- Email: [email protected]
Activation of the Hepatitis B Virus S Promoter by Transcription Factor NF-Y via a CCAAT Element
VIROLOGY
225, 387–394 (1996) 0613
ARTICLE NO.
Activation of the Hepatitis B Virus S Promoter by Transcription Factor NF-Y via a CCAAT Element CHENG-CHAN LU*,1 and T. S. BENEDICT YEN*,†,2 *Department of Pathology, University of California School of Medicine, San Francisco, California; and †Anatomic Pathology Service, Veterans Affairs Medical Center, 4150 Clement Street, San Francisco, California 94121 Received August 1, 1996; accepted September 18, 1996 The middle and small surface proteins of hepatitis B virus are translated from 5*-heterogeneous transcripts specified by the S promoter. We have generated a series of linker-substitution mutants that encompass the 130 base pairs comprising this promoter and measured the amount of transcripts and protein products synthesized from each mutant. The results confirm our previous finding that a CCAAT element is an important up-stream activating element for this promoter, as mutation of this element leads to a ú20-fold decrease in promoter activity. In vitro binding assays showed that the cellular transcription factor NF-Y (CCAAT-binding factor) binds to this element, and expression of a dominant-negative NF-Y subunit in transfected cells specifically reduced surface protein expression from the S promoter via the CCAAT element. In addition, two Sp1 sites also contribute to S promoter activity by a total of approximately 6-fold. Therefore, the S promoter is activated by both NF-Y and Sp1, but more strongly by the former factor. q 1996 Academic Press, Inc.
Hepatitis B virus (HBV) infection can give rise to chronic hepatitis, cirrhosis, and hepatocellular carcinoma and is estimated to cause 2 million deaths worldwide every year (reviewed by Hollinger, 1990). Its genome in the infected cell is a circular double-stranded DNA episome 3.2 kb in length that depends on host factors for transcription from its four promoters (reviewed by Ganem, 1996; Yen, 1993). As part of our continuing efforts to understand the cis- and trans-acting factors that regulate HBV gene expression, we have been mapping the viral surface gene (S) promoter. This promoter gives rise to 5*-heterogenous transcripts that straddle an ATG codon in the S open-reading frame (ORF). Thus, the largest transcript is translated into the so-called middle surface protein, while the other transcripts are translated into the small (major) surface protein, which initiates at the next ATG codon in the ORF (Fig. 1). Both of these surface proteins, together with the large surface protein specified by a separate, up-stream promoter, are incorporated into progeny virion particles, as the protein components of the viral envelope (Ganem, 1991). Previously, we showed that a host factor binds to a CCAAT element in the S promoter and that this element is important for high level expression of this promoter (Zhou and Yen, 1991). However, we did not characterize the cellular factor that binds to this element, nor did our work rule out a role for other elements, the existence
of which has been demonstrated by other studies (DeMedina et al., 1988; Raney et al., 1991, 1992). Therefore, we have gone on to study the S promoter in greater detail, by constructing a series of linker-substitution mutants that cover 130 bp down-stream of the HincII site, including the CCAAT element and the sites of transcriptional initiation. Quantitation of RNA and protein synthesis directed by these mutants confirms the importance of the CCAAT element for the activity of this promoter. We also demonstrate that NF-Y (CCAAT-binding factor) is the major host factor that binds to and activates transcription via this element. In addition, studies with additional mutants reveal that two Sp1 sites identified by Raney et al. (1992) also contribute to S promoter activity, by further boosting transcription approximately sixfold in the presence of the CCAAT element. Thus, these two cellular transcription proteins are both necessary for full S promoter function. MATERIALS AND METHODS Plasmid constructions The plasmid pHBV1.2, also called adwR9, was obtained from T. J. Liang, Harvard Medical School, and contains slightly more than one complete copy (with terminal redundancy of nucleotides 1420–2187) of the HBV genome, strain adw2, that is competent for replication (Blum et al., 1992). The plasmid pSAgDHin contains the HincII– BglII fragment (Fig. 1) of HBV DNA strain adw2 (Valenzuela et al., 1980) in the vector pTZ19U, as described previously (Zhou and Yen, 1990). A two-part polymerase chain reaction (PCR) procedure was used to substitute
1 Current address: Department of Pathology, Cheng Kung School of Medicine, Kaohsiung, Taiwan, ROC. 2 To whom correspondence and reprint requests should be addressed. Fax: (415) 750-6947. E-mail: [email protected].
0042-6822/96 $18.00
387
AID
VY 8243
/
6a23$$$121
10-17-96 14:43:46
Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
vira
AP: Virology
388
LU AND YEN
successive 10-bp regions of the S promoter in this plasmid with a linker containing a BstEII site (GGTGACCTAA), to create the LS series of mutants (Fig. 1). The first set of primers had the sequences 5*TTGGTTGAGTACTCACCAGTC and 5*TTAGGTCACC(X)16 , where X represents nucleotides homologous to the antisense strand of the S promoter immediately up-stream of the substituted region. The amplified product from this reaction was digested with ScaI and BstEII restriction endonucleases. The second set of primers has the sequences 5*GCCTAGCAGCCATGGAAATG and 5*CGCGGTGACCTAA(Y)16 , where Y represents nucleotides homologous to the sense strand of the S promoter immediately down-stream of the substituted region. The amplified product from this reaction was digested with NcoI and BstEII restriction endonucleases. The two digested PCR products were then ligated together with the small ScaI–NcoI fragment of pSAgDHin. Each resulting mutant plasmid was digested with several restriction endonucleases, to confirm the presence of the introduced BstEII site and the gross integrity of the plasmid. In addition, the S promoter region of each plasmid was sequenced, to verify that the desired mutations were present and that no other mutations had been generated. For each experiment, at least two different clones of each mutant were used to insure that any differences with the wild-type plasmid was not due to cryptic mutations introduced elsewhere into the plasmid during the mutagenesis procedure. A similar strategy was used to generate substitution mutants that knocked out each of the Sp1-binding sites identified by Raney et al. (1992) (see Fig. 1 for sequences). The plasmid with mutations in the up-stream Sp1 site is called LSp1-1 (originally called L73-3), while the plasmid with mutations in the down-stream Sp1 site is called LSp1-2 (originally called L77-1). An additional mutant with both Sp1 sites mutated is called LSp1-12 (originally called L75A-3). The plasmid pCCAAT-S (originally called p33-7) contains a lone CCAAT element in place of the S promoter driving the surface gene and was constructed as follows. The S promoter was excised from pSAg (Zhou and Yen, 1991) with HindIII and PstI restriction endonucleases, and a 64-bp fragment of ‘‘nonspecific’’ sequence was inserted between the HindIII and PstI sites. This fragment was synthesized as two pairs of partially overlapping double-stranded oligonucleotides with HindIII and PstI compatible ends, and correspond in sequence to nucleotides 4201–4254 (antisense) of the chloramphenicol acetyltransferase gene in the plasmid pSV2CAT (Gorman, 1985). Subsequently, a double-stranded oligonucleotide with the sequence
by restriction enzyme digestion patterns and partial sequencing. Cell culture and transfection HuH-7 cells, which are well-differentiated human hepatoma cells free of endogenous HBV DNA, were cultured at 377 in Dulbecco’s modified Eagle’s medium with 10% fetal bovine serum, under an atmosphere of 93% air and 7% CO2 . Cells were transfected in 60- or 100-mm plates with the calcium phosphate coprecipitation method (Gorman, 1985). On the second day after the start of transfection, the cells were harvested for analysis of RNA and/ or protein expression. RNA and protein analysis Total RNA was extracted from cultured cells with RNAzolB (Biotecx), and subjected to primer extension with 32 P-end-labeled oligonucleotide probes, as described previously (Huang and Yen, 1993; Zheng and Yen, 1994; Zhou and Yen, 1990). The primer for S gene transcripts has the sequence 5*AGAGGCAATATTCGGAGCAGGGTTTAC, and the primer for human growth hormone gene transcripts has the sequence 5*GCCACTGCAGCTAGGTGAGCGTCC. The amount of HBV surface protein (HBsAg) in spent media was quantitated by a commercial radioimmunoassay kit (Abbott); when necessary, the samples were diluted with medium to insure that assays were performed in the linear range. To control for variations in transfection efficiency, in most experiments the cells were cotransfected with pTKGH, which contains the herpes simplex thymidine kinase promoter driving the human growth hormone gene (Selden et al., 1986), and secreted growth hormone was quantitated by a commercial radioimmunoassay kit (Nichols). The amount of secreted surface protein was then normalized to the amount of growth hormone, although variations in the transfection efficiency was usually minor. Electrophoretic mobility shift assays
was inserted into the HindIII site, up-stream of the ‘‘nonspecific’’ sequence. The final plasmid was again checked
Nuclear extracts from HuH-7 cells were prepared according to the procedure of Dignam et al. (1983). For detection of protein–DNA binding with the electrophoretic mobility shift assay (EMSA) (Fried and Crothers, 1981), 7 ml of nuclear extracts were preincubated with 1 ml of 1 mg/ml poly[dI-dC] (Boehringer Mannheim) and 1 ml of 1 mg/ml sonicated salmon sperm DNA (Sigma) and then incubated with approximately 50 femtomols (30,000–40,000 cpm) of a double-stranded oligonucleotide, labeled at the 5* ends by polynucleotide kinase and [g-32P]ATP (Sambrook et al., 1989). After 30 min of incubation at room temperature, the mixture was electrophoresed on a 5% polyacrylamide gel in 0.51 Tris–borate–EDTA buffer at 150 V for 1 hr.
