promoters late in infection (rpo3O E2 versus El and 7.5K versus TK) ... (data not shown). In order to exclude nonspecific effects of. A. A. El- "P j}. B. E2-. 1 2 4 6 8 ...
JOURNAL OF VIROLOGY, Sept. 1993, p. 5394-5401
Vol. 67, No. 9
0022-538X/93/095394-08$02.00/0 Copyright © 1993, American Society for Microbiology
Reactivation of Transcription from a Vaccinia Virus Early Promoter Late in Infection JAVIER GARCES, KRZYSZTOF MASTERNAK, BEATRICE KUNZ, AND RICCARDO WITTEK*
Institut de Biologie Animale, Universite6 de Lausanne, 1015 Lausanne, Switzerland Received 17 February 1993/Accepted 11 June 1993
We have studied the kinetics of RNA synthesis from the vaccinia virus 7,500-molecular-weight gene (7.5K gene) which is regulated by early and late promoters arranged in tandem. Unexpectedly, after a first burst of RNA synthesis early in infection, transcription was reactivated late in infection. Reactivation was not dependent on the location of the promoter in the genome or on the presence of the upstream late regulatory sequences. The mRNA synthesized from the reactivated promoter in the late phase had the same 5' and 3' ends as the molecules transcribed in the early phase. Interestingly, these molecules were efficiently translated despite the absence of the poly(A) leader characteristic of late mRNAs. Reactivation appears to be dependent on virus assembly since it is prevented by rifampin, a specific inhibitor of morphogenesis. Finally, analysis of various other early genes showed that reactivation is not unique to the 7.5K early promoter. Vaccinia virus is a complex animal DNA virus which replicates in the cytoplasm of infected host cells. The double-stranded DNA genome of the Copenhagen strain has been sequenced and consists of 191,636 bp (22). A sizable fraction of this large genome encodes enzymes and transcription factors involved in the synthesis and modification of viral mRNA (31). This renders vaccinia virus largely, and perhaps entirely, independent of host nuclear functions for gene transcription. The virus replication cycle can be summarized as follows: immediately after infection, the first uncoating step results in the release of the viral core into the cytoplasm of the cell. The RNA polymerase contained within the core then transcribes the early genes. Following expression of these genes, the DNA is released from the core and replicated. This allows expression of the intermediate genes, whose transcription follows DNA replication (46). Upon expression of the intermediate genes, the late genes, many of which encode structural proteins, are expressed. Finally, progeny virions are assembled in a complex morphogenesis process. The transition between consecutive phases of gene expression is mediated by phase-specific transcription factors which are encoded by genes of the preceding temporal class. The promoter elements of representative examples of all three temporal classes of vaccinia virus genes have been characterized. Early promoters are about 30 bp long and contain an important sequence element located between positions -28 and -13, which has been designated the critical region (16). Intermediate and late promoters are also short; they extend about 30 to 40 bp upstream of the transcription initiation site and possess distinct sequence elements characteristic of a given temporal class (5, 17, 24). Early transcription requires the presence of the vaccinia virus early transcription factor (VETF) and an RNA polymerase-associated protein, designated RAP 94 (2). VETF binds specifically to the early promoter critical region and determines the site of transcription initiation (10, 16, 50). VETF and RAP 94 are made late in infection, packaged into progeny virions, and used in a subsequent round of infection. *
Like their cellular counterparts, vaccinia virus mRNAs capped and polyadenylated. Early mRNAs usually have short 5' untranslated regions, while late mRNAs possess an unusual capped poly(A) stretch at their 5' ends. The size of this so-called poly(A) leader ranges from a few to more than 30 A residues that are not encoded by the genome. Instead, they are added by the virus RNA polymerase via a backward slippage mechanism (8, 39, 40). Termination of early transcription occurs about 20 to 50 bp downstream of the sequence 1TLTTNT (51) and requires the presence of the viral capping enzyme (41). However, in vivo studies suggest that termination directed by this motif is not 100% efficient (18). In contrast to the early transcripts, late mRNAs apparently fail to terminate at specific sites and as a consequence are long and heterogeneous in size (14, 29). It is generally believed that early promoters are silent late in infection. Interestingly, extracts prepared from cells in the late phase of infection are also able to transcribe templates containing the vaccinia virus 7,500-molecular-weight early promoter (7.5K early promoter) (34, 48, 49). This observation led us to investigate the activity of this particular promoter late in infection in vivo. The promoter of interest is part of a more complex regulatory region, as it is preceded by an upstream late promoter element. Thus, transcription of the 7.5K gene occurs from two distinct initiation sites separated by about 55 bp (13). In this paper, we present an analysis of the kinetics of transcription from the 7.5K early promoter and show that after a first burst of RNA synthesis early in infection, a second burst occurs very late in infection, well after late gene expression has reached maximal levels. Similar analyses performed on other early genes show that such a reactivation of transcription is not restricted to the 7.5K early promoter. are
MATERIALS AND METHODS Viruses and cells. The WR strain of vaccinia virus was used. A rifampin-resistant mutant virus (43) was kindly provided by E. Paoletti (Virogenetics Corp., Troy, N.Y.). CV-1 and HeLa cells were grown in Dulbecco's modified Eagle's medium supplemented with 8% fetal bovine serum. Construction of recombinant plasmids. The basic promoter
Corresponding author. 5394
VOL. 67, 1993 constructs were obtained as follows: the 7.5K tandem promoters and flanking sequences were cloned as a 324-bp
