JOURNAL OF VIROLOGY, Oct. 2003, p. 10975–10983 0022-538X/03/$08.00⫹0 DOI: 10.1128/JVI.77.20.10975–10983.2003 Copyright © 2003, American Society for Microbiology. All Rights Reserved.
Vol. 77, No. 20
Induction of the Upstream Regulatory Region of Human Papillomavirus Type 31 by Dexamethasone Is Differentiation Dependent Jennifer L. Bromberg-White,† Ellora Sen, Samina Alam, Jason M. Bodily, and Craig Meyers* Department of Microbiology and Immunology, The Pennsylvania State University College of Medicine, Hershey, Pennsylvania 17033 Received 2 December 2002/Accepted 10 July 2003
Glucocorticoids have been shown to play a role in the transforming abilities of human papillomaviruses (HPVs), and glucocorticoid response elements (GREs) have been identified in the upstream regulatory regions (URRs) of various HPV types. These findings have made glucocorticoids potential therapeutic targets for HPV infection. We have previously shown that the URR of HPV type 31 (HPV31) is insensitive to induction by the synthetic glucocorticoid dexamethasone (dex) in monolayer culture, despite the identification of three potential GREs in the 5ⴕ region of the URR. Due to the fact that the HPV life cycle is intimately linked to the differentiation of the host tissue, we chose to determine whether the URR of HPV31 was inducible by dex under differentiating conditions. Upon suspension of cells in a semisolid medium of methylcellulose, we found that the URR of HPV31 was inducible by dex. The three GREs appear to play roles as independent repressors of this inducibility. By 5ⴕ deletion analysis, the element(s) responsible for this induction was localized to nucleotides (nt) 7238 to 7557. Furthermore, we found that the region between nt 7883 and 7900 appears to act as a repressor of dex inducibility. These findings indicate that epithelial differentiation has a profound effect on the action of dex on the URR of HPV31, suggesting that glucocorticoids play an important role in the differentiation-dependent life cycle of HPV. Human papillomavirus (HPV) comprises a group of over 100 types of small, double-stranded DNA viruses that infect differentiating squamous epithelium. HPV infection is thought to occur primarily through microabrasions in the epithelium, upon which HPV DNA is maintained in the basal cells of the epithelium as extrachromosomal episomes. Viral early-gene expression is controlled by promoter and enhancer elements located in the upstream regulatory region (URR) of the genome (21). Early viral genes E1 and E2 regulate viral DNA replication, which occurs along with cellular DNA replication in the mitotically active basal cells (27, 45). Expression of the late viral genes occurs as HPV-infected cells migrate up through the epithelium and undergo a complex program of differentiation (7, 17, 35). The life cycle of HPV is tightly linked to the differentiation of the epithelium it infects (41). In undifferentiated keratinocytes, viral DNA is maintained at low copy numbers and viral gene expression is limited (9). Upon epithelial differentiation, viral DNA is amplified to over 1,000 copies per cell and an increase in viral gene expression is observed. While the majority of early viral genes are expressed in both undifferentiated and differentiated epithelial cells, expression of the late genes occurs only in the terminally differentiated layers of the epithelium (23). These differentiation-dependent events occur in
* Corresponding author. Mailing address: Department of Microbiology and Immunology, The Pennsylvania State University College of Medicine, Hershey, PA 17033. Phone: (717) 531-6240. Fax: (717) 5314600. E-mail:
[email protected]. † Present address: Laboratory of Tumor Metastasis and Angiogenesis, The Van Andel Research Institute, Grand Rapids, MI 49503.
preparation for virion morphogenesis. Late-gene expression is controlled by a differentiation-dependent promoter; in HPV31, it is p742 (22), and in HPV16, it is p670 (20). Both p742 promoter activity and late-gene expression have been shown to be upregulated in differentiating organotypic (raft) cultures compared with what occurs in monolayers of the CIN612-9E (9E) cell line (35, 36). This cell line is a clonal isolate of a cervical intraepithelial neoplasia grade I lesion that maintains episomal copies of HPV type 31b (HPV31b) (7) and has been shown to produce infectious virus in raft cultures (32). Cellular transcription factors have been shown to modulate the activity of the early viral promoter by binding to cis elements in the URR. Changes in the expressions of viral genes upon epithelial differentiation can be explained in part by differences in the expressions and activities of these cellular factors. Numerous studies have demonstrated that changes in viral gene expression upon differentiation are strongly influenced by the levels and activities of cellular transcription factors and that this influence is mediated by the binding, or lack thereof, to the viral URR (1). The dependence of a productive HPV infection on epithelial differentiation has caused difficulties in the study of the complete HPV life cycle in vitro. Advances in the organotypic (raft) culture system have allowed for the growth of differentiating epithelium in vitro and have made it possible to provide all of the factors necessary for the complete viral life cycle (10, 32, 33, 35). However, preparation of raft cultures can be time consuming, requiring up to 12 days in culture prior to analysis of late viral events, which can limit the rapid analysis of multiple constructs. Suspension of epithelial cells in a semisolid medium of methylcellulose has been shown to allow for the
10975
10976
BROMBERG-WHITE ET AL.
