, 155–184. Vales, L. D., and Darnell, J. E., Jr. (1989). Promoter occlusion prevents transcription of adenovirus polypeptide IX mRNA until after DNA replication. Genes Dev. 3, 49–59. Wagner, E. K. (1994). Herpes simplex virus—Molecular biology. In ‘‘Encyclopedia of Virology’’ (R. G. Webster and A. Granoff, Eds.), pp. 593–603. Academic Press, London.

Weinheimer, S. P., and McKnight, S. L. (1987). Transcriptional and posttranscriptional controls establish the cascade of herpes simplex virus protein synthesis. J. Mol. Biol. 195, 819–833. Wightman, B., Ha, I., and Ruvkun, G. (1993). Posttransciptional regulation of the heterochronic gene lin-14 by lin-4 mediates temporal pattern formation in C. elegans. Cell 75, 855–862. Wobbe, K. K., Digard, P., Staknis, D., and Coen, D. M. (1993). Unusual regulation of expression of the herpes simplex virus DNA polymerase gene. J. Virol. 67, 5419–5425. Zhang, Y.-F., and Wagner, E. K. (1987). The kinetics of expression of individual herpes simplex virus type 1 transcripts. Virus Genes 1, 49–60.

AID

vira

VY 7841

/

6a14$$$$81

03-18-96 10:50:57

AP: Virology