AID
vira
5*AGCTCCACCAATCGGCA3* w w w w w www w w w w w 3* GGTGGTTAGCCGTTCGA5*
VY 8243
/
6a23$$$122
10-17-96 14:43:46
AP: Virology
ACTIVATION OF HBV S PROMOTER BY NF-Y
389
FIG. 1. Top, HBV DNA fragment present in pSAgDHin. The middle surface protein is translated from the ORF that includes both the preS2 and surface regions, while the small surface protein is translated from the ORF that includes only the surface region. H, HincII site; B, BamHI site; An , polyadenylation signal. Bottom, HBV sequence, subtype adw2, in the region of the S promoter, between the HincII and BamHI sites. Each of the LS mutants contains the sequence 5*GGTGACCTAA replacing the sequence between the bars. The five major start sites are marked by asterisks. The CCAAT element is shown in bold type. The two Sp1 sites are underlined (Raney et al., 1991, 1992), and the sequences inserted in their place in the Sp1 mutant plasmids (LSp1-1 and LSp1-2) are shown above the main sequence. The nucleotides shown in italic type are different in the ayw subtype used by Raney et al. (1991, 1992).
RESULTS S gene expression from LS mutants To localize all of the important cis-acting elements within the S promoter, we constructed the LS series of mutants, each of which contains a 10-bp linker sequence (GGTGACCTAA) replacing a 10-bp region of the 130-bp S promoter, in the context of the plasmid pSAgDHin, which contains the S promoter driving the entire S gene in its native configuration (Fig. 1). HuH-7 cells were transfected with each of these plasmids, and the S promoter activity was assessed by the amount of surface proteins secreted into the medium, as quantitated by radioimmunoassay. This analysis revealed that LS3 gave rise to õ5% of the wild-type level of secreted HBsAg, while all other mutants gave rise to ú50% of the wild-type level (Fig. 2A). Similar results were obtained when S transcript levels were directly quantitated by primer extension and phosphor-imaging (Fig. 2B). Therefore, the major activating element appeared to be contained within region 3. Region 3 contains the sequence CCAAT (Fig. 1), which has been shown by us previously to act as a positive element and bind a cellular factor. Mutations in several other regions had smaller positive or negative effects (Fig. 2), but because of the relative weakness of these effects, we have not studied them in more detail. In addition, some of the mutations downstream of region 6 affected start site selection (data not shown), in agreement with previous data that sequences in this region have initiator-like functions (Zhou and Yen, 1991). Previous data from Raney et al. (1992) had demonstrated the presence of two Sp1-binding sites within our 130-bp S promoter region (underlined sequences in Fig. 1) that were able to activate this promoter. Our studies seemed to indicate a modest effect of these sites on S promoter activity (LS2 and LS7 in Fig. 2). However, be-
FIG. 2. (A) The amount of surface protein secreted by cells transfected with each of the LS mutants, compared with the parent pSAgDHin (normalized to 100), as quantitated by radioimmunoassay. The results shown are the mean { standard deviation of 3 independent transfections, after normalization to growth hormone synthesized by the cotransfected plasmid pTKGH. (B) The amount of S transcripts synthesized in cells transfected with each of the LS mutants, compared with the parent pSAgDHin (normalized to 100). The S transcripts were detected by primer extension (Zhou and Yen, 1991) and then quantitated with a Molecular Dynamics Phosphorimager (courtesy of Howard Hughes Medical Institute, UCSF).
AID
vira
VY 8243
/
6a23$$$122
10-17-96 14:43:46
AP: Virology
390
LU AND YEN
FIG. 3. The amount of surface protein secreted by cells transfected with each of the Sp1 mutants, compared with the parent pSAgDHin (normalized to 100), after normalization to growth hormone synthesis. The results shown are the mean { standard deviation of 3 independent transfections, after normalization to growth hormone synthesized by the cotransfected plasmid pTKGH.
cause these Sp1-binding sites straddle the LS1/2 and LS6/7 boundaries (Fig. 1), it was possible that no single linker substitution mutation was able to prevent completely the binding of Sp1 to these sites. To examine specifically the role of these Sp1 sites, we generated two new mutants, each of which mutated one of the Sp1 sites (Fig. 1). As seen in Fig. 3, mutation of the up-stream Sp1 site (LSp1-1) had a small negative effect on S promoter activity, as measured by surface protein secretion, while mutation of the down-stream Sp1 site (LSp1-2) led to a drop of approximately fourfold. A double mutant that knocked out both sites (LSp1-12) retained approximately 17% of the activity (Fig. 3). Therefore, these Sp1 sites, especially the down-stream one, do contribute to S promoter activity, by a total of approximately sixfold. To rule out the presence of yet other elements that may have been missed by our linker substitution mutants, we removed the entire S promoter from the plasmid pSAgDHin and inserted in its place the CCAAT element up-stream of 64 bp of irrelevant sequence, to preserve the approximate distance between the CCAAT element and the S ORF; in other words, we deleted all S promoter sequences both 5* and 3* of the CCAAT element. When this plasmid (pCCAAT-S) was transfected into HuH-7 cells, it gave rise to 12 { 3% of the amount of secreted surface protein as the parent pSAgDHin, as determined by radioimmunoassay of media from three independent transfections. This figure is similar to the 17% residual activity of the S promoter with the double Sp1 mutation (Fig. 3). Therefore, it is unlikely that there are other elements that contribute significantly to S promoter activity.
the factor(s). At least five different classes of cellular transcription factors bind to CCAAT or related DNA sequences. Their names and consensus binding sites are shown in Table 1. To determine which, if any, of these factors may be binding to the S promoter CCAAT element, we performed competition EMSA, using the HBV CCAAT element as the labeled probe, and the various consensus binding sites as cold competitors. As seen in Fig. 4, a single shifted band was produced upon incubation of the CCAAT element with HuH-7 nuclear extracts, as expected from previous studies (Zhou and Yen, 1991). The c/EBP-binding site did not compete with the CCAAT element (lanes 13 and 14), while the CTF-binding site competed only weakly (lanes 11 and 12). In contrast, the binding sites for NF-Y, YB-1, and CBF (Fig. 4, lanes 5 and 6, 7 and 8, and 9 and 10, respectively) all competed as well as the CCAAT element itself (Fig. 4, lanes 3 and 4). Therefore, the cellular factor that binds to the S promoter CCAAT element is presumably NF-Y, YB-1, or CBF. To distinguish among these factors, we employed rabbit antibodies that could specifically recognize the A subunit of NF-Y (courtesy of B. de Crombrugghe, M. D. Anderson Cancer Center) in the gel shift analysis. As seen in Fig. 5, lane 3, inclusion of these antibodies led to the complete disappearance of the shifted band (SB) and the appearance of a new band with slower migration rate (SS), consistent with a ternary complex of the oligonucleotide, NF-Y, and immunoglobulins (so-called ‘‘supershift’’). This new band is weaker, presumably reflecting the ability of some of the antibodies to compete for DNA binding by NF-Y. In contrast, an antibody against YB-1 (Wright et al., 1994; courtesy of J. Ting, University of North Carolina) did not produce this change (Fig. 5, lane 4), nor did an irrelevant antibody (data not shown). Therefore, the shifted band must have resulted from NFY binding to the CCAAT element. Since no other shifted bands were present, it is unlikely that other cellular factors bound this element. Based on the above data, we conclude that NF-Y is the only HuH-7 cellular factor that binds to the S promoter CCAAT element in vitro. Activation of the S promoter by NF-Y
We had previously shown by foot-print and electrophoretic mobility shift analyses (EMSA) that cellular factor(s) present in HuH-7 cells bind to the S promoter CCAAT element (Zhou and Yen, 1991), but we did not identify
While we have demonstrated that NF-Y is the major factor present in HuH-7 cells that binds to the CCAAT element in vitro, such binding does not necessarily reflect activation in vivo. Unfortunately, NF-Y is expressed at high levels in all eucaryotic cells examined. Thus, we could not look for activation of the S promoter by cotransfected NF-Y. As an alternative, we used the plasmid D4YA13m29 (courtesy of R. Mantovani, Universita degli Studi di Milano), which expresses a dominant negative mutant of the A sub-unit of NF-Y and specifically represses transcription from promoters containing NF-Y binding sites (Jackson et al., 1995; Mantovani et al., 1994). Indeed, cotransfection of this plasmid caused an almost
AID
vira
Binding of CCAAT element by NF-Y
VY 8243
/
6a23$$$122
10-17-96 14:43:46
AP: Virology
ACTIVATION OF HBV S PROMOTER BY NF-Y
391
TABLE 1 Oligonucleotides Used for EMSA Namea
Sequenceb
Referencesc
HBV CCAAT Element CBF c/EBP CTF (NF-1) NF-Y (CBF)a YB-1 (EFIA , dbp)
CTCCACCAATCGGCAGTCAG CTGAGCCAATCAGCAGTCAG ATTGCGCAATCTGCAGTGCA CTCCGAAATTTGGCATCATG CTTAACCAATCAGCAGTCAG CTTGGCCAATCAGCAGTCAG
Zhou and Yen (1991) Lum et al. (1990) Cao et al. (1991) Santoro et al. (1988) Vuorio et al. (1990) Ozer et al. (1990)
a The names used in this paper are given first, and other names commonly used in the literature are given in parentheses. It should be noted that while NF-Y is also called CBF, it bears no relationship at all to the other factors named CBF. b The double-stranded oligonucleotides used consist of two direct repeats of the sequences shown and flanked by EcoRI sites. The sequence differences between the concensus factor-binding sites and the HBV CCAAT element are in italics. c The references cited refer only to papers describing the cDNA clones used in this paper, in cases where multiple groups independently cloned the cDNA.