rapid induction of differentiation in only 48 h (16, 24). HPV31positive epithelial cells have been shown to express the differentiation markers involucrin and transglutaminase after suspension in methylcellulose as well as to express transcripts from p742, although no capsid proteins have been detected in this system (38). Therefore, suspension in methylcellulose is a convenient system for use in a quick analysis of the effect of differentiation on a variety of life cycle events, even though it cannot facilitate the final steps of the viral life cycle. In addition to the other transcription factors mentioned, glucocorticoids have been shown to affect viral early-gene expression and promoter activity by binding to glucocorticoid response elements (GREs) in the URR (12, 14, 19, 30). Bromberg-White and Meyers have previously shown that the URR of HPV31 is not inducible by the synthetic glucocorticoid dexamethasone (dex) in monolayer culture (10). In the present study, we demonstrate that suspension of the HPV31b-positive 9E cell line in the semisolid medium methylcellulose results in a dex-inducible activation of a luciferase promoter construct driven by the URR of HPV31 that was not seen in monolayer culture. The three GREs we identified by sequence analysis and their ability to bind the glucocorticoid receptor (GR) in the URR of HPV31 do not appear to be responsible for this inducibility. However, loss of all three GREs resulted in an increase in the induction, suggesting that each of the three GREs acts as a repressor of glucocorticoid action. By 5⬘ deletion analysis, we localized the element(s) responsible for the dex inducibility of the URR of HPV31 to nucleotides (nt) 7238 to 7557. Furthermore, replacement of wild-type sequences between nt 7883 and 7900 by linker scanning mutagenesis resulted in an increased dex inducibility. Taken together, these results show that differentiation has a profound effect on the action of dex on the URR of HPV31, suggesting that glucocorticoids may play an important role in the differentiation-dependent life cycle of HPV. MATERIALS AND METHODS Cell culture. The 9E cell line is a clonal isolate derived from a cervical intraepithelial neoplasia grade I lesion that maintains 50 episomal copies of HPV31b per cell (7). Primary human foreskin keratinocyte (HFK) cultures were derived from newborn foreskin via trypsin digestion at 37°C (38). 9E cells were grown in E medium in the presence of mitomycin-C-treated J2 fibroblasts as previously described (7, 31, 32). HFK cells were maintained in KGM-2 (BioWhittaker, Walkersville, Md.). Deficient E medium was used to culture 9E cells for transfection experiments as described previously (10). Plasmid constructs. The full-length URR of HPV31 or HPV18 was cloned into the pGL2-Basic luciferase reporter vector as described previously (10, 40). All other constructs were created as described previously (10, 40). Transfection and luciferase assay. For monolayer studies, 9E cells were transfected as previously described (10). HFK cells were seeded at a density of 2 ⫻ 105 cells per 35-mm-diameter dish in KGM-2. Twelve hours later, cells were transfected by using Lipofectamine PLUS (Invitrogen-Life Technologies, Carlsbad, Calif.). Each well was transfected in KGM-2 with 1 g of experimental construct, 6 l of PLUS reagent, and 4 l of Lipofectamine for 3 h. The transfection solution was replaced with fresh KGM-2 that contained either 0.1% ethyl alcohol (vehicle) or 10⫺6 M dex. After transfection, monolayer cells were incubated for 48 h at 37°C in an atmosphere of 5% CO2. For differentiation studies, 9E or HFK cells were seeded at a density of 8 ⫻ 105 cells per 100-mm-diameter dish. 9E cells were first seeded in deficient E medium (which contains no hydrocortisone or phenol red and is supplemented with dextran-coated, charcoal-stripped serum) for 3 h, at which point the medium was switched to KGM-2. HFK cells were seeded and maintained in KGM-2 for the duration of the transfection. At 12 h (9E) or 16 h (HFK) postseeding, cells were transfected in KGM-2 with 4 g of experimental construct, 20 l of PLUS reagent, and 30 l of Lipofectamine per