SalI-ScaI fragment (44) into pBluescriptll KS+ digested with SalI and EcoRV, and the construct was designated pBL7.5K. To delete the late promoter, pBL7.5K was digested with NdeI, partially digested withAsel, and religated, resulting in the pBL7.5KA plasmid. The NdeI site is located at position -80, and the AseI site is located at position +17 with respect to the late transcription start site. The promoter regions of pBL7.5K and pBL7.5KA were excised with AccI and EcoRI by using the EcoRI site in the polylinker of the plasmid and cloned in front of the chloramphenicol acetyltransferase (CAT) gene. The promoter-CAT fusions were inserted between the ClaI and EcoRI sites of the vaccinia virus thymidine kinase (TK) gene contained in a recombinant plasmid. The resulting plasmids, p7.5KCAT and p7.5ECAT, were used for insertion of the chimeric CAT gene into the viral genome (see below). Construction of recombinant vaccinia virus. To obtain recombinant vaccinia virus, CV-1 cells were infected with the vaccinia virus temperature-sensitive mutant ts7 (44) and then transfected with a calcium phosphate precipitate of the vaccinia virus wild-type DNA and the insertion plasmid p7.5KCAT or p7.5ECAT. TK-negative recombinant virus carrying the CAT gene under control of either the complete 7.5K early/late promoter (r7.5KCAT) or only the early promoter (r7.5ECAT) was isolated as described previously (44). Recombinant virus contained the CAT gene in the same orientation as the TK gene. Primer extension analysis. Total RNA was isolated according to the method of Sambrook et al. (36) from cells infected with a multiplicity of infection of 10 at the times indicated. Briefly, the cells were washed twice with ice-cold phosphate-buffered saline buffer and lysed in 4 M guanidinium thiocyanate-0.1 M Tris-hydrochloride (pH 7.5)-1% 1-mercaptoethanol. The RNA was subsequently purified by centrifugation through a CsCl cushion as described previously (36). For analysis of specific transcripts, 10 ,ug of total RNA was mixed with a molar excess of 5'-end-labeled primers in 10 mM Tris-hydrochloride (pH 7.5)-250 mM NaCl-1 mM EDTA, heat denatured at 95°C for 5 min, and cooled down slowly to 40°C. Primer extensions were performed with the mouse mammary leukemia virus reverse transcriptase lacking RNase H activity (Superscript; GIBCO-BRL) as described previously (19). CAT assay. Whole-cell extracts were prepared from duplicate dishes in parallel with the RNA preparations and used to determine CAT activity according to the method of Gorman et al. (23), with slight modifications. Briefly, the cellular extracts were diluted in 250 mM Tris-hydrochloride (pH 7.8)-0.1 mg of bovine serum albumin per ml, and aliquots were incubated with 4 mM acetyl coenzyme A-0.1 ,uCi of ['4C]chloramphenicol in 250 mM Tris-hydrochloride (pH 7.8) at 37°C for 1 h. Analysis of the reaction products was performed as described previously (23). Northern blotting analysis. Aliquots containing 10 ,ug of total RNA were separated on formaldehyde-containing agarose gels (36), blotted onto nylon membranes, and probed with either a CAT-specific or a TK-specific riboprobe. The probes were obtained by in vitro transcription of appropriate DNA fragments with SP6 or T7 RNA polymerase, according to the conditions described by Sambrook et al. (36). Hybridizations were performed with 6x SSC (lx SSC is 0.15 M NaCl and 0.015 M sodium citrate)-5 x Denhardt's solution0.1% sodium dodecyl sulfate (SDS) at 68°C for 5 h. After
REACTIVATION OF EARLY TRANSCRIPTION
5395
TABLE 1. Oligonucleotides used for primer extension analysis Gene
Oligonucleotide (5'-3')
TK AlL
CATGOGGCCGATTATCAACTG AGTACAAGCGAGTGCTTCTTCT
l1K 7.5K CAT
CAGTTCTAACATCGGATACTTTAACGG GGCACATGCATGCCAGGACG GGTGGTATATCCAGTGATTTTTTTCTCCAT CTCGGATATCAGTAGCGGTTACCGCCA
rpo3O
Position
+23 to +71 to +80 to +65 to +5 to +28 to
+44a +97a +106' +84a +34" +54a
a Position with respect to the first early transcription initiation site. bPosition with respect to the EcoRI site of the CAT cartridge.
hybridization, the membrane was washed for 10 min in 1 x SSC-0.1% SDS at room temperature and then in 0.2x SSC-0.1% SDS at 680C. Nuclease Si mapping. The probe for Si mapping of 3' ends was derived from plasmid p7.5ECAT. A 950-bp NcoI-PvuII fragment containing 175 bp of vector sequences at the 5' end was 3' end labeled at the NcoI site in the CAT gene coding sequences with [a-32P]dATP by using the Klenow fragment of DNA polymerase. The gel-purified probe (approximately 5,000 cpm) was hybridized to 10 ,ug of total RNA isolated at 2 h postinfection. The hybridization and nuclease Si digestion were performed as described previously (36). RESULTS
Expression pattern of genes of different temporal classes. The kinetics of RNA synthesis from the 7.5K tandem early/late promoter were investigated during a 12-h period of infection by primer extensions. Also included in the analysis were representative examples of all three temporal classes of vaccinia virus genes, namely, the TK early gene (25, 47), the AlL intermediate gene (27), and the ilK late gene (7). Oligonucleotides used in this analysis are shown in Table 1. As seen in Fig. 1, TK early mRNA was first detected at 1 h after infection and reached its highest levels at about 3 h. At later times, only trace amounts were observed. As expected, transcription of the AlL intermediate gene was delayed by about 2 h with respect to the TK gene. RNA levels peaked at 4 to 5 h and then decreased rapidly. Finally, transcription of the ilK late gene started at about 5 h, and by 6 h RNA levels had reached peak values, which stayed constant for the remaining period investigated. Primer extension of the ilK and AlL RNAs resulted in multiple bands because of the presence of the poly(A) leader, which is heterogeneous in length. Transcription from the 7.5K gene was also analyzed. Since this gene is regulated by both an early and a late promoter with corresponding transcription initiation sites separated by about 55 bp (13), two temporal classes of RNA were observed. Again, extension products of RNA from the late promoter yielded a broad smear. Primer extension of the 7.5K early RNA yielded an unexpected result. Small amounts of the RNA were first detected at 1 h after infection, as was also the case for the TK early RNA. The amount of RNA then increased and decreased with kinetics characteristic of early transcripts. Surprisingly, late in infection a second increase in the amount of 7.5K early RNA was detected, increasing until 12 h, the last time point investigated. In other experiments in which synthesis was monitored for longer periods, RNA levels continued to rise slightly until 15 h and then decreased, although relatively large amounts of transcripts were still detectable at 24 h (data not shown). Reactivation
GARCtS ET AL.