10-fold reduction of surface protein synthesis from the wild-type pSAgDHin (Fig. 6A). The level of repression is similar to that seen with other promoters with NF-Y sites (Jackson et al., 1995). In contrast, this dominant negative mutant induced no significant decrease in surface protein expression from a plasmid containing a linker-substitution mutation of the CCAAT element (Fig. 6A). These results confirmed that NF-Y is the major factor that transactivates the S promoter via the CCAAT element. In addition, we found that cotransfection of CTF, c/EBP, CBF, or YB-1 expression plasmids had no measurable effect on S promoter activity (data not shown). Thus, none of these other factors can activate the S promoter, in agreement with the lack of binding seen in vitro.
Since the plasmid pSAgDHin used in the above studies contains a subgenomic fragment of HBV DNA, it was possible that cis-elements not present in this fragment may modulate the effect of NF-Y on the S promoter CCAAT element. To rule out this possibility, we also examined the effect of the dominant negative NF-YA protein on surface gene expression from pHBV1.2, a plasmid containing a
FIG. 4. Binding of host cell factors to the HBV CCAAT element, and competition by the concensus sites of various mammalian transcription factors that bind to CCAAT or related sequences, as assessed by EMSA. See Table 1 for the DNA sequences of the various elements. FP, free probe; SB, specifically shifted band.
FIG. 5. Effect of antisera against NF-Y and YB-1 on the binding of host cell factors to the HBV CCAAT element. In the indicated lanes, 1 ml of rabbit antiserum against the A subunit of NF-Y (courtesy of B. de Crombrugghe) or against YB-1 (courtesy of J. Ting) was included in the mixture. FP, free probe; SB, shifted band; SS, ‘‘supershifted’’ band specifically induced by rabbit antiserum against NF-Y.
AID
vira
VY 8243
/
6a23$$$122
10-17-96 14:43:46
AP: Virology
392
LU AND YEN
DISCUSSION
replication-competent, greater-than-full-length clone of the HBV genome (Blum et al., 1992). As seen in Fig. 6B, the mutant NF-YA protein also drastically down-regulated surface gene expression from this plasmid, as measured by surface antigen secretion. As expected, a CCAAT element mutant of pHBV1.2 produced much lower amounts of surface antigen, but was no longer down-regulated by the dominant negative NF-YA (Fig. 6B). There was no significant difference in transfection efficiency between the wildtype and mutant pHBV1.2 plasmids, since the amount of secreted e antigen (a product of the core gene) was comparable (data not shown).
Using a series of plasmids containing linker-substitution mutations in the S promoter of a subgenomic HBV fragment, we demonstrated that the CCAAT element upstream of the start sites is the major positive-acting cis-element of this promoter. In addition, two Sp1 sites together further increased S promoter activity by approximately sixfold. No other activating elements with significant activity were found. However, our studies have not addressed the functions of the elements responsible for correct initiation, which are down-stream of the NF-Y and Sp1 sites and appear to be very complex (Lu and Yen, unpublished data). We have gone on to demonstrate that the major cellular transcription factor that binds in vitro to the S promoter CCAAT element is the heterotrimeric protein, NF-Y (also known as CCAAT-binding factor) (Becker et al., 1991; Maity et al., 1992; Sinha et al., 1995; van Huijsduijnen et al., 1990; Vuorio et al., 1990). In addition, experiments with a dominant negative mutant of NF-Y indicated that NF-Y also activate via this CCAAT element in cultured cells. Since NF-Y is ubiquitously expressed, this finding correlates with the relatively non-cell-type-specific activity of this promoter (Antonucci and Rutter, 1989). The question arises as to why two other cellular factors, YB1 and CBF, neither bind to nor activate via this CCAAT element, since its sequence is extremely similar to the published concensus binding sites for these two factors (Table 1). In fact, concensus binding sites for YB-1 and CBF compete efficiently for NF-Y binding to the HBV CCAAT element (Fig. 5), and bind NF-Y in HuH-7 extracts when used in EMSA experiments (Lu and Yen, published data). Since CBF appears to be a heat shock-induced factor that has specialized functions (Lum et al., 1990), it is not entirely surprising that it is not active in unshocked cells and does not play a role in S promoter activity. However, normal hepatocytes (Spitkovsky et al., 1992) and HuH-7 cells (Lu and Yen, unpublished data) are known to express high levels of YB-1 mRNA. Thus, it is at first glance somewhat surprising that YB-1 apparently neither binds to the CCAAT element, nor influences S promoter activity. Interestingly, however, YB-1 has been shown to bind RNA and single-stranded DNA (Gai et al., 1992; Hasegawa et al., 1991), and one of its homologs in frog oocytes is involved in RNA storage and translational control (Bouvet and Wolffe, 1994). Furthermore, over-expression of YB-1 actually decreases expression from class II major histocompatibility promoters (Ting et al., 1994; Lloberas et al., 1995). Therefore, it is possible that YB-1 may not function as a classical double-stranded DNA-binding transcription factor. Finally, it should be noted that similar CCAAT elements within the albumin, grp78, farnesyl diphosphate synthase, and 3-hydroxy-3methylglutaryl-coenzyme A synthase promoters have all been shown to be specifically bound by NF-Y but not by
AID
vira
FIG. 6. (A) Effect of a dominant negative mutant of NF-Y, A subunit, on surface protein expression from the wild-type pSAgDHin, or a CCAAT element mutant (LS3). Each 100-mm plate of HuH-7 cells was cotransfected with 4 mg of the surface gene plasmid and with either 2 mg of the mutant NF-YA expression plasmid (D4YA13m29) (Jackson et al., 1995; Mantovani et al., 1994) (right half) or 2 mg of pUC18 (left half). The results are expressed as the percentage relative to surface protein secretion from cells cotransfected with wild-type pSAgDHin and pUC18 and represent the mean { standard deviation of 3 independent transfections. (B) Effect of a dominant negative mutant of NF-Y, A subunit, on surface protein expression from the wild-type pHBV1.2, or a CCAAT element mutant (LS3). Each 100-mm plate of HuH-7 cells was cotransfected with 4 mg of the HBV plasmid and with either 2 mg of the mutant NF-YA expression plasmid (right half) or 2 mg of pUC18 (left half). The results are expressed as the percentage relative to surface protein secretion from cells cotransfected with wild-type pHBV1.2 and pUC18 and represent the mean { standard deviation of 3 independent transfections.
VY 8243
/
6a23$$$123
10-17-96 14:43:46
AP: Virology
ACTIVATION OF HBV S PROMOTER BY NF-Y
other CCAAT-binding proteins, including YB-1 (Jackson et al., 1995; Roy and Lee, 1995; Tronche et al., 1991). Thus, our results on the S promoter CCAAT element are entirely consistent with previous studies. In summary, we have shown that the HBV S promoter is relatively compact and is mainly activated by a single CCAAT element that binds NF-Y, with additional activity contributed by two Sp1 sites. Our conclusions are somewhat similar but not identical to those reached by Raney et al. (1991, 1992), in that their data revealed approximately threefold activation of the S promoter by the upstream Sp1 site (site 2 in their nomenclature), and little effect of the down-stream Sp1 site (their site 3). Thus, the relative contribution of these Sp1 sites to S promoter activity is the reverse of what we observe (stronger activity of the down-stream versus the up-stream site). More significantly, Raney et al. (1991, 1992) found only a twoto fourfold activation of this promoter by the CCAAT element, but they found that a third Sp1 site (their site 1), up-stream of our 130-bp minimal S promoter, activated this promoter by approximately fourfold. In contrast, we do not find evidence for any important positive-acting element up-stream of the minimal promoter region (Zhou and Yen, 1991). The reason for these disrepancies is not clear and cannot be attributed to our using the minimal S promoter fragment, since the CCAAT element is also important for S promoter activity in the context of the entire HBV genome that is competent for replication (Fig. 6B). Nor can cell-type difference be the cause, since both groups used HuH-7 cells. Perhaps part of the differences can be attributed to the fact that we used the adw2 subtype of HBV in these experiments, while Raney et al. (1991, 1992) used the ayw sub-type. While the CCAAT element is conserved among all sequenced HBV genomes (Zhou and Yen, 1991), the sequences in the Sp1 sites are not (nucleotides shown in italics in Fig. 1). Therefore, it is possible that some of the Sp1 sites in the ayw sub-type bind Sp1 more strongly than the corresponding sites in adw2. It should also be noted that Raney et al. (1991, 1992) used heterologous reporter gene constructs that give rise to an interfering up-stream ATG codon in some but not all of the S promoter transcripts, that may complicate data analysis based solely on the amount of reporter protein synthesized. Finally, it remains a possibility that the Sp1 sites bind other, as yet unidentified transcription factor(s), in addition to Sp1, and that it is these other factor(s) that are mainly responsible for activating the S promoter. Resolution of these and other questions on S promoter function await further experimentation.
393
on a shared instrument purchased with UCSF Academic Senate Opportunity funds. This work was funded by NIH Grant R01CA55578.