J. VIROL. 100-mm-diameter dish. After a 3-h incubation, cells were trypsinized, pelleted, and resuspended in 0.5 ml of deficient E medium. Cells were then suspended in 15 ml of 1.6% methylcellulose and treated with vehicle or 10⫺6 M dex. Methylcellulose was prepared exactly as described previously, using Deficient E medium (38, 40). After a 48-h incubation, cells in methylcellulose were recovered by three rounds of washes in phosphate-buffered saline. Luciferase lysates of all transfections were prepared using passive lysis buffer (Promega Corp., Madison, Wis.). Luciferase assays were performed by using the Luciferase Assay System (Promega) with a Turner Designs TD 20/20 luminometer as recommended by the manufacturer. For methylcellulose experiments, luciferase lysates were normalized to total protein by the Bradford assay by using a protein assay dye reagent (Bio-Rad Laboratories, Hercules, Calif.). Nuclear-extract preparation and EMSA. Nuclear extracts from 9E cells were prepared as described previously (26). Briefly, 2 ⫻ 106 cells were suspended in 400 l of buffer A (10 mM HEPES [pH 7.9], 10 mM KCl, 0.1 mM EDTA, 0.1 mM EGTA, 1 mM dithiothreitol, 0.5 mM phenylmethylsulfonyl fluoride). The cells were allowed to swell on ice for 20 min, after which 25 l of IGEPAL CA-630 (Sigma) was added, followed by vigorous mixing for 20 s. Following centrifugation for 30 s, the pellet was recovered and resuspended in buffer B (20 mM HEPES [pH 7.9], 400 mM KCl, 1 mM EDTA, 0.5 mM phenylmethylsulfonyl fluoride), shaken at 4°C for 30 min, and centrifuged for 5 min at 4°C. The supernatants were recovered, and protein concentration was measured by using a BC protein assay kit (Pierce, Rockford, Ill.). Blunt-ended, double-stranded oligonucleotides were end labeled using [␥32-P]ATP and T4 polynucleotide kinase as previously described (10). Competition electrophoretic mobility shift assays (EMSA) were performed by incubating 10 g of nuclear extract with 2 g of poly(dI-dC) (Pharmacia, Kalamazoo, Mich.) in the presence or absence of a 100-fold excess of unlabeled homologous oligonucleotide or oligonucleotide corresponding to the consensus GRE (GREc) sequence for 20 min at room temperature in a 20-l reaction volume containing 5% glycerol, 150 mM KCl, 0.5 mM EDTA, and 0.5 mM dithiothreitol. Following incubation, 0.75 ng of 32Plabeled oligonucleotide probe was added to the reaction, followed by incubation at room temperature for an additional 30 min. The DNA-protein complexes were separated from free probe by electrophoresis on a 6% Tris-glycine-EDTA polyacrylamide gel (acrylamide to bisacrylamide, 37.5:1) at 200 V at 4°C. Gels were dried at 80°C for 2 h and subjected to autoradiography.
RESULTS HPV31 URR is inducible by dex in methylcellulose. It has previously been shown that the URR of HPV31 was insensitive to induction by dex in monolayer 9E culture (10) compared with the URR of HPV18, which has been shown previously to contain a functional GRE (12, 14, 30). Because the life cycle of HPV is intimately linked to the differentiation of the epithelium it infects, we were interested in determining whether the URR of HPV31 was inducible by dex in a differentiating system. Therefore, we chose to analyze the induction of transcriptional activity of the URR of HPV31 by dex in a semisolid medium, methylcellulose. Suspension in methylcellulose provides a simplified system for differentiation of epithelial cells in a shorter time frame than that with the organotypic (raft) culture system (16, 24). While the growth of raft cultures is necessary in order to obtain the complete viral life cycle, including infectious-virus production, cells suspended in methylcellulose are able to undergo a variety of differentiation-dependent late events of the viral life cycle, including viral-genome amplification and late-gene transcript expression (38). To determine if cells suspended in methylcellulose are in fact undergoing differentiation, differentiation-specific cell markers are commonly analyzed (15, 32, 38, 40). We have previously shown that the 9E cells are able to differentiate when suspended in methylcellulose, as evidenced by an increase in the expression of the differentiation markers K10 and involucrin (40). We wanted to determine whether suspension of these cells in methylcellulose would alter the dex inducibility of the wild-
VOL. 77, 2003
HPV31 URR DIFFERENTIATION-DEPENDENT DEX INDUCTION
10977
FIG. 1. (A) Analysis of the luciferase activity of the wild-type HPV18 URR (pGL2-18URR), a GRE LSM of HPV18 (p18LSMGRE), a mutant containing a 1-base change in the GRE of HPV18 (18URR-C7843A), the wild-type HPV31 URR (pGL2-31URR), or a single, double, or triple GRE LSM of HPV31 (p31LSMGRE) after suspension in methylcellulose and treatment with 10⫺6 M dex. Relative luciferase activity is expressed as a change in activity (n-fold) over that of pGL2-31URR under vehicle conditions (0.1% ethyl alcohol), which was set to 1. Luciferase values were normalized to total protein, and all experiments were performed in triplicate. Error bars indicate standard deviations (SD). (B) Sequence comparison of the GREc to the GRE of HPV18 and the three GREs of HPV31. Included for comparison is the linker scanning mutagenesis (LSM) sequence.