5396
A
J. VIROL. -232
n.i.0,5 1
2
3
4
5
6
8
9
SaI -TK
-r I I
AlL
orU
_ ,
B
.11K
~7.5L
w _
150
-I-
---*
.0
'1
10 TIMEI (hrs p.i.)
A
15
FIG. 1. Kinetics of gene expression in vaccinia virus-infected cells. (A) Total RNA was prepared from infected or noninfected (n.i.) CV1 cells at the time (in hours postinfection) indicated above each lane. Aliquots of 10 p.g were subjected to primer extension analysis with primers specific for an early gene (TK), an intermediate gene (AlL), a late gene (1lK), or the 7.5K gene. Extension products of early and late RNAs are designated 7.5 E and 7.5 L, respectively. Only the relevant parts of the autoradiographs are shown. (B) The autoradiographs were scanned, and the values were normalized to either the value at 12 h postinfection (7.5 E, 7.5 L, and 11K) or to the maximal value (TK), which was set to 100.
observed not only in the CV-1 cells which we have used in most experiments but also in HeLa cell monolayers and is thus not restricted to a particular cell line (data not shown). Analysis of transcription from the isolated 7.5K early promoter. Reactivation of early transcription late in infection occurs from the early promoter of the 7.5K gene, but not from that of the TK gene, which contains only an early regulatory region. We therefore considered the possibility that the close vicinity of the late promoter somehow mediates the reactivation of transcription from the 7.5K early promoter. This was investigated by analyzing RNA synthesis from the early promoter separated from the upstream late
was
+1
+32
E I SaL
S
12
7.5K promoter region
ScaI
-138
-41
L 4E
+1
SaI
CiT EcoRI
\\ Sail
L
-
r7.5KCAT
-
r7.5ECAT
E
(Ne/se
R ScoRI
CAT
FIG. 2. Relevant portions of recombinant virus. The authentic 7.5K gene promoter region is shown at the top. The r7.5KCAT recombinant contains the CAT gene under control of the dual early/late promoter. In the r7.5ECAT recombinant, a region located between the NdeI and AseI sites containing the late promoter has been deleted. E, early; L, late promoter element. The direction of transcription is indicated by the arrows.
element. We also tested whether reactivation was related to the particular position of the 7.5K gene in the genome. Toward this goal we constructed two recombinant viruses (Fig. 2) in which the CAT gene under control of the complete 7.5K early/late promoter (r7.5KCAT) or only the early promoter (r7.5ECAT) was inserted into the TK locus. Cell cultures were infected with recombinant virus, and total RNA was isolated at various times after infection and analyzed by primer extension. Cell extracts prepared from parallel cultures were used to determine CAT activity. When RNA isolated from cells infected with the r7.5KCAT recombinant virus was analyzed by primer extension, both temporal classes of RNA were detected, and these accumulated with the expected kinetics (Fig. 3A). However, although reactivation of transcription from the early promoter was always observed, it was less pronounced than in the case of the 7.5K early promoter in its original position, and the extent of reactivation varied from experiment to experiment (data not shown). CAT activity increased steadily during the entire period of infection (Fig. 3B). The results of the RNA analysis and CAT assays were quantified and are represented in Fig. 3C. To test whether reactivation of early transcription depends on the presence of the upstream 7.5K late promoter sequences, cells were infected with the r7.5ECAT recombinant virus. The results of RNA analysis and CAT assays are shown in Fig. 4. Significantly, a strong reactivation of early transcription was still observed. Thus, reactivation of early transcription from the 7.5K early promoter is independent of the presence of the late promoter element. CAT activity increased up to 4 h after infection and then reached a plateau. After 8 h, a second rise in enzyme activity was observed (Fig. 4B and C), demonstrating that the RNA was translated. Moreover, at early as well as late time points, when similar amounts of CAT mRNAs were detected, the rates of CAT accumulation were comparable, indicating that the translational efficiency of the RNA did not change throughout the infection cycle. Transcription termination. Transcription termination of vaccinia virus early RNAs occurs downstream of the sequence TITTITNT (51). In contrast, late transcripts apparently fail to terminate at specific sites and as a consequence are long and heterogeneous in size (14, 29). The molecular basis for this difference is currently not known. Reactivation of transcription from an early promoter late in infection
VOL. 67, 1993
REACTIVATION OF EARLY TRANSCRIPTION
B
A 7.5 L-
..A
_.
A
B
*
*
fk
6 8 (h p.i.1
9.5
11
5397
w.