REFERENCES
We thank Z. Cao, L. Guarente, J. Liang, R. Mantovani, S. McKnight, L. Sealy, R. Tjian, B. Wu, and Y. W. Zheng for plasmids and B. deCrombrugghe and J. Ting for antisera. Oligonucleotides were synthesized
Antonucci, T. K., and Rutter, W. J. (1989). Hepatitis B virus (HBV) promoters are regulated by the HBV enhancer in a tissue-specific manner. J. Virol. 63, 579–583. Becker, D. M., Fikes, J. D., and Guarente, L. (1991). A cDNA encoding a human CCAAT-binding protein cloned by functional complementation in yeast. Proc. Natl. Acad. Sci. USA 88, 1968–1972. Blum, H. E., Zhang, Z. S., Galun, E., von, W. F., Garner, B., Liang, T. J., and Wands, J. R. (1992). Hepatitis B virus X protein is not central to the viral life cycle in vitro. J. Virol. 66, 1223–1227. Bouvet, P., and Wolffe, A. P. (1994). A role for transcription and FRGY2 in masking maternal mRNA within Xenopus oocytes. Cell 77, 931– 941. Cao, Z., Umek, R. M., and McKnight, S. L. (1991). Regulated expression of three C/EBP isoforms during adipose conversion of 3T3-L1 cells. Genes Dev. 5, 1538–1552. De-Medina, T., Faktor, O., and Shaul, Y. (1988). The S promoter of hepatitis B virus is regulated by positive and negative elements. Mol. Cell Biol. 8, 2449–2455. Dignam, J. D., Lebovitz, R. M., and Roeder, R. G. (1983). Accurate transcription initiation by RNA polymerase II in a soluble extract from isolated mammalian nuclei. Nucleic Acids Res. 11, 1475–1489. Fried, M., and Crothers, D. (1981). Equilibiria and kinetics of lac repressor-operator interactions by polyacrylamide gel electrophoresis. Nucleic Acids Res. 9, 6505–6525. Gai, X. X., Lipson, K. E., and Prystowsky, M. B. (1992). Unusual DNA binding characteristics of an in vitro translation product of the CCAAT binding protein mYB-1. Nucleic Acids Res. 20, 601–606. Ganem, D. (1991). Assembly of hepadnaviral virions and subviral particles. Curr. Top. Microbiol. Immunol. 168, 61–83. Ganem, D. (1996). Hepadnaviridae and their replication. In ‘‘Virology’’ (B. N. Fields and D. M. Knipe, Eds.), pp. 2703–2737. Raven Press, New York. Gorman, C. (1985). High efficiency gene transfer into mammalian cells. In ‘‘DNA Cloning’’ (D. M. Glover, Ed.), pp. 143–190. IRL, Oxford. Hasegawa, S. L., Doetsch, P. W., Hamilton, K. K., Martin, A. M., Okenquist, S. A., Lenz, J., and Boss, J. M. (1991). DNA binding properties of YB-1 and dbpA: Binding to double-stranded, single-stranded, and abasic site containing DNAs. Nucleic Acids Res. 19, 4915–4920. Hollinger, F. B. (1990). Hepatitis B Virus. In ‘‘Virology’’ (B. N. Fields and D. M. Knipe, Eds.), pp. 2171–2238. Raven Press, New York. Huang, Z. M., and Yen, T. S. B. (1993). Dys-regulated surface gene expression from disrupted hepatitis B virus genomes. J. Virol. 67, 7032– 7040. Jackson, S. M., Ericsson, J., Osborne, T. F., and Edwards, P. A. (1995). NF-Y has a novel role in sterol-dependent transcription of two cholesterogenic genes. J. Biol. Chem. 270, 21445–21448. Lloberas, J., Maki, R. A., and Celada, A. (1995). Repression of major histocompatibility complex I-A beta gene expression by dbpA and dbpB (mYB-1) proteins. Mol. Cell Biol. 15, 5092–5099. Lum, L. S., Sultzman, L. A., Kaufman, R. J., Linzer, D. I., and Wu, B. J. (1990). A cloned human CCAAT-box-binding factor stimulates transcription from the human hsp70 promoter. Mol. Cell Biol. 10, 6709– 6717. Maity, S. N., Sinha, S., Ruteshouser, E. C., and de Crombrugghe, B. (1992). Three different polypeptides are necessary for DNA binding of the mammalian heteromeric CCAAT binding factor. J. Biol. Chem. 267, 16574–16580. Mantovani, R., Li, X. Y., Pessara, U., van, H. R., Benoist, C., and Mathis, D. (1994). Dominant negative analogs of NF-YA. J. Biol. Chem. 269, 20340–20346. Ozer, J., Faber, M., Chalkley, R., and Sealy, L. (1990). Isolation and characterization of a cDNA clone for the CCAAT transcription factor
AID
vira
ACKNOWLEDGMENTS
VY 8243
/
6a23$$$123
10-17-96 14:43:46
AP: Virology
394
LU AND YEN
EFIA reveals a novel structural motif. J. Biol. Chem. 265, 22143– 22152. Raney, A. K., Le, H. B., and McLachlan, A. (1992). Regulation of transcription from the hepatitis B virus major surface antigen promoter by the Sp1 transcription factor. J. Virol. 66, 6912–6921. Raney, A. K., Milich, D. R., and McLachlan, A. (1991). Complex regulation of transcription from the hepatitis B virus major surface antigen promoter in human hepatoma cell lines. J. Virol. 65, 4805–4011. Roy, B., and Lee, A. S. (1995). Transduction of calcium stress through interaction of the human transcription factor CBF with the proximal CCAAT regulatory element of the grp78/BiP promoter. Mol. Cell Biol. 15, 2263–2274. Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989). ‘‘Molecular Cloning: A Laboratory Manual,’’ 2nd ed., Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. Santoro, C., Mermod, N., Andrews, P. C., and Tjian, R. (1988). A family of human CCAAT-box-binding proteins active in transcription and DNA replication: Cloning and expression of multiple cDNAs. Nature 334, 218–224. Selden, R. F., Howie, K. B., Rowe, M. E., Goodman, H. M., and Moore, D. D. (1986). Human growth hormone as a reporter gene in regulation studies employing transient gene expression. Mol. Cell Biol. 6, 3173– 3179. Sinha, S., Maity, S. N., Lu, J., and de Crombrugghe, B. (1995). Recombinant rat CBF-C, the third subunit of CBF/NFY, allows formation of a protein-DNA complex with CBF-A and CBF-B and with yeast HAP2 and HAP3. Proc. Natl. Acad. Sci. USA 92, 1624–1628. Spitkovsky, D. D., Royer, P. B., Delius, H., Kisseljov, F., Jenkins, N. A., Gilbert, D. J., Copeland, N. G., and Royer, H. D. (1992). Tissue restricted expression and chromosomal localization of the YB-1 gene encoding a 42 kD nuclear CCAAT binding protein. Nucleic Acids Res. 20, 797–803. Ting, J. P., Painter, A., Zeleznik, L. N., MacDonald, G., Moore, T. M.,
Brown, A., and Schwartz, B. D. (1994). YB-1 DNA-binding protein represses interferon gamma activation of class II major histocompatibility complex genes. J. Exp. Med. 179, 1605–1611. Tronche, F., Rollier, A., Sourdive, D., Cereghini, S., and Yaniv, M. (1991). NFY or a related CCAAT binding factor can be replaced by other transcriptional activators for cooperation with HNF1 in driving the rat albumin promoter in vivo. J. Mol. Biol. 222, 31–43. Valenzuela, P., Quiroga, M., Zaldivar, J., Gray, P., and Rutter, W. (1980). The nucleotide sequence of the hepatitis B genome and the identification of the major viral genes. In ‘‘Animal Virus Genetics’’ (B. Fields, R. Jaenisch, and C. Fox, Eds.), pp. 57–70. Academic Press, New York. van Huijsduijnen, R., Li, X. Y., Black, D., Matthes, H., Benoist, C., and Mathis, D. (1990). Co-evolution from yeast to mouse: cDNA cloning of the two NF-Y (CP-1/CBF) subunits. EMBO J. 9, 3119–3127. Vuorio, T., Maity, S. N., and de Crombrugghe, B. (1990). Purification and molecular cloning of the ‘‘A’’ chain of a rat heteromeric CCAAT-binding protein. Sequence identity with the yeast HAP3 transcription factor. J. Biol. Chem. 265, 22480–22486. Wright, K. L., Vilen, B. J., Itoh, L. Y., Moore, T. L., Li, G., Criscitiello, M., Cogswell, P., Clarke, J. B., and Ting, J. P. (1994). CCAAT box binding protein NF-Y facilitates in vivo recruitment of upstream DNA binding transcription factors. EMBO J. 13, 4042–4053. Yen, T. S. B. (1993). Regulation of hepatitis B virus gene expression. Semin. Virol. 4, 33–42. Zheng, Y. W., and Yen, T. S. B. (1994). Negative regulation of hepatitis B virus gene expression and replication by oxidative stress. J. Biol. Chem. 269, 8857–8862. Zhou, D. X., and Yen, T. S. B. (1990). Differential regulation of the hepatitis B viral surface gene promoters by a second viral enhancer. J. Biol. Chem. 265, 20731–20734. Zhou, D. X., and Yen, T. S. B. (1991). The hepatitis B virus S promoter comprises a CCAAT motif and two initiation regions. J. Biol. Chem. 266, 23416–23421.