type HPV31 URR construct, pGL2-31URR. It has previously been shown that this construct is insensitive to glucocorticoid induction in monolayer culture (10). Following transfection and suspension in methylcellulose, analysis of luciferase activity showed that the URR of HPV31 was induced approximately threefold after treatment with dex compared with what was seen with vehicle treatment (Fig. 1). Surprisingly, loss of any of the GREs by linker scanning mutagenesis had little effect on this induction, with the exception of the triple linker scanning mutant (LSM) p31LSMGRE1,2,3, which showed a marked increase (eightfold) in dex induction compared with that of the wild-type construct (Fig. 1A). These data suggest that each of the three GREs acts as a repressor of dex inducibility of the HPV31 URR. As previous studies have shown (10, 12, 30), the URR of HPV18 is inducible by dex in monolayer 9E culture. Therefore, we analyzed the effect of suspension in methylcellulose on this inducibility. As shown in Fig. 1A, the wild-type URR of HPV18, pGL2-18URR, was inducible by dex in methylcellulose, as is the GRE LSM p18LSMGRE and the subtle GRE mutant 18URR-C7843A, although the dex inducibility of these GRE mutants in monolayer culture was reduced by 60% (10). These data suggest that the URR of HPV18 contains another GRE and that differentiation may result in stronger induction by this element. An element within nt 7238 to 7557 is responsible for dex inducibility of the HPV31 URR. As shown in Fig. 1, the URR of HPV31 is inducible by dex in methylcellulose suspension. Surprisingly, none of the three GREs we identified appear to be responsible for this induction. Therefore, to localize the region conferring dex inducibility, we utilized 5⬘
deletions of the HPV31 URR whereby the URR of HPV31 was 5⬘ truncated to nt 7238 (p31URR⌬6921-7238), nt 7557 (p31URR⌬6921-7557), and nt 7799 (p31URR⌬6921-7799) (Fig. 2A). After transfection of these constructs into 9E cells and suspension in methylcellulose, we found that the induction of activity of the URR of HPV31 by dex was lost with construct p31URR⌬6921-7557 (Fig. 2B). This indicates that the element(s) responsible for the dex induction seen with the fulllength URR either lies between nt 7238 and 7557 or spans the nt 7557 deletion junction. The locations of GRE2 and GRE3 in this region (Fig. 2A) point to either or both of these sites as the explanation for the loss of dex inducibility. However, linker scanning mutagenesis demonstrated that loss of either or both of these GREs had no effect on the dex induction within the context of the whole URR (Fig. 1), suggesting that another cis regulatory element in this region is responsible. Other cis elements in this region include two AP1 sites and E2 binding site no. 1 (E2BS1) (26, 38). As AP1 is a known antagonist of GR action (25, 39, 47) and as there is no evidence that E2 or E2 binding sites play a role in glucocorticoid action, it seemed plausible to believe that an unidentified site in this region is responsible for dex inducibility of the URR of HPV31 under differentiating conditions. To determine what element or elements were responsible for differentiation-dependent dex induction in methylcellulose within nt 7238 to 7557, LSMs encompassing this region were utilized. Sen et al. have previously divided the URR of HPV31 into four regions (named A, B, C, and D), with each region containing 14 potential LSMs (1 through 14) (40). Region A spans nt 7008 to 7259, region B spans nt 7260 to 7511, region C spans nt 7512 to 7763, and region D spans nt 7764 to 103.
10978
BROMBERG-WHITE ET AL.
J. VIROL.
FIG. 2. The dex inducibility of 5⬘ deletion constructs of the URR of HPV31. (A) Diagram of the URR of HPV31, with GREs, E2 binding sites (E2BS), and the nucleotides used for the creation of 5⬘ deletion constructs indicated. (B) Wild-type HPV31 URR (pGL2-31URR) or 5⬘ deletion constructs. The nucleotides deleted from each construct are indicated by deltas. 9E cells were transfected with each construct, suspended in methylcellulose, and treated with 10⫺6 M dex. Relative luciferase activity is expressed as a change in activity (n-fold) over that of pGL2-31URR under vehicle conditions, which was set to 1. Luciferase values were normalized to total protein, and all experiments were performed in triplicate. Error bars indicate SD.
LSMs A14 (nt 7242 to 7259) through C3 (nt 7548 to 7565) were transfected into 9E cells and then suspended in methylcellulose to determine what specific region of nt 7238 to 7557 was responsible for dex induction. As shown in Fig. 3, we found that mutants A14 (nt 7242 to 7259) and B1 (nt 7260 to 7278) resulted in a loss of dex inducibility, suggesting that the element responsible is contained within this region. Surprisingly,
GRE2, which was previously identified by sequence analysis, overlaps this region of the URR (Fig. 2A) (10). However, linker scanning mutagenesis of this GRE did not result in a loss of dex inducibility upon suspension in methylcellulose (Fig. 1), indicating that other elements in this region are responsible. Interestingly, we also found that mutants B5 (nt 7332 to 7349), B10 (nt 7458 to 7475), and C3 (nt 7548 to 7565) resulted in a
FIG. 3. The dex inducibility of LSMs for the region between nt 7238 and 7557. Numbering of mutants has been described previously (40). All constructs were transfected into 9E cells, suspended in methylcellulose, and treated with 10⫺6 M dex. Relative luciferase activity is expressed as a change in activity (n-fold) over that of pGL2-31URR under vehicle conditions, which was set to 1. Asterisks over mutants A14 and B1 denote no change between vehicle and dex treatments. Luciferase values were normalized to total protein, and all experiments were performed in triplicate. Error bars indicate SD.
VOL. 77, 2003
HPV31 URR DIFFERENTIATION-DEPENDENT DEX INDUCTION
10979
FIG. 4. The dex inducibility of various HPV31 URR constructs: pGL2-31URR, wild-type HPV31 URR construct; 31URR-3⬘LSM, nt 7883 to 7900 removed by linker scanning mutagenesis; p31LSMGRE1,2,3-3⬘LSM, linker scanning mutagenesis of all three HPV31 GREs in the region with nt 7883 to 7900 removed by linker scanning mutagenesis; 31URR-3⬘LSM(18GRE), LSM of 31URR-3⬘LSM replaced with the GRE of HPV18; p31LSMGRE1,2,3-3⬘LSM(18GRE), linker scanning mutagenesis of the three HPV31 GREs in the 5⬘ region of the URR in the context of p31-3⬘LSM(18GRE). All constructs were transfected into 9E cells, suspended in methylcellulose, and treated with 10⫺6 M dex. Relative luciferase activity is expressed as a change in activity (n-fold) over that of pGL2-31URR under vehicle conditions, which was set to 1. Luciferase values were normalized to total protein, and all experiments were performed in triplicate. Error bars indicate SD.