2 2
4
6
8
10 12
(h p.i.)
7.5 E-I
4
-
C 7.5E-?0
C
:l _f
c-
-1
I
2 4 6 8 9.511
{h p.i.)
0
4
T
2
4 6
8
(h p.i.)
1012
0
5 10 TIME Ih p.i.)
FIG. 3. Levels of RNA and CAT activity expressed by recombinant virus r7.5K CAT. (A) RNA isolated at the times indicated was analyzed by primer extension. 7.5E and 7.5L indicate primer extension products from mRNA made from the 7.5K early and late promoters, respectively. (B) Analysis of CAT activity in extracts prepared in parallel. (C) RNA levels were determined by scanning the autoradiograph, and the values were normalized to the maximal values. CAT activity is expressed as the percentage of acetylation obtained in a standard reaction with 10 ng of protein extract.
provided a means of answering the question of whether recognition of the termination signal is determined by the phase of gene expression or by the class of promoter used. To address this question, Northern blot analysis of RNA isolated from cells infected with r7.5ECAT recombinant virus was performed. RNA isolated at different times was blotted and hybridized to radioactively labeled riboprobes (Fig. 5B). With the TK-specific riboprobe (Fig. SA), we expected to obtain an RNA of about 1,500 nucleotides resulting from transcription termination at the TK gene termination signal, and such a band was observed (lanes 1, 2). Unexpectedly, a second species consisting of about 1,100 nucleotides was also present. The size of this RNA suggested that termination occurred in the CAT gene coding sequences. Indeed, inspection of the sequence revealed a stretch of seven T residues at the expected position, corresponding to the vaccinia virus transcription termination signal. That this sequence actually functions as a terminator was demonstrated by S1 mapping of RNA isolated 2 h postinfection (Fig. SC). With a 3'-end-labeled probe, protected fragments with sizes of about 560 and 180 nucleotides were observed, consistent with the existence of two termination sites. When the blot was probed with a CAT genespecific riboprobe (Fig. SB, lanes 7 to 12), three RNA species consisting of 1,S00, 1,100, and 700 nucleotides were detected in early RNA. Transcription initiation at either the TK or 7.5K early promoter and termination at either of the two identified sites were expected to result in four RNA
iME
6 8 (h p.i.)
10
12
FIG. 4. Levels of RNA and CAT activity expressed by recombinant virus r7.5E CAT. (A) Primer extension products of RNA isolated at the indicated times from infected cells (7.5 E). (B) Analysis of CAT activity from extracts prepared in parallel. (C) RNA levels were quantified by densitometric scanning of the autoradiograph, and the values obtained were normalized to the maximal values. CAT activity is expressed as the percentage of acetylation obtained in a standard reaction with 100 ng of protein extract.
species. However, two of these were very similar in size (Fig. 5A) and would not be expected to be resolved by agarose gel electrophoresis. Importantly, both the signals for the 700- and 1,100-nucleotide RNA originating from the reactivated promoter increased in intensity late in infection, indicating that the early termination signals were also recognized late in infection. The smear observed in Fig. 5B in the lanes containing late RNA is due to long heterogeneous read-through transcripts initiated upstream of the TK gene (4). Reactivation of other early promoters. The experiments described above demonstrate that transcription from the 7.5K early promoter is reactivated late in infection. To test whether other early promoters also have this property, we analyzed transcription of several other genes. Primer extension analysis was performed with RNA transcribed from KlL (9, 21), HSR (35), VGF (45), and three genes coding for RNA polymerase subunits: rpol47 (11), rpol32 (3) (data not shown), and rpo3O (1) (Fig. 6A). Of these, we found that transcription from only the rpo3O gene was reactivated late in infection. This gene has a complex pattern of expression, with two early transcription initiation sites and a late RNA start site located in between. Analysis of RNA transcribed from this gene showed a marked difference in the amount of transcripts made from the two early initiation sites, consistent with the results of Si analysis reported by others (1). The amount of RNA transcribed from the distal early promoter (El) reached its highest values between 2 and 4 h after infection. No evidence of reactivation of early transcription from this promoter was seen late in infection. In contrast, transcription from the proximal early promoter (E2) showed two peaks of activity. A first peak of activity was observed
GARCtS ET AL.
5398
J. VIROL.
A
A PrK
T1
P7.5E
X12
Nrol
Prombes:
El*TK-
"P
riboprohes
CAT
*-
0
145
S I probe
\
5(06
j}
T'ranscripts: -
1500 nt
-
-1 100 nt-
-1100 nt -_ -700 iot -*-
o
I TK-transcripts 7.5E-transcripts
v
E2-
C
B 2 4 6 8 10 12 2 4 6 8 1(0 12
1 2 4 6 8 12
12 3 4
(h p.i.) -
-
959
-
345
B
28S-
i'l,ikQ-
1w:7r
-48
J LT -
lS
El
-
262
L
-26
TAAATG Metl
X~~~~~~~"0
E2
+1
F* ATG Met17.
FIG. 6. Levels of RNA synthesized from the rpo3O gene promot(A) RNA isolated at the indicated times was analyzed by primer extension. El, E2, and L, extension products of RNAs made from the distal early, proximal early, and late promoters, respectively. (B) Scheme of the rpo3O gene promoters with the distances between El and the two upstream initiation sites E2 and L indicated. ers.