AID
vira
VY 8243
/
6a23$$$123
10-17-96 14:43:46
AP: Virology
225, 387–394 (1996) 0613
ARTICLE NO.
Activation of the Hepatitis B Virus S Promoter by Transcription Factor NF-Y via a CCAAT Element CHENG-CHAN LU*,1 and T. S. BENEDICT YEN*,†,2 *Department of Pathology, University of California School of Medicine, San Francisco, California; and †Anatomic Pathology Service, Veterans Affairs Medical Center, 4150 Clement Street, San Francisco, California 94121 Received August 1, 1996; accepted September 18, 1996 The middle and small surface proteins of hepatitis B virus are translated from 5*-heterogeneous transcripts specified by the S promoter. We have generated a series of linker-substitution mutants that encompass the 130 base pairs comprising this promoter and measured the amount of transcripts and protein products synthesized from each mutant. The results confirm our previous finding that a CCAAT element is an important up-stream activating element for this promoter, as mutation of this element leads to a ú20-fold decrease in promoter activity. In vitro binding assays showed that the cellular transcription factor NF-Y (CCAAT-binding factor) binds to this element, and expression of a dominant-negative NF-Y subunit in transfected cells specifically reduced surface protein expression from the S promoter via the CCAAT element. In addition, two Sp1 sites also contribute to S promoter activity by a total of approximately 6-fold. Therefore, the S promoter is activated by both NF-Y and Sp1, but more strongly by the former factor. q 1996 Academic Press, Inc.
Hepatitis B virus (HBV) infection can give rise to chronic hepatitis, cirrhosis, and hepatocellular carcinoma and is estimated to cause 2 million deaths worldwide every year (reviewed by Hollinger, 1990). Its genome in the infected cell is a circular double-stranded DNA episome 3.2 kb in length that depends on host factors for transcription from its four promoters (reviewed by Ganem, 1996; Yen, 1993). As part of our continuing efforts to understand the cis- and trans-acting factors that regulate HBV gene expression, we have been mapping the viral surface gene (S) promoter. This promoter gives rise to 5*-heterogenous transcripts that straddle an ATG codon in the S open-reading frame (ORF). Thus, the largest transcript is translated into the so-called middle surface protein, while the other transcripts are translated into the small (major) surface protein, which initiates at the next ATG codon in the ORF (Fig. 1). Both of these surface proteins, together with the large surface protein specified by a separate, up-stream promoter, are incorporated into progeny virion particles, as the protein components of the viral envelope (Ganem, 1991). Previously, we showed that a host factor binds to a CCAAT element in the S promoter and that this element is important for high level expression of this promoter (Zhou and Yen, 1991). However, we did not characterize the cellular factor that binds to this element, nor did our work rule out a role for other elements, the existence
of which has been demonstrated by other studies (DeMedina et al., 1988; Raney et al., 1991, 1992). Therefore, we have gone on to study the S promoter in greater detail, by constructing a series of linker-substitution mutants that cover 130 bp down-stream of the HincII site, including the CCAAT element and the sites of transcriptional initiation. Quantitation of RNA and protein synthesis directed by these mutants confirms the importance of the CCAAT element for the activity of this promoter. We also demonstrate that NF-Y (CCAAT-binding factor) is the major host factor that binds to and activates transcription via this element. In addition, studies with additional mutants reveal that two Sp1 sites identified by Raney et al. (1992) also contribute to S promoter activity, by further boosting transcription approximately sixfold in the presence of the CCAAT element. Thus, these two cellular transcription proteins are both necessary for full S promoter function. MATERIALS AND METHODS Plasmid constructions The plasmid pHBV1.2, also called adwR9, was obtained from T. J. Liang, Harvard Medical School, and contains slightly more than one complete copy (with terminal redundancy of nucleotides 1420–2187) of the HBV genome, strain adw2, that is competent for replication (Blum et al., 1992). The plasmid pSAgDHin contains the HincII– BglII fragment (Fig. 1) of HBV DNA strain adw2 (Valenzuela et al., 1980) in the vector pTZ19U, as described previously (Zhou and Yen, 1990). A two-part polymerase chain reaction (PCR) procedure was used to substitute
1 Current address: Department of Pathology, Cheng Kung School of Medicine, Kaohsiung, Taiwan, ROC. 2 To whom correspondence and reprint requests should be addressed. Fax: (415) 750-6947. E-mail: [email protected].
0042-6822/96 $18.00
387
AID
VY 8243
/
6a23$$$121
10-17-96 14:43:46
Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
vira
AP: Virology
388
LU AND YEN
successive 10-bp regions of the S promoter in this plasmid with a linker containing a BstEII site (GGTGACCTAA), to create the LS series of mutants (Fig. 1). The first set of primers had the sequences 5*TTGGTTGAGTACTCACCAGTC and 5*TTAGGTCACC(X)16 , where X represents nucleotides homologous to the antisense strand of the S promoter immediately up-stream of the substituted region. The amplified product from this reaction was digested with ScaI and BstEII restriction endonucleases. The second set of primers has the sequences 5*GCCTAGCAGCCATGGAAATG and 5*CGCGGTGACCTAA(Y)16 , where Y represents nucleotides homologous to the sense strand of the S promoter immediately down-stream of the substituted region. The amplified product from this reaction was digested with NcoI and BstEII restriction endonucleases. The two digested PCR products were then ligated together with the small ScaI–NcoI fragment of pSAgDHin. Each resulting mutant plasmid was digested with several restriction endonucleases, to confirm the presence of the introduced BstEII site and the gross integrity of the plasmid. In addition, the S promoter region of each plasmid was sequenced, to verify that the desired mutations were present and that no other mutations had been generated. For each experiment, at least two different clones of each mutant were used to insure that any differences with the wild-type plasmid was not due to cryptic mutations introduced elsewhere into the plasmid during the mutagenesis procedure. A similar strategy was used to generate substitution mutants that knocked out each of the Sp1-binding sites identified by Raney et al. (1992) (see Fig. 1 for sequences). The plasmid with mutations in the up-stream Sp1 site is called LSp1-1 (originally called L73-3), while the plasmid with mutations in the down-stream Sp1 site is called LSp1-2 (originally called L77-1). An additional mutant with both Sp1 sites mutated is called LSp1-12 (originally called L75A-3). The plasmid pCCAAT-S (originally called p33-7) contains a lone CCAAT element in place of the S promoter driving the surface gene and was constructed as follows. The S promoter was excised from pSAg (Zhou and Yen, 1991) with HindIII and PstI restriction endonucleases, and a 64-bp fragment of ‘‘nonspecific’’ sequence was inserted between the HindIII and PstI sites. This fragment was synthesized as two pairs of partially overlapping double-stranded oligonucleotides with HindIII and PstI compatible ends, and correspond in sequence to nucleotides 4201–4254 (antisense) of the chloramphenicol acetyltransferase gene in the plasmid pSV2CAT (Gorman, 1985). Subsequently, a double-stranded oligonucleotide with the sequence
by restriction enzyme digestion patterns and partial sequencing. Cell culture and transfection HuH-7 cells, which are well-differentiated human hepatoma cells free of endogenous HBV DNA, were cultured at 377 in Dulbecco’s modified Eagle’s medium with 10% fetal bovine serum, under an atmosphere of 93% air and 7% CO2 . Cells were transfected in 60- or 100-mm plates with the calcium phosphate coprecipitation method (Gorman, 1985). On the second day after the start of transfection, the cells were harvested for analysis of RNA and/ or protein expression. RNA and protein analysis Total RNA was extracted from cultured cells with RNAzolB (Biotecx), and subjected to primer extension with 32 P-end-labeled oligonucleotide probes, as described previously (Huang and Yen, 1993; Zheng and Yen, 1994; Zhou and Yen, 1990). The primer for S gene transcripts has the sequence 5*AGAGGCAATATTCGGAGCAGGGTTTAC, and the primer for human growth hormone gene transcripts has the sequence 5*GCCACTGCAGCTAGGTGAGCGTCC. The amount of HBV surface protein (HBsAg) in spent media was quantitated by a commercial radioimmunoassay kit (Abbott); when necessary, the samples were diluted with medium to insure that assays were performed in the linear range. To control for variations in transfection efficiency, in most experiments the cells were cotransfected with pTKGH, which contains the herpes simplex thymidine kinase promoter driving the human growth hormone gene (Selden et al., 1986), and secreted growth hormone was quantitated by a commercial radioimmunoassay kit (Nichols). The amount of secreted surface protein was then normalized to the amount of growth hormone, although variations in the transfection efficiency was usually minor. Electrophoretic mobility shift assays
was inserted into the HindIII site, up-stream of the ‘‘nonspecific’’ sequence. The final plasmid was again checked
Nuclear extracts from HuH-7 cells were prepared according to the procedure of Dignam et al. (1983). For detection of protein–DNA binding with the electrophoretic mobility shift assay (EMSA) (Fried and Crothers, 1981), 7 ml of nuclear extracts were preincubated with 1 ml of 1 mg/ml poly[dI-dC] (Boehringer Mannheim) and 1 ml of 1 mg/ml sonicated salmon sperm DNA (Sigma) and then incubated with approximately 50 femtomols (30,000–40,000 cpm) of a double-stranded oligonucleotide, labeled at the 5* ends by polynucleotide kinase and [g-32P]ATP (Sambrook et al., 1989). After 30 min of incubation at room temperature, the mixture was electrophoresed on a 5% polyacrylamide gel in 0.51 Tris–borate–EDTA buffer at 150 V for 1 hr.