strong upregulation of dex induction of the URR of HPV31 (Fig. 3), suggesting the presence of repressors of dex induction in this region. Region between nt 7883 and 7900 acts as a repressor of dex action. We had previously constructed an LSM in a 3⬘ location of the URR of HPV31 (nt 7883 to 7900) that corresponds to the location of the HPV18 GRE in the HPV18 URR (10). This construct, 31URR-3⬘LSM, was created in part to understand why the URR of HPV31 was not inducible by dex in 9E monolayer cultures. It has previously been shown that the basal activity and dex inducibility of 31URR-3⬘LSM were identical to those of the wild-type pGL2-31URR construct in 9E monolayer cultures (10). However, we discovered that the induction of the URR of HPV31 by dex in methylcellulose was further upregulated by this linker scanning mutation over that of the wild-type construct, suggesting that the region between nt 7883 and 7900 is acting as a repressor of dex function in a differentiating system (Fig. 4). Surprisingly, replacement of this region with the GRE of HPV18 resulted in a loss of dex induction (Fig. 4), even though this region in the URR of HPV31 corresponds to the location of the HPV18 GRE in the URR of HPV18. These data indicate that the GRE of HPV18 acts as a repressor in the URR of HPV31 in methylcellulose when placed in the URR of HPV31 in a location corresponding to its location in HPV18 URR. Loss of all three GREs in the 5⬘ region of the URR of HPV31 by linker scanning mutagenesis did not result in a recovery of dex inducibility (Fig. 4), indicating that these three sites were not acting as repressors of possible dex action by the HPV18 GRE in the 3⬘ region of the HPV31 URR. HPV31 URR is not dex inducible in HFK cells. All experiments to this point have been performed with the 9E cell line, which contains episomally maintained HPV31b. During epithelial differentiation, viral genes, including E1 and E2 (36) and the late genes L1 and L2 (23), are upregulated. Therefore, it is possible that differentiation-dependent glucocorticoid inducibility of the URR of HPV31 is due to changes in viral gene expression. To determine whether viral gene products play a role in the differentiation-dependent dex inducibility of the URR of HPV31, we next performed transfection experiments
using primary HFK cells derived from newborn foreskin. In monolayer culture, as a positive control, we found that the URR of HPV18 was inducible by dex treatment by a factor of fourfold, with a 75% loss in this induction upon loss of the characterized GRE (Fig. 5A) (12, 14). Neither the wild-type URR of HPV31 nor the triple GRE LSM construct were affected by dex treatment. These results are identical to what has previously been seen in 9E monolayer experiments (10).
FIG. 5. The dex inducibilities of the wild-type HPV31 URR (pGL2-31URR), the triple GRE LSM of HPV31 (p31LSMGRE1,2,3), the wild-type HPV18 URR (pGL2-18URR), and the GRE LSM of HPV18 (p18LSMGRE) in HFK monolayer cultures (A) or methylcellulose cultures (B). HFK cells were transfected with each construct and treated with 10⫺6 M dex. Relative luciferase activity is expressed as a change in activity (n-fold) over that of pGL2-31URR under vehicle conditions, which was set to 1. Luciferase values were normalized to total protein, and all experiments were performed in triplicate. Error bars indicate SD.
10980
BROMBERG-WHITE ET AL.
FIG. 6. Comparison of the basal activities of the URRs of HPV31 and HPV18 in monolayer and methylcellulose culture conditions in 9E and HFK cells. Wild-type HPV31 URR (pGL2-31URR) or HPV18 URR (pGL2-18URR) was transfected into 9E or HFK monolayer (white columns) or methylcellulose (black columns) cultures and assayed for luciferase activity. Methylcellulose luciferase values were normalized to total protein. Relative luciferase activity is expressed as a change in activity (n-fold) over that of pGL2-31URR in 9E monolayer cultures, which was set to 1. All experiments were performed in triplicate, and error bars indicate SD.