FIG. 5. Termination of transcription from the reactivated
pro-
moter. (A) A schematic representation of the TK locus of recombinant virus r7.5E CAT. PTK, TK gene promoter; P7.5E, 7.5K early
promoter; T1 and T2, transcription termination signals. The positions of TK- and CAT-specific riboprobes are indicated by arrows. The Si probe labeled at the 3' end (solid circle) and the distance (in nucleotides) of the two termination signals from the labeled end are also shown. RNA species transcribed from the TK and 7.5K early promoters with their respective sizes are shown (transcripts). (B) Northern blot analysis of RNA isolated at the times (in hours postinfection) indicated above the lanes and probed with TK and CAT riboprobes. Bands corresponding to the RNA species shown in panel A are marked with arrows. The positions of the 28S and 18S rRNAs are also shown. (C) Si nuclease analysis. Lane 1, probe. The probe was hybridized to total RNA isolated from noninfected (lane 2) or infected (lane 3) cells at 2 h postinfection and digested with nuclease Si. The protected fragments are indicated by arrows. Lane 4, marker. The sizes of the fragments (in nucleotides) are indicated on the right.
between 2 and 4 h postinfection, after which RNA levels decreased. However, at 12 h, RNA levels were again as high as during the first wave of transcriptional activity. Transcripts from the late promoter (L), which is weak, were not easily detected and produced a smear just below the position of the 5' ends of the major early transcripts. The apparent differences in the activity of various early promoters late in infection (rpo3O E2 versus El and 7.5K versus TK) strongly argue against the possibility that reactivation is dependent upon the progeny virus superinfecting another infected cells. To formally rule out this possibility, we performed experiments with the drug N1-isonicotinoyl-
N2-3-methyl-4-chlorobenzoylhydrazine (IMCBH) (26). This compound prevents virus envelopment and release without significantly affecting the production of intracellular naked virus (37). Analysis of RNA from cells infected in presence of the drug (10 p,gIml) demonstrated that IMCBH did not inhibit reactivation. Rifampin prevents reactivation. Since reactivation of early transcription occurs at a time when the first progeny virions are assembled, we tested whether reactivation might be linked to morphogenesis. Rifampin at a concentration of 100 ,ug/ml, but not at 50 ,ug/ml, specifically inhibits the maturation of progeny virions but has little or no effect on viral DNA, RNA, or protein synthesis (6, 32, 33; 42). To test the influence of rifampin on reactivation, cells were infected with wild-type (rifampin-sensitive) virus and grown in medium containing either 50 or 100 ,g of the drug per ml. At various times after infection, 7.5K RNA was analyzed by primer extension. The results of the densitometric scanning of the signals corresponding to transcripts initiated at the early promoter are shown in Fig. 7A. Both in the absence and presence of 50 ,ug of rifampin per ml, reactivation was apparent. In the absence of the drug, the degree of reactivation at 12 h postinfection was higher than that usually observed. More importantly, with 100 ,ug of rifampin per ml, no reactivation occurred. Late transcription was apparently unaffected by the presence of rifampin, as the levels of transcripts initiated at the upstream late promoter were similar, irrespective of the drug concentration (data not shown). In order to exclude nonspecific effects of
REACTIVATION OF EARLY TRANSCRIPTION
VOL. 67, 1993
A Ri6virus
-o--- 5Opg/ ml pg /ml
- x-
-n+ -+100 pg / ml
200-
100
IX B
A~~~xX
---
RiR RfRVius
I
+
OpgI/ml loo pg / l _
200
nD~~ 100
/ +
+~~~ U
0
2
4
6 8 TIME (h p.i.)
10
12
14
FIG. 7. Effect of rifampin on reactivation. Rifampin-sensitive (Rif5) (A) and rifampin-resistant mutant (Rif) (B) virus was grown in presence of different concentrations of the drug. RNA harvested at the times indicated was analyzed by primer extension. The autoradiographs were scanned, and the values obtained were normalized to the value at 2 h postinfection, which was set at 100.
rifampin we also tested its effects on a drug-resistant mutant virus (43) and found that rifampin at a concentration of 100 ,ug/ml did not prevent reactivation (Fig. 7B). DISCUSSION Transcription from the early 7.5K promoter is reactivated late in infection. In analyzing RNA made from the 7.5K early promoter region, we were surprised to find that after the expected first burst of RNA synthesis early in infection, a second peak appeared late in infection (Fig. 1). Reactivation of early transcription was not observed for several other early promoters, and moreover, reactivation was not affected by an inhibitor of virus release, excluding the possibility that the second burst of transcription from the early 7.5K element was due to a second round of infection. Apparently, reactivation is not due to a position effect but is an intrinsic property of the 7.5K gene promoter region, as this phenomenon still occurs when the promoter is translocated into the TK locus. Also, reactivation was not dependent on the presence of the upstream late promoter, as deletion of this region did not abolish the reactivation of the early promoter. A subset of early promoters is reactivated. We initially
5399
suspected that reactivation might be a consequence of promoter strength and therefore selected some additional early promoters for analysis on the basis of homology to the critical region consensus sequence as established by Davison and Moss (16). However, the results of these analyses showed no such correlation. Of the six early genes analyzed, only transcription from the rpo3O gene was found to be reactivated. This gene is regulated by two early promoters and one late promoter (1). One early promoter (El) has a high degree of homology to the consensus sequence and is stronger than the other (E2), which has a lower degree of homology. Significantly, it is the weaker promoter which is reactivated (Fig. 6), demonstrating that reactivation is not a direct function of promoter strength. Given that the transcripts originating from the 7.5K early promoter initiate and terminate (see below) at the same sites in the early phase and during reactivation, we assume that the factors specific for early transcription play similar roles in reactivation. There must, therefore, be additional elements that