AID
vira
5*AGCTCCACCAATCGGCA3* w w w w w www w w w w w 3* GGTGGTTAGCCGTTCGA5*
VY 8243
/
6a23$$$122
10-17-96 14:43:46
AP: Virology
ACTIVATION OF HBV S PROMOTER BY NF-Y
389
FIG. 1. Top, HBV DNA fragment present in pSAgDHin. The middle surface protein is translated from the ORF that includes both the preS2 and surface regions, while the small surface protein is translated from the ORF that includes only the surface region. H, HincII site; B, BamHI site; An , polyadenylation signal. Bottom, HBV sequence, subtype adw2, in the region of the S promoter, between the HincII and BamHI sites. Each of the LS mutants contains the sequence 5*GGTGACCTAA replacing the sequence between the bars. The five major start sites are marked by asterisks. The CCAAT element is shown in bold type. The two Sp1 sites are underlined (Raney et al., 1991, 1992), and the sequences inserted in their place in the Sp1 mutant plasmids (LSp1-1 and LSp1-2) are shown above the main sequence. The nucleotides shown in italic type are different in the ayw subtype used by Raney et al. (1991, 1992).
RESULTS S gene expression from LS mutants To localize all of the important cis-acting elements within the S promoter, we constructed the LS series of mutants, each of which contains a 10-bp linker sequence (GGTGACCTAA) replacing a 10-bp region of the 130-bp S promoter, in the context of the plasmid pSAgDHin, which contains the S promoter driving the entire S gene in its native configuration (Fig. 1). HuH-7 cells were transfected with each of these plasmids, and the S promoter activity was assessed by the amount of surface proteins secreted into the medium, as quantitated by radioimmunoassay. This analysis revealed that LS3 gave rise to õ5% of the wild-type level of secreted HBsAg, while all other mutants gave rise to ú50% of the wild-type level (Fig. 2A). Similar results were obtained when S transcript levels were directly quantitated by primer extension and phosphor-imaging (Fig. 2B). Therefore, the major activating element appeared to be contained within region 3. Region 3 contains the sequence CCAAT (Fig. 1), which has been shown by us previously to act as a positive element and bind a cellular factor. Mutations in several other regions had smaller positive or negative effects (Fig. 2), but because of the relative weakness of these effects, we have not studied them in more detail. In addition, some of the mutations downstream of region 6 affected start site selection (data not shown), in agreement with previous data that sequences in this region have initiator-like functions (Zhou and Yen, 1991). Previous data from Raney et al. (1992) had demonstrated the presence of two Sp1-binding sites within our 130-bp S promoter region (underlined sequences in Fig. 1) that were able to activate this promoter. Our studies seemed to indicate a modest effect of these sites on S promoter activity (LS2 and LS7 in Fig. 2). However, be-
FIG. 2. (A) The amount of surface protein secreted by cells transfected with each of the LS mutants, compared with the parent pSAgDHin (normalized to 100), as quantitated by radioimmunoassay. The results shown are the mean { standard deviation of 3 independent transfections, after normalization to growth hormone synthesized by the cotransfected plasmid pTKGH. (B) The amount of S transcripts synthesized in cells transfected with each of the LS mutants, compared with the parent pSAgDHin (normalized to 100). The S transcripts were detected by primer extension (Zhou and Yen, 1991) and then quantitated with a Molecular Dynamics Phosphorimager (courtesy of Howard Hughes Medical Institute, UCSF).
AID
vira
VY 8243
/
6a23$$$122
10-17-96 14:43:46
AP: Virology
390
LU AND YEN
FIG. 3. The amount of surface protein secreted by cells transfected with each of the Sp1 mutants, compared with the parent pSAgDHin (normalized to 100), after normalization to growth hormone synthesis. The results shown are the mean { standard deviation of 3 independent transfections, after normalization to growth hormone synthesized by the cotransfected plasmid pTKGH.
cause these Sp1-binding sites straddle the LS1/2 and LS6/7 boundaries (Fig. 1), it was possible that no single linker substitution mutation was able to prevent completely the binding of Sp1 to these sites. To examine specifically the role of these Sp1 sites, we generated two new mutants, each of which mutated one of the Sp1 sites (Fig. 1). As seen in Fig. 3, mutation of the up-stream Sp1 site (LSp1-1) had a small negative effect on S promoter activity, as measured by surface protein secretion, while mutation of the down-stream Sp1 site (LSp1-2) led to a drop of approximately fourfold. A double mutant that knocked out both sites (LSp1-12) retained approximately 17% of the activity (Fig. 3). Therefore, these Sp1 sites, especially the down-stream one, do contribute to S promoter activity, by a total of approximately sixfold. To rule out the presence of yet other elements that may have been missed by our linker substitution mutants, we removed the entire S promoter from the plasmid pSAgDHin and inserted in its place the CCAAT element up-stream of 64 bp of irrelevant sequence, to preserve the approximate distance between the CCAAT element and the S ORF; in other words, we deleted all S promoter sequences both 5* and 3* of the CCAAT element. When this plasmid (pCCAAT-S) was transfected into HuH-7 cells, it gave rise to 12 { 3% of the amount of secreted surface protein as the parent pSAgDHin, as determined by radioimmunoassay of media from three independent transfections. This figure is similar to the 17% residual activity of the S promoter with the double Sp1 mutation (Fig. 3). Therefore, it is unlikely that there are other elements that contribute significantly to S promoter activity.
the factor(s). At least five different classes of cellular transcription factors bind to CCAAT or related DNA sequences. Their names and consensus binding sites are shown in Table 1. To determine which, if any, of these factors may be binding to the S promoter CCAAT element, we performed competition EMSA, using the HBV CCAAT element as the labeled probe, and the various consensus binding sites as cold competitors. As seen in Fig. 4, a single shifted band was produced upon incubation of the CCAAT element with HuH-7 nuclear extracts, as expected from previous studies (Zhou and Yen, 1991). The c/EBP-binding site did not compete with the CCAAT element (lanes 13 and 14), while the CTF-binding site competed only weakly (lanes 11 and 12). In contrast, the binding sites for NF-Y, YB-1, and CBF (Fig. 4, lanes 5 and 6, 7 and 8, and 9 and 10, respectively) all competed as well as the CCAAT element itself (Fig. 4, lanes 3 and 4). Therefore, the cellular factor that binds to the S promoter CCAAT element is presumably NF-Y, YB-1, or CBF. To distinguish among these factors, we employed rabbit antibodies that could specifically recognize the A subunit of NF-Y (courtesy of B. de Crombrugghe, M. D. Anderson Cancer Center) in the gel shift analysis. As seen in Fig. 5, lane 3, inclusion of these antibodies led to the complete disappearance of the shifted band (SB) and the appearance of a new band with slower migration rate (SS), consistent with a ternary complex of the oligonucleotide, NF-Y, and immunoglobulins (so-called ‘‘supershift’’). This new band is weaker, presumably reflecting the ability of some of the antibodies to compete for DNA binding by NF-Y. In contrast, an antibody against YB-1 (Wright et al., 1994; courtesy of J. Ting, University of North Carolina) did not produce this change (Fig. 5, lane 4), nor did an irrelevant antibody (data not shown). Therefore, the shifted band must have resulted from NFY binding to the CCAAT element. Since no other shifted bands were present, it is unlikely that other cellular factors bound this element. Based on the above data, we conclude that NF-Y is the only HuH-7 cellular factor that binds to the S promoter CCAAT element in vitro. Activation of the S promoter by NF-Y
We had previously shown by foot-print and electrophoretic mobility shift analyses (EMSA) that cellular factor(s) present in HuH-7 cells bind to the S promoter CCAAT element (Zhou and Yen, 1991), but we did not identify
While we have demonstrated that NF-Y is the major factor present in HuH-7 cells that binds to the CCAAT element in vitro, such binding does not necessarily reflect activation in vivo. Unfortunately, NF-Y is expressed at high levels in all eucaryotic cells examined. Thus, we could not look for activation of the S promoter by cotransfected NF-Y. As an alternative, we used the plasmid D4YA13m29 (courtesy of R. Mantovani, Universita degli Studi di Milano), which expresses a dominant negative mutant of the A sub-unit of NF-Y and specifically represses transcription from promoters containing NF-Y binding sites (Jackson et al., 1995; Mantovani et al., 1994). Indeed, cotransfection of this plasmid caused an almost
AID
vira
Binding of CCAAT element by NF-Y
VY 8243
/
6a23$$$122
10-17-96 14:43:46
AP: Virology
ACTIVATION OF HBV S PROMOTER BY NF-Y
391
TABLE 1 Oligonucleotides Used for EMSA Namea
Sequenceb
Referencesc
HBV CCAAT Element CBF c/EBP CTF (NF-1) NF-Y (CBF)a YB-1 (EFIA , dbp)
CTCCACCAATCGGCAGTCAG CTGAGCCAATCAGCAGTCAG ATTGCGCAATCTGCAGTGCA CTCCGAAATTTGGCATCATG CTTAACCAATCAGCAGTCAG CTTGGCCAATCAGCAGTCAG
Zhou and Yen (1991) Lum et al. (1990) Cao et al. (1991) Santoro et al. (1988) Vuorio et al. (1990) Ozer et al. (1990)
a The names used in this paper are given first, and other names commonly used in the literature are given in parentheses. It should be noted that while NF-Y is also called CBF, it bears no relationship at all to the other factors named CBF. b The double-stranded oligonucleotides used consist of two direct repeats of the sequences shown and flanked by EcoRI sites. The sequence differences between the concensus factor-binding sites and the HBV CCAAT element are in italics. c The references cited refer only to papers describing the cDNA clones used in this paper, in cases where multiple groups independently cloned the cDNA.