Upon suspension of transfected cell in methylcellulose, we found that the URR of HPV18 was only weakly inducible by dex and that the URR of HPV31 was not inducible at all (Fig. 5B). This is in direct contrast to what was seen with 9E cells, where the URR of HPV18 and the URR of HPV31 were inducible by dex (Fig. 1 and 2). Furthermore, we also noted that the triple LSM GRE construct of HPV31 and the LSM GRE construct of HPV18 were not strongly inducible by dex in HFK methylcellulose cultures (Fig. 5B). These differences between 9E and HFK cells suggest that one or more viral factors plays a role in the differentiationdependent inducibility of the URR of HPV31 and HPV18. To determine whether the URRs of HPV31 and HPV18 behave differently in the different cell types, we compared the basal activities of the wild-type HPV31 URR (pGL2-31URR) and HPV18 URR (pGL2-18URR) constructs in monolayer and methylcellulose cultures (Fig. 6). We found that the basal activities of the URR of HPV18 were nearly identical between monolayer and methylcellulose conditions in the two cell lines, suggesting that differences in dex inducibility of the URR of HPV18 is not due to basal activity differences between culturing conditions. Furthermore, we found that while the basal activities of the URR of HPV31 were similar in monolayer culture between the two cell lines, the basal activity was more than twofold higher in HFK cells than in 9E cells in methylcellulose cultures (Fig. 6). These data suggest that the lack of dex induction of the URR of HPV31 in HFK methylcellulose cultures may not be due to a lack of viral factors but may be due to the inability of dex to induce HPV31 URR activity, as this activity is already induced by an unknown factor(s). Competition EMSA analysis indicates changes in DNA-protein interactions in undifferentiated and differentiated conditions. We had previously discovered that regions A14 and B1, which are contained within nt 7238 to 7557, appeared to be responsible for dex inducibility of the URR of HPV31 in differentiating cultures (Fig. 3). Although GRE2 overlaps these
J. VIROL.
regions, we determined by linker scanning mutagenesis that GRE2 is not responsible for this differentiation-dependent dex inducibility (Fig. 1). Analysis of the sequences of these regions identified two potential GRE half sites compared to the GREc (Fig. 7A): one residing in region A14 and one residing in region B1 (Fig. 7B). Therefore, we chose to analyze the binding abilities of regions A14 and B1, as well as of B2, B3, B4, and B5, which are all contained within the region we identified as having GRE-induced activity, in both undifferentiated (monolayer) conditions and differentiated (methylcellulose) conditions. Each region was used as a probe in binding reactions with 9E nuclear extracts from both monolayer and methylcellulose cultures. Furthermore, we analyzed the specificity of binding utilizing both the homologous oligonucleotide as well as the GREc oligonucleotide as cold competitors. As shown in Fig. 8, we saw formation of a complex bound to all regions in undifferentiated conditions (Fig. 8, lanes 2, 6, 10, 14, 18, and 22). While the homologous cold competitor was able to compete away this complex for all regions (Fig. 8, lanes 3, 7, 11, 15, 19, and 23), GREc as cold competitor was unable to clearly compete away the complex and had no effect on complex formation with region A14 (Fig. 8, lanes 4, 8, 12, 16, 20, and 24). These data suggest that the GR does not directly bind these regions of the URR of HPV31 but may cooperatively bind with other factors. In differentiating conditions, complexes formed with regions A14 and B3 were not affected by the consensus cold competitor, GREc (Fig. 8, lanes 28 and 40).
FIG. 7. Analysis of the dexamethasone-inducible region of the URR of HPV31. Shown are the GREc sequence, which consists of 6-bp imperfect palindromes separated by 3 nt (A); nt 7242 to 7277, consisting of LSMs A14 and B1, each of which is necessary for the differentiation-dependent induction of the URR of HPV31 (also indicated is GRE2, which overlaps A14 and B1) (B); and the proposed model for dex inducibility by A14 and B1 GRE half sites (C).
VOL. 77, 2003
HPV31 URR DIFFERENTIATION-DEPENDENT DEX INDUCTION
10981
FIG. 8. Competition EMSA of regions A14 to B5 (nt 7242 to 7349) with 9E nuclear extracts in undifferentiated (monolayer) and differentiated (methylcellulose) conditions. Undifferentiated: 32P-end-labeled probe (A14, B1, B2, B3, B4, or B5) was incubated without protein (lanes 1, 5, 9, 13, 17, and 21), with 9E nuclear extracts from monolayer cultures (lanes 2, 6, 10, 14, 18, and 22), with monolayer extracts plus homologous cold competitor (lanes 3, 7, 11, 15, 19, and 23), or with monolayer extracts plus GREc as cold competitor (lanes 4, 8, 12, 16, 20, and 24). Differentiated: 32 P-end-labeled probe (A14, B1, B2, B3, B4, or B5) was incubated without protein (lanes 25, 29, 33, 37, 41, and 45), with 9E nuclear extracts from methylcellulose cultures (lanes 26, 30, 34, 38, 42, and 46), with methylcellulose extracts plus homologous cold competitor (lanes 27, 31, 35, 39, 43, and 47), or with methylcellulose extracts plus GREc as cold competitor (lanes 28, 32, 36, 40, 44, and 48). C, bound complex formation; NS, nonspecific complex.