are responsible for the differences between various early promoters with respect to reactivation. We consider several mechanisms to explain these differences. For instance, one could imagine the existence of a repressor that prevents the nonreactivated promoters from being active late in infection by binding to a particular DNA sequence. Alternatively, an activator protein could act selectively on promoters subject to reactivation. In an attempt to identify putative sequence elements responsible for the difference between the two classes of early promoters, we performed sequence comparisons between reactivated and nonreactivated promoters. On the basis of the few examples available, we were not able to find any sequence elements particular to either class of early promoters. The regulatory elements that determine reactivation, although not necessarily located within the 30-bp-minimal early promoter region, would be likely to be quite close to it and exert their effect over only a short distance, as suggested by the fact that the two early promoters of the rpo3O gene, which are separated by only 50 bp, behave quite differently with respect to reactivation. Experiments are under way to identify these regulatory elements. Termination of early transcription in the late phase. Northern blot analysis and nuclease Si mapping showed that the 7.5K early RNAs made late in infection have discrete 3' ends resulting from specific transcription termination. The ability to recognize the transcription termination signal is therefore promoter specific rather than specific for the early phase of virus gene expression. The viral capping enzyme plays an important role in transcription termination early in infection but is apparently not sufficient, since intermediate and late RNAs are not terminated specifically despite the constant presence of the capping enzyme. Thus, other factors specific for early transcription, such as VETF or RAP 94 might also participate in the termination process. Early mRNAs synthesized late in infection are translated. Vaccinia virus late mRNAs possess a capped poly(A) leader which represents an unusual 5' untranslated region. Various speculations about the possible functions of the poly(A) leader have been formulated. One of these is based on the fact that the poly(A) leader is devoid of secondary structures, which may render the process of translation initiation less dependent upon recognition of the cap (28). Viral RNAs would be selectively translated if vaccinia virus infection resulted in the inactivation of factors involved in capdependent translation. This possibility was recently addressed in experiments designed to analyze whether the
5400
GARCES ET AL.
factors p220 and eIF-4E are inactivated in the course of infection, but no changes in protein levels or in the phosphorylation state were detected (20, 38). Nevertheless, these experiments do not exclude that vaccinia virus might achieve selective translation of poly(A)-headed mRNA by some different mechanism. This idea is supported by the recent observation that the early mRNA encoding the viral DNA polymerase is not translated late in infection (30). Our results argue against the role of the poly(A) leader in selective translation of late vaccinia virus mRNAs. The mRNA synthesized from the reactivated early promoter does not have a poly(A) leader but was nonetheless translated efficiently at late times, as was demonstrated by the increase in CAT activity correlating with the level of CAT mRNA. Furthermore, there was no apparent change in translational efficiency of these mRNAs with the switch from early phase to late phase. Reactivation is inhibited by rifampin. Reactivation of the 7.5K early promoter is first observed at about 7 to 8 h postinfection. At this time, late gene expression has already reached its maximal levels and the early stages of virus morphogenesis occur (15). We therefore tested whether reactivation is linked to virus assembly. Toward this goal we used rifampin, which specifically prevents virus maturation without significantly influencing levels of DNA, RNA, or protein synthesis (6, 32, 33, 42). We found that rifampin inhibits reactivation of early transcription in wild-type virus but not in a rifampin-resistant mutant. This indicates that this inhibition by rifampin is not an unrelated effect of the drug and that reactivation is directly related to virus assem-
bly. Among the effects of rifampin on virus maturation is the inhibition of the transition of the viral DNA from a DNasesensitive state to a DNase-insensitive state, suggesting that the drug prevents the packaging of viral DNA into virions (32). It is therefore tempting to speculate that reactivation takes place on DNA templates that are already packaged into immature nucleoprotein particles together with VETF and other components of the early transcription machinery. Thus, reactivated transcription from early promoters may take place on templates different from those for transcription from late promoters occurring at the same time. This is supported by the observation that the reactivation event does not seem to interfere with the activity of the upstream late promoter in spite of the limited distance between the two initiation sites (55 and 30 bp for the 7.5K gene and rpo3O, respectively). Biological role of reactivation. Of the perhaps 100 genes expressed early in infection, only 8 were analyzed in this study, 2 of which were found to be reactivated late in infection. Therefore, at the present time we do not know how general this phenomenon is, and it is difficult to evaluate its biological significance. Since the mRNA transcribed from the 7.5K early promoter is translated late in infection,
reactivation leads to an increase in synthesis of this protein at the time of virus maturation. It is conceivable that such an increase may contribute to the fine tuning of the virus replication cycle. However, as the function of the 7.5K protein is not known, it is difficult to speculate about the importance of reactivation of this particular gene. Concerning rpo3O, reactivation of transcription from the E2 promoter leads to the synthesis of an N terminally truncated version of the RNA polymerase subunit. It is not known whether the two forms of the rpo3O protein differ functionally. Nevertheless, it is possible that some proteins, which are made early in infection are needed again very late. If this were the case,