10-fold reduction of surface protein synthesis from the wild-type pSAgDHin (Fig. 6A). The level of repression is similar to that seen with other promoters with NF-Y sites (Jackson et al., 1995). In contrast, this dominant negative mutant induced no significant decrease in surface protein expression from a plasmid containing a linker-substitution mutation of the CCAAT element (Fig. 6A). These results confirmed that NF-Y is the major factor that transactivates the S promoter via the CCAAT element. In addition, we found that cotransfection of CTF, c/EBP, CBF, or YB-1 expression plasmids had no measurable effect on S promoter activity (data not shown). Thus, none of these other factors can activate the S promoter, in agreement with the lack of binding seen in vitro.
Since the plasmid pSAgDHin used in the above studies contains a subgenomic fragment of HBV DNA, it was possible that cis-elements not present in this fragment may modulate the effect of NF-Y on the S promoter CCAAT element. To rule out this possibility, we also examined the effect of the dominant negative NF-YA protein on surface gene expression from pHBV1.2, a plasmid containing a
FIG. 4. Binding of host cell factors to the HBV CCAAT element, and competition by the concensus sites of various mammalian transcription factors that bind to CCAAT or related sequences, as assessed by EMSA. See Table 1 for the DNA sequences of the various elements. FP, free probe; SB, specifically shifted band.
FIG. 5. Effect of antisera against NF-Y and YB-1 on the binding of host cell factors to the HBV CCAAT element. In the indicated lanes, 1 ml of rabbit antiserum against the A subunit of NF-Y (courtesy of B. de Crombrugghe) or against YB-1 (courtesy of J. Ting) was included in the mixture. FP, free probe; SB, shifted band; SS, ‘‘supershifted’’ band specifically induced by rabbit antiserum against NF-Y.
AID
vira
VY 8243
/
6a23$$$122
10-17-96 14:43:46
AP: Virology
392
LU AND YEN
DISCUSSION
replication-competent, greater-than-full-length clone of the HBV genome (Blum et al., 1992). As seen in Fig. 6B, the mutant NF-YA protein also drastically down-regulated surface gene expression from this plasmid, as measured by surface antigen secretion. As expected, a CCAAT element mutant of pHBV1.2 produced much lower amounts of surface antigen, but was no longer down-regulated by the dominant negative NF-YA (Fig. 6B). There was no significant difference in transfection efficiency between the wildtype and mutant pHBV1.2 plasmids, since the amount of secreted e antigen (a product of the core gene) was comparable (data not shown).
Using a series of plasmids containing linker-substitution mutations in the S promoter of a subgenomic HBV fragment, we demonstrated that the CCAAT element upstream of the start sites is the major positive-acting cis-element of this promoter. In addition, two Sp1 sites together further increased S promoter activity by approximately sixfold. No other activating elements with significant activity were found. However, our studies have not addressed the functions of the elements responsible for correct initiation, which are down-stream of the NF-Y and Sp1 sites and appear to be very complex (Lu and Yen, unpublished data). We have gone on to demonstrate that the major cellular transcription factor that binds in vitro to the S promoter CCAAT element is the heterotrimeric protein, NF-Y (also known as CCAAT-binding factor) (Becker et al., 1991; Maity et al., 1992; Sinha et al., 1995; van Huijsduijnen et al., 1990; Vuorio et al., 1990). In addition, experiments with a dominant negative mutant of NF-Y indicated that NF-Y also activate via this CCAAT element in cultured cells. Since NF-Y is ubiquitously expressed, this finding correlates with the relatively non-cell-type-specific activity of this promoter (Antonucci and Rutter, 1989). The question arises as to why two other cellular factors, YB1 and CBF, neither bind to nor activate via this CCAAT element, since its sequence is extremely similar to the published concensus binding sites for these two factors (Table 1). In fact, concensus binding sites for YB-1 and CBF compete efficiently for NF-Y binding to the HBV CCAAT element (Fig. 5), and bind NF-Y in HuH-7 extracts when used in EMSA experiments (Lu and Yen, published data). Since CBF appears to be a heat shock-induced factor that has specialized functions (Lum et al., 1990), it is not entirely surprising that it is not active in unshocked cells and does not play a role in S promoter activity. However, normal hepatocytes (Spitkovsky et al., 1992) and HuH-7 cells (Lu and Yen, unpublished data) are known to express high levels of YB-1 mRNA. Thus, it is at first glance somewhat surprising that YB-1 apparently neither binds to the CCAAT element, nor influences S promoter activity. Interestingly, however, YB-1 has been shown to bind RNA and single-stranded DNA (Gai et al., 1992; Hasegawa et al., 1991), and one of its homologs in frog oocytes is involved in RNA storage and translational control (Bouvet and Wolffe, 1994). Furthermore, over-expression of YB-1 actually decreases expression from class II major histocompatibility promoters (Ting et al., 1994; Lloberas et al., 1995). Therefore, it is possible that YB-1 may not function as a classical double-stranded DNA-binding transcription factor. Finally, it should be noted that similar CCAAT elements within the albumin, grp78, farnesyl diphosphate synthase, and 3-hydroxy-3methylglutaryl-coenzyme A synthase promoters have all been shown to be specifically bound by NF-Y but not by
AID
vira
FIG. 6. (A) Effect of a dominant negative mutant of NF-Y, A subunit, on surface protein expression from the wild-type pSAgDHin, or a CCAAT element mutant (LS3). Each 100-mm plate of HuH-7 cells was cotransfected with 4 mg of the surface gene plasmid and with either 2 mg of the mutant NF-YA expression plasmid (D4YA13m29) (Jackson et al., 1995; Mantovani et al., 1994) (right half) or 2 mg of pUC18 (left half). The results are expressed as the percentage relative to surface protein secretion from cells cotransfected with wild-type pSAgDHin and pUC18 and represent the mean { standard deviation of 3 independent transfections. (B) Effect of a dominant negative mutant of NF-Y, A subunit, on surface protein expression from the wild-type pHBV1.2, or a CCAAT element mutant (LS3). Each 100-mm plate of HuH-7 cells was cotransfected with 4 mg of the HBV plasmid and with either 2 mg of the mutant NF-YA expression plasmid (right half) or 2 mg of pUC18 (left half). The results are expressed as the percentage relative to surface protein secretion from cells cotransfected with wild-type pHBV1.2 and pUC18 and represent the mean { standard deviation of 3 independent transfections.
VY 8243
/
6a23$$$123
10-17-96 14:43:46
AP: Virology
ACTIVATION OF HBV S PROMOTER BY NF-Y
other CCAAT-binding proteins, including YB-1 (Jackson et al., 1995; Roy and Lee, 1995; Tronche et al., 1991). Thus, our results on the S promoter CCAAT element are entirely consistent with previous studies. In summary, we have shown that the HBV S promoter is relatively compact and is mainly activated by a single CCAAT element that binds NF-Y, with additional activity contributed by two Sp1 sites. Our conclusions are somewhat similar but not identical to those reached by Raney et al. (1991, 1992), in that their data revealed approximately threefold activation of the S promoter by the upstream Sp1 site (site 2 in their nomenclature), and little effect of the down-stream Sp1 site (their site 3). Thus, the relative contribution of these Sp1 sites to S promoter activity is the reverse of what we observe (stronger activity of the down-stream versus the up-stream site). More significantly, Raney et al. (1991, 1992) found only a twoto fourfold activation of this promoter by the CCAAT element, but they found that a third Sp1 site (their site 1), up-stream of our 130-bp minimal S promoter, activated this promoter by approximately fourfold. In contrast, we do not find evidence for any important positive-acting element up-stream of the minimal promoter region (Zhou and Yen, 1991). The reason for these disrepancies is not clear and cannot be attributed to our using the minimal S promoter fragment, since the CCAAT element is also important for S promoter activity in the context of the entire HBV genome that is competent for replication (Fig. 6B). Nor can cell-type difference be the cause, since both groups used HuH-7 cells. Perhaps part of the differences can be attributed to the fact that we used the adw2 subtype of HBV in these experiments, while Raney et al. (1991, 1992) used the ayw sub-type. While the CCAAT element is conserved among all sequenced HBV genomes (Zhou and Yen, 1991), the sequences in the Sp1 sites are not (nucleotides shown in italics in Fig. 1). Therefore, it is possible that some of the Sp1 sites in the ayw sub-type bind Sp1 more strongly than the corresponding sites in adw2. It should also be noted that Raney et al. (1991, 1992) used heterologous reporter gene constructs that give rise to an interfering up-stream ATG codon in some but not all of the S promoter transcripts, that may complicate data analysis based solely on the amount of reporter protein synthesized. Finally, it remains a possibility that the Sp1 sites bind other, as yet unidentified transcription factor(s), in addition to Sp1, and that it is these other factor(s) that are mainly responsible for activating the S promoter. Resolution of these and other questions on S promoter function await further experimentation.
393
on a shared instrument purchased with UCSF Academic Senate Opportunity funds. This work was funded by NIH Grant R01CA55578.