Furthermore, we found that the complexes formed with regions B4 and B5 were completely blocked in differentiated conditions (Fig. 8, compare lanes 18 and 42 for B4 and lanes 22 and 46 for B5). Taken together, these data indicate that while differentiation does affect binding of factors to the regions analyzed in the URR of HPV31, GR and, specifically, GREs do not appear to directly play a role in the regulation of these regions. DISCUSSION In this report, we have shown that the URR of HPV31 is inducible by the synthetic glucocorticoid dex in differentiating keratinocytes. This is in direct contrast to what occurs in nondifferentiating keratinocytes, where dex treatment had no effect on the transcriptional activity of the URR. As the life cycle of HPV is tightly linked to the differentiation of the host tissue, these studies suggest that glucocorticoids play an important role in the differentiation-dependent life cycle of HPV. The findings presented here further indicate the differences between the URRs of different HPV types. The URR of HPV18, which has previously been shown by others to contain a functional GRE, was inducible by dex in both monolayer and methylcellulose cultures in both 9E and HFK cells, while the URR of HPV31 was only inducible in methylcellulose cultures in 9E cells. Since the URR contains numerous cis elements for the binding of both viral and cellular transcription factors, these findings implicate the binding activity and/or the endogenous levels of these factors as being responsible for the differences observed. In fact, changes in both the expression and binding activity upon differentiation of various transcription factors, including AP1 (48), YY1 (1), Sp1 (4, 48), and CDP (2), have been demonstrated by others. Therefore, it seems plausible to suggest that the URR of HPV31 is inducible by dex in a differentiating system and not in an undifferentiated system because of these changes. As the URR of HPV31 is not in-
ducible by dex in HFK cells (Fig. 5), it appears that viral factors may also play a role. Why the URR of HPV18 is not similarly affected is unclear, although differences in binding sites and binding affinities of transcription factors are logical explanations. CDP may be the factor to focus on, as it has been shown to be downregulated in differentiated keratinocytes, thereby relieving repression on the URR (1). CDP has also been shown to repress the dex-inducible transcription of the mouse mammary tumor virus long terminal repeat, which is a well-characterized dex-inducible enhancer (49). Therefore, it is possible that the URR of HPV31 is more sensitive to CDP repression than the URR of HPV18, so that CDP represses dex activity on the URR of HPV31 in monolayer culture but not that on the HPV18 URR. A portion of the URR of HPV31, nt 7238 to 7557, appears to contain one or more elements that are responsible for the dex inducibility seen in methylcellulose cultures (Fig. 2). While two GREs that we identified by sequence analysis are located in this region, these elements do not appear to play a role in this inducibility, as dex induction was not lost upon mutations of these two GRE sites (Fig. 1). Utilizing LSMs encompassing this region, we were able to determine that nt 7242 to 7259 (mutant A14) and nt 7260 to 7277 (mutant B1) were both required for this induction (Fig. 3). Surprisingly, GRE2 is contained in this region (nt 7254 to 7268) (Fig. 2A) (10). However, we showed that this GRE is not responsible for dex induction, as specific mutagenesis of this GRE failed to abolish dex induction of the URR of HPV31 (Fig. 1). Sequence analysis of this region revealed two potential GRE half sites compared to the GREc sequence (Fig. 7A), one 5⬘ of GRE2 (and contained in the A14 region) and one 3⬘ of GRE2 (and contained in the B1 region) (Fig. 7B). A functional GRE is most commonly arranged as a 15-bp incomplete palindrome separated by a 3-bp spacer (Fig. 7A) (6, 42). However, other possible organizations have been reported, including overlapping
10982
BROMBERG-WHITE ET AL.
GREs (18), directly repeating half sites (13), and variable spacing of half sites (44). Furthermore, studies have shown that the rotational positioning of a GRE in a nucleosome is critical for its accessibility (28) and that GR binding to the GRE induces DNA bending (37). We have found that DNA bending of the URR of HPV31 in the region containing A14 and B1 could result in the creation of a complete GRE similar to the GREc (Fig. 7C). We were surprised to find that loss of nt 7883 to 7900 by linker scanning mutagenesis led to an increased induction of the URR of HPV31 by dex in differentiating cells (Fig. 4). This region in the URR of HPV31 corresponds to the location of the GRE in the URR of HPV18. Sequence analysis of this region identified a weakly consensus GRE (7 of 12 nt consensus) as well as an overlapping, perfectly consensus GRE half site (nt 7898 to 7903). Whether GR can bind any of these sites is unknown. We also identified a potential STAT5 binding site in this region of the URR. STATs (signal transducers and activators of transcription) are a family of transcription factors that are activated by cytokines such as interleukin-2 (8), interleukin-6 (43), and alpha interferon (29). This activation can result in either positive or negative regulation of glucocorticoid-induced transcription, the outcome of which is dependent on promoter context and cell type (8). It is possible that STAT5 may bind this region of the URR and repress dex activity. The context of the URR in this region appears to be important, as placement of the HPV18 GRE in this location results in a repression of dex induction (Fig. 4). This result was not expected, as nt 7883 to 7900 in the URR of HPV31 correspond to the native location of the HPV18 GRE in the URR of HPV18. The fact that the HPV18 GRE acts as a positive regulator of dex inducibility in the URR of HPV18 and a negative regulator in the URR of HPV31 suggests differences in the overall context of the URRs of these two viral types. The URR of HPV31 was not inducible by dex in HFK cells either in monolayer or methylcellulose cultures (Fig. 5). It is possible that viral gene products are necessary for this induction, as the URR is inducible in 9E methylcellulose cultures, which contain endogenous HPV31b and express early viral genes. E7 has been shown to act as a transcriptional activator (50) and can complex with AP1 factors such as c-jun, Jun B, and Jun D and can transactivate c-jun-induced transcription (3). As AP1 is a well-known antagonist of GR action, it seemed unlikely that E7 could cause induction of dex action. The viral protein E2 could be important in glucocorticoid-dependent activation of the URR of HPV31, as an E2 binding site is located in the region of the URR responsible for dex inducibility (Fig. 2A). Analysis of protein binding to regions A14 and B1, as well as regions B2, B3, B4, and B5, indicated that the GR did not directly bind these regions, as the GREc sequence did not clearly compete away the bound complex (Fig. 8), suggesting that GR either (i) cooperatively binds these regions, resulting in the strengthening of binding of a second transcription factor, or (ii) strengthens the binding of another factor but does not itself bind these regions. Studies have shown that induction of the mouse mammary tumor virus MMTV promoter by glucocorticoids requires functional synergism between GR and nuclear factor 1, with simultaneous occupancy of the promoter binding sites for these factors (5). In addition, hormonal in-