J. VIROL.
reactivation of transcription may be considered as a novel mechanism by which this is achieved. Alternatively, reactivation could be regarded as a consequence of the presence of all of the factors and enzymes needed for early transcription, and their availability to the DNA template during morphogenesis. The question then remains as to why only a subset of early promoters is reactivated. If indeed a repressor prevented the expression of most early genes during the late phase, reactivation could simply be the consequence of some genes escaping this silencing. Perhaps it is not a coincidence that both of the identified reactivated early promoters are present downstream of late promoters for the same gene, since binding of a repressor to the early promoter in such a configuration would probably not be compatible with transcription from the upstream late promoter. Whatever the biological significance of reactivation is, we believe that by studying the mechanisms by which only selected early promoters become reactivated late in infection we will be able to learn more about gene regulation in vaccinia virus. ACKNOWLEDGMENTS We thank Enzo Paoletti for the rifampin-resistant virus mutant and Stephanie Child, Nicolas Mermod, and Walter Wahli for critical reading of the manuscript. This work was supported by the Swiss National Science Foundation. REFERENCES 1. Ahn, B.-Y., P. D. Gershon, E. V. Jones, and B. Moss. 1990. Identification of rpo3O, a vaccinia virus RNA polymerase gene with structural similarity to a eucaryotic transcription elongation factor. Mol. Cell. Biol. 10:5433-5441. 2. Ahn, B.-Y., and B. Moss. 1992. RNA polymerase-associated transcription specificity factor encoded by vaccinia virus. Proc. Natl. Acad. Sci. USA 89:3536-3540. 3. Amegadzie, B. Y., M. H. Holmes, N. B. Cole, E. V. Jones, P. L. Earl, and B. Moss. 1991. Identification, sequence and expression of the gene encoding the second largest subunit of the vaccinia virus DNA-dependent RNA polymerase. Virology 180: 88-98. 4. Bajszar, G., R. Wittek, J. P. Weir, and B. Moss. 1983. Vaccinia virus thymidine kinase and neighboring genes: mRNA and polypeptides of wild-type virus and putative nonsense mutants. J. Virol. 45:62-73. 5. Baldick, C. J., Jr., J. G. Keck, and B. Moss. 1992. Mutational analysis of the core, spacer, and initiator regions of vaccinia virus intermediate-class promoters. J. Virol. 66:4710-4719. 6. Ben-Ishai, Z., E. Heller, N. Goldblum, and Y. Becker. 1969. Rifampicin poxvirus and trachoma agent. Nature (London) 224:29-32. 7. Bertholet, C., R. Drillien, and R. Wittek. 1985. One hundred base pairs of flanking sequence of a vaccinia late gene are sufficient to temporally regulate late transcription. Proc. Natl. Acad. Sci. USA 82:2096-2100. 8. Bertholet, C., E. Van Meir, B. ten Heggeler-Bordier, and R. Wittelk 1987. Vaccinia virus produces late mRNAs by discontinuous synthesis. Cell 50:153-162. 9. Boursnell, M. E., I. J. Foulds, J. Campbell, and M. M. Binns. 1988. Non-essential genes in the vaccinia virus HindIII K fragment: a gene related to serine protease inhibitors and a gene related to the 37K vaccinia virus major envelope protein. J. Gen. Virol. 69:2995-3003. 10. Broyles, S. S., J. Li, and B. Moss. 1991. Promoter-DNA contacts made by the vaccinia virus early transcription factor. J. Biol. Chem. 266:15539-15544. 11. Broyles, S. S., and B. Moss. 1986. Homology between RNA polymerases of poxviruses, prokaryotes, and eukaryotes: nucleotide sequence and transcriptional analysis of vaccinia virus
REACTIVATION OF EARLY TRANSCRIPTION
VOL. 67, 1993
12. 13. 14.
15. 16. 17.
18.
19.
20.
21.
22. 23. 24.
25.
26.
27.
28. 29. 30.
31.
32.
genes encoding 147-kDa and 22-kDa subunits. Proc. Natl. Acad. Sci. USA 83:3141-3145. Broyles, S. S., L. Yuen, S. Shuman, and B. Moss. 1988. Purification of a factor for transcription of vaccinia virus early genes. J. Biol. Chem. 263:10754-10760. Cochran, M. A., B. M. Puckett, and B. Moss. 1985. In vitro mutagenesis of the promoter region for a vaccinia virus gene: evidence for tandem regulatory signals. J. Virol. 54:30-37. Cooper, J. A., R. Wittek, and B. Moss. 1981. Hybridization selection and cell-free translation of mRNA encoded within the inverted terminal repetition of the vaccinia virus genome. J. Virol. 37:284-294. Dales, S., and B. G. T. Pogo. 1981. Biology of poxviruses. Monogr. Virol. 18:1-109. Davison, A. J., and B. Moss. 1989. Structure of vaccinia virus early promoters. J. Mol. Biol. 210:749-769. Davison, A. J., and B. Moss. 1989. Structure of vaccinia virus late promoters. J. Mol. Biol. 210:771-784. Earl, P. L., A. W. Hugin, and B. Moss. 1990. Removal of cryptic transcription termination signals from the human immunodeficiency virus type 1 envelope gene enhances expression and immunogenicity of a recombinant vaccinia virus. J. Virol. 64:2448-2451. Garces, J., and R. Wittek. 1991. Reverse transcriptase-associated RNaseH activity mediates template switching during reverse transcription in vitro. Proc. R. Soc. Lond. Ser. B Biol. Sci. 243:235-239. Gierman, T. M., R. M. Frderickson, N. Sonenberg, and D. Pickup. 1992. The eukaryotic translation initiation factor 4E is not modified during the course of vaccinia virus replication. Virology 188:934-937. Gillard, S., D. Spehner, R. Drillien, and A. Kirn. 1986. Localization and sequence of a vaccinia virus gene required for multiplication in human cells. Proc. Natl. Acad. Sci. USA 83:5573-5577. Goebel, S. J., G. P. Johnson, M. E. Perkus, S. W. Davis, J. P. Winslow, and E. Paoletti. 