REFERENCES
We thank Z. Cao, L. Guarente, J. Liang, R. Mantovani, S. McKnight, L. Sealy, R. Tjian, B. Wu, and Y. W. Zheng for plasmids and B. deCrombrugghe and J. Ting for antisera. Oligonucleotides were synthesized
Antonucci, T. K., and Rutter, W. J. (1989). Hepatitis B virus (HBV) promoters are regulated by the HBV enhancer in a tissue-specific manner. J. Virol. 63, 579–583. Becker, D. M., Fikes, J. D., and Guarente, L. (1991). A cDNA encoding a human CCAAT-binding protein cloned by functional complementation in yeast. Proc. Natl. Acad. Sci. USA 88, 1968–1972. Blum, H. E., Zhang, Z. S., Galun, E., von, W. F., Garner, B., Liang, T. J., and Wands, J. R. (1992). Hepatitis B virus X protein is not central to the viral life cycle in vitro. J. Virol. 66, 1223–1227. Bouvet, P., and Wolffe, A. P. (1994). A role for transcription and FRGY2 in masking maternal mRNA within Xenopus oocytes. Cell 77, 931– 941. Cao, Z., Umek, R. M., and McKnight, S. L. (1991). Regulated expression of three C/EBP isoforms during adipose conversion of 3T3-L1 cells. Genes Dev. 5, 1538–1552. De-Medina, T., Faktor, O., and Shaul, Y. (1988). The S promoter of hepatitis B virus is regulated by positive and negative elements. Mol. Cell Biol. 8, 2449–2455. Dignam, J. D., Lebovitz, R. M., and Roeder, R. G. (1983). Accurate transcription initiation by RNA polymerase II in a soluble extract from isolated mammalian nuclei. Nucleic Acids Res. 11, 1475–1489. Fried, M., and Crothers, D. (1981). Equilibiria and kinetics of lac repressor-operator interactions by polyacrylamide gel electrophoresis. Nucleic Acids Res. 9, 6505–6525. Gai, X. X., Lipson, K. E., and Prystowsky, M. B. (1992). Unusual DNA binding characteristics of an in vitro translation product of the CCAAT binding protein mYB-1. Nucleic Acids Res. 20, 601–606. Ganem, D. (1991). Assembly of hepadnaviral virions and subviral particles. Curr. Top. Microbiol. Immunol. 168, 61–83. Ganem, D. (1996). Hepadnaviridae and their replication. In ‘‘Virology’’ (B. N. Fields and D. M. Knipe, Eds.), pp. 2703–2737. Raven Press, New York. Gorman, C. (1985). High efficiency gene transfer into mammalian cells. In ‘‘DNA Cloning’’ (D. M. Glover, Ed.), pp. 143–190. IRL, Oxford. Hasegawa, S. L., Doetsch, P. W., Hamilton, K. K., Martin, A. M., Okenquist, S. A., Lenz, J., and Boss, J. M. (1991). DNA binding properties of YB-1 and dbpA: Binding to double-stranded, single-stranded, and abasic site containing DNAs. Nucleic Acids Res. 19, 4915–4920. Hollinger, F. B. (1990). Hepatitis B Virus. In ‘‘Virology’’ (B. N. Fields and D. M. Knipe, Eds.), pp. 2171–2238. Raven Press, New York. Huang, Z. M., and Yen, T. S. B. (1993). Dys-regulated surface gene expression from disrupted hepatitis B virus genomes. J. Virol. 67, 7032– 7040. Jackson, S. M., Ericsson, J., Osborne, T. F., and Edwards, P. A. (1995). NF-Y has a novel role in sterol-dependent transcription of two cholesterogenic genes. J. Biol. Chem. 270, 21445–21448. Lloberas, J., Maki, R. A., and Celada, A. (1995). Repression of major histocompatibility complex I-A beta gene expression by dbpA and dbpB (mYB-1) proteins. Mol. Cell Biol. 15, 5092–5099. Lum, L. S., Sultzman, L. A., Kaufman, R. J., Linzer, D. I., and Wu, B. J. (1990). A cloned human CCAAT-box-binding factor stimulates transcription from the human hsp70 promoter. Mol. Cell Biol. 10, 6709– 6717. Maity, S. N., Sinha, S., Ruteshouser, E. C., and de Crombrugghe, B. (1992). Three different polypeptides are necessary for DNA binding of the mammalian heteromeric CCAAT binding factor. J. Biol. Chem. 267, 16574–16580. Mantovani, R., Li, X. Y., Pessara, U., van, H. R., Benoist, C., and Mathis, D. (1994). Dominant negative analogs of NF-YA. J. Biol. Chem. 269, 20340–20346. Ozer, J., Faber, M., Chalkley, R., and Sealy, L. (1990). Isolation and characterization of a cDNA clone for the CCAAT transcription factor
AID
vira
ACKNOWLEDGMENTS
VY 8243
/
6a23$$$123
10-17-96 14:43:46
AP: Virology
394
LU AND YEN
EFIA reveals a novel structural motif. J. Biol. Chem. 265, 22143– 22152. Raney, A. K., Le, H. B., and McLachlan, A. (1992). Regulation of transcription from the hepatitis B virus major surface antigen promoter by the Sp1 transcription factor. J. Virol. 66, 6912–6921. Raney, A. K., Milich, D. R., and McLachlan, A. (1991). Complex regulation of transcription from the hepatitis B virus major surface antigen promoter in human hepatoma cell lines. J. Virol. 65, 4805–4011. Roy, B., and Lee, A. S. (1995). Transduction of calcium stress through interaction of the human transcription factor CBF with the proximal CCAAT regulatory element of the grp78/BiP promoter. Mol. Cell Biol. 15, 2263–2274. Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989). ‘‘Molecular Cloning: A Laboratory Manual,’’ 2nd ed., Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. Santoro, C., Mermod, N., Andrews, P. C., and Tjian, R. (1988). A family of human CCAAT-box-binding proteins active in transcription and DNA replication: Cloning and expression of multiple cDNAs. Nature 334, 218–224. Selden, R. F., Howie, K. B., Rowe, M. E., Goodman, H. M., and Moore, D. D. (1986). Human growth hormone as a reporter gene in regulation studies employing transient gene expression. Mol. Cell Biol. 6, 3173– 3179. Sinha, S., Maity, S. N., Lu, J., and de Crombrugghe, B. (1995). Recombinant rat CBF-C, the third subunit of CBF/NFY, allows formation of a protein-DNA complex with CBF-A and CBF-B and with yeast HAP2 and HAP3. Proc. Natl. Acad. Sci. USA 92, 1624–1628. Spitkovsky, D. D., Royer, P. B., Delius, H., Kisseljov, F., Jenkins, N. A., Gilbert, D. J., Copeland, N. G., and Royer, H. D. (1992). Tissue restricted expression and chromosomal localization of the YB-1 gene encoding a 42 kD nuclear CCAAT binding protein. Nucleic Acids Res. 20, 797–803. Ting, J. P., Painter, A., Zeleznik, L. N., MacDonald, G., Moore, T. M.,
Brown, A., and Schwartz, B. D. (1994). YB-1 DNA-binding protein represses interferon gamma activation of class II major histocompatibility complex genes. J. Exp. Med. 179, 1605–1611. Tronche, F., Rollier, A., Sourdive, D., Cereghini, S., and Yaniv, M. (1991). NFY or a related CCAAT binding factor can be replaced by other transcriptional activators for cooperation with HNF1 in driving the rat albumin promoter in vivo. J. Mol. Biol. 222, 31–43. Valenzuela, P., Quiroga, M., Zaldivar, J., Gray, P., and Rutter, W. (1980). The nucleotide sequence of the hepatitis B genome and the identification of the major viral genes. In ‘‘Animal Virus Genetics’’ (B. Fields, R. Jaenisch, and C. Fox, Eds.), pp. 57–70. Academic Press, New York. van Huijsduijnen, R., Li, X. Y., Black, D., Matthes, H., Benoist, C., and Mathis, D. (1990). Co-evolution from yeast to mouse: cDNA cloning of the two NF-Y (CP-1/CBF) subunits. EMBO J. 9, 3119–3127. Vuorio, T., Maity, S. N., and de Crombrugghe, B. (1990). Purification and molecular cloning of the ‘‘A’’ chain of a rat heteromeric CCAAT-binding protein. Sequence identity with the yeast HAP3 transcription factor. J. Biol. Chem. 265, 22480–22486. Wright, K. L., Vilen, B. J., Itoh, L. Y., Moore, T. L., Li, G., Criscitiello, M., Cogswell, P., Clarke, J. B., and Ting, J. P. (1994). CCAAT box binding protein NF-Y facilitates in vivo recruitment of upstream DNA binding transcription factors. EMBO J. 13, 4042–4053. Yen, T. S. B. (1993). Regulation of hepatitis B virus gene expression. Semin. Virol. 4, 33–42. Zheng, Y. W., and Yen, T. S. B. (1994). Negative regulation of hepatitis B virus gene expression and replication by oxidative stress. J. Biol. Chem. 269, 8857–8862. Zhou, D. X., and Yen, T. S. B. (1990). Differential regulation of the hepatitis B viral surface gene promoters by a second viral enhancer. J. Biol. Chem. 265, 20731–20734. Zhou, D. X., and Yen, T. S. B. (1991). The hepatitis B virus S promoter comprises a CCAAT motif and two initiation regions. J. Biol. Chem. 266, 23416–23421.
AID
vira
VY 8243
/
6a23$$$123
10-17-96 14:43:46
AP: Virology