J. VIROL.
duction of the MMTV promoter requires binding of other transcription factors to adjacent promoter sequences in addition to the binding of hormone receptors to their binding sites (11). Others have shown cooperative transactivation of genes such as -casein by STAT5, CCAAT/enhancer binding protein  (C/EBP), and GR, with this cooperativity lost in the absence of GR (46), as well as transcriptional cooperativity of the whey acidic protein by NF1, STAT5, and GR (34). The present study suggests that regions in the URR of HPV31 within nt 7238 to 7557, especially regions A14 and B1, are involved in GR-mediated cooperative transactivation, as loss of these regions by linker scanning mutagenesis results in loss of glucocorticoid-dependent activation (Fig. 3), without any direct involvement of GR binding to these regions (Fig. 8). In conclusion, the URR of HPV31 has been shown to be inducible by the glucocorticoid dex in a differentiating system. From these findings, it is clear that there is a complex interplay of negative and positive factors on the URR of HPV. Furthermore, we have shown that glucocorticoid-dependent activation of the URR of HPV is a complex process and is not the same between different HPV types. ACKNOWLEDGMENTS We thank David Spector and the Meyers laboratory for helpful discussions. This work was supported by NIH training grants CA60395-06 (J.L.B.-W.) and CA79006 and PA-HEALTH 98-07-17 (C.M.). REFERENCES 1. Ai, W., J. Marahari, and A. Roman. 2000. Yin Yang 1 negatively regulates the differentiation-specific E1 promoter of human papillomavirus type 6. J. Virol. 74:5198–5205. 2. Ai, W., E. Toussaint, and A. Roman. 1999. CCAAT displacement protein binds to and negatively regulates human papillomavirus type 6 E6, E7, and E1 promoters. J. Virol. 73:4220–4229. 3. Antinore, M. J., M. J. Birrer, D. Patel, L. Nader, and D. J. McCance. 1996. The human papillomavirus type 16 E7 gene product interacts with and trans-activates the AP1 family of transcription factors. EMBO J. 15:1950– 1960. 4. Apt, D., R. Watts, G. Suske, and H. Bernard. 1996. High Sp1/Sp3 ratios in epithelial cells during epithelial differentiation and cellular transformation correlate with the activation of the HPV-16 promoter. Virology 224:281–291. 5. Aumais, J. P., H. S. Lee, C. DeGannes, J. Horsford, and J. H. White. 1996. Function of directly repeated half-sites as response elements for steroid hormone receptors. J. Biol. Chem. 271:12568–12577. 6. Becker, P. B., B. Gloss, W. Schmid, U. Strahle, and G. Schutz. 1986. In vivo protein-DNA interactions in a glucocorticoid response element require the presence of the hormone. Nature 324:686–688. 7. Bedell, M., J. Hudson, T. Golub, M. Turyk, M. Hosken, G. Wilbanks, and L. Laimins. 1991. Amplification of human papillomavirus genomes in vitro is dependent on epithelial differentiation. J. Virol. 65:2254–2260. 8. Biola, A., P. Lefebvre, M. Perrin-Wolff, M. Sturm, J. Bertoglio, and M. Pallardy. 2001. Interleukin-2 inhibits glucocorticoid receptor transcriptional activity through a mechanism involving STAT5 (signal transducer and activator of transcription 5) but not AP-1. Mol. Endocrinol. 15:1062–1076. 9. Broker, T., and M. Botchan. 1986. Papillomaviruses: retrospectives and prospectives. Cancer Cells 4:17–36. 10. Bromberg-White, J. L., and C. Meyers. 2002. The upstream regulatory region of human papillomavirus type 31 is insensitive to glucocorticoid induction. J. Virol. 76:9702–9715. 11. Bruggemeier, U., M. Kalff, S. Franke, C. Scheidereit, and M. Beato. 1991. Ubiquitous transcription factor OTF-1 mediates induction of the MMTV promoter through synergistic interaction with hormone receptors. Cell 64: 565–572. 12. Butz, K., and F. Hoppe-Seyler. 1993. Transcriptional control of human papillomavirus (HPV) oncogene expression: composition of the HPV type 18 upstream regulatory region. J. Virol. 67:6476–6486. 13. Chan, G. C., P. Hess, T. Meenakshi, J. Carlstedt-Duke, J. A. Gustafsson, and F. Payvar. 1991. Delayed secondary glucocorticoid response elements. Unusual nucleotide motifs specify glucocorticoid receptor binding to transcribed regions of ␣2u-globulin DNA. J. Biol. Chem. 