1990. The complete DNA sequence of vaccinia virus. Virology 179:247-266. Gorman, C. M., L. F. Moffat, and B. H. Howard. 1982. Recombinant genomes which express chloramphenicol acetyltransferase in mammalian cells. Mol. Cell. Biol. 2:1044-1051. Hirschmann, P., J. C. Vos, and H. G. Stunnenberg. 1990. Mutational analysis of a vaccinia virus intermediate promoter in vivo and in vitro. J. Virol. 64:6063-6069. Hruby, D. E., R. A. Maki, D. B. Miller, and A. Ball. 1983. Fine structure analysis and nucleotide sequence of the vaccinia virus thymidine kinase gene. Proc. Natl. Acad. Sci. USA 80:34113415. Kato, N., H. J. Eggers, and H. Rolly. 1969. Inhibition of release of vaccinia virus by Nl-isonicotinoyl-N2-3-methyl-4-chlorobenzoylhydrazine. J. Exp. Med. 129:795-808. Keck, J. G., C. J. Baldick, and B. Moss. 1990. Role of DNA replication in vaccinia virus gene expression: a naked template is required for transcription of three late trans-activator genes. Cell 61:801-809. Kozak, M. 1991. Effects of long 5' leader sequences on initiation by eucaryotic ribosomes in vitro. Gene Expression 1:117-125. Mahr, A., and B. E. Roberts. 1984. Arrangement of late RNAs transcribed from a 7.1-kilobase EcoRI vaccinia virus DNA fragment. J. Virol. 49:510-520. McDonald, W. F., V. Crozel-Goudot, and P. Traktman. 1992. Transient expression of the vaccinia virus DNA polymerase is an intrinsic feature of the early phase of infection and is unlinked to DNA replication and late gene expression. J. Virol. 66:534-547. Moss, B. 1990. Regulation of vaccinia virus transcription. Annu. Rev. Biochem. 59:661-688. Moss, B., E. N. Rosenblum, E. Katz, and P. M. Grimley. 1969.
33. 34.
35.
36.
37.
38.
39. 40.
41.
42.
43. 44.
45.
46.
47. 48. 49.
50. 51.
5401
Rifampicin: a specific inhibitor of vaccinia virus assembly. Nature (London) 224:1280-1284. Pennington, T. H. 1973. Vaccinia virus morphogenesis: a comparison of virus-induced antigens and polypeptides. J. Gen. Virol. 19:65-79. Puckett, C., and B. Moss. 1983. Selective transcription of vaccinia virus genes in template dependent soluble extracts of infected cells. Cell 35:441-448. Rosel, J. L., P. L. Earl, J. P. Weir, and B. Moss. 1986. Conserved TAAATG sequence at the transcriptional translational initiation sites of vaccinia virus late genes deduced by structural and functional analysis of the HindIII H genome fragment. J. Virol. 60:436-439. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. Schmutz, C., L. G. Payne, J. Gubser, and R. Wittek. 1991. A mutation in the gene encoding the vaccinia virus 37,000Mr protein confers resistance to an inhibitor of virus envelopment and release. J. Virol. 65:3435-3442. Schnierle, B. S., and B. Moss. 1992. Vaccinia virus-mediated inhibition of host protein synthesis involves neither degradation nor underphosphorylation of components of the cap-binding eukaryotic translation initiation factor complex eIF-4F. Virology 188:931-933. Schwer, B., and H. G. Stunnenberg. 1988. Vaccinia virus late transcripts generated in vitro have a poly(A) head. EMBO J. 7:1183-1190. Schwer, B., P. Visca, J. C. Vos, and H. G. Stunnenberg. 1987. Discontinuous transcription or RNA processing of vaccinia virus late messengers result in a 5' poly(A) leader. Cell 50:163169. Shuman, S., S. S. Broyles, and B. Moss. 1987. Purification and characterization of a transcription termination factor from vaccinia virions. J. Biol. Chem. 262:12372-12380. Subak-Sharpe, J. H., T. H. Pennington, J. F. Szilagyi, M. C. Timbury, and J. F. Williams. 1970. The effect of rifampicin on mammalian viruses and cells, p. 260-286. In L. Silvestre (ed.), RNA polymerase and transcription. North Holland, Amsterdam. Tartaglia, J., and E. Paoletti. 1985. Physical mapping and DNA sequence analysis of the rifampicin resistance locus in vaccinia virus. Virology 147:394-404. Venkatesan, S., B. M. Baroudy, and B. Moss. 1981. Distinctive nucleotide sequences adjacent to multiple initiation and termination sites of an early vaccinia virus gene. Cell 25:805-813. Venkatesan, S., A. Gershovitz, and B. Moss. 1982. Complete nucleotide sequences of two adjacent early vaccinia virus genes located within the inverted terminal repetition. J. Virol. 44:637646. Vos, J. C., and H. G. Stunnenberg. 1988. Derepression of a novel class of vaccinia virus genes upon DNA replication. EMBO J. 7:3487-3492. Weir, J. P., and B. Moss. 1983. Nucleotide sequence of the vaccinia virus thymidine kinase gene and the nature of spontaneous frameshift mutations. J. Virol. 46:530-537. Wright, C., and B. Moss. 1987. In vitro synthesis of vaccinia virus late mRNA containing a 5' poly(A) leader sequence. Proc. Natl. Acad. Sci. USA 84:8883-8887. Wright, C., and B. Moss. 1989. Identification of factors specific for transcription of the late class of vaccinia virus genes. J. Virol. 63:4224-4233. Yuen, L., A. J. Davison, and B. Moss. 1987. Early promoterbinding factor from vaccinia virions. Proc. Natl. Acad. Sci. USA 84:6069-6073. Yuen, L., and B. Moss. 1987. Oligonucleotide sequence signaling transcriptional termination of vaccinia virus early genes. Proc. Natl. Acad. Sci. USA 84:6417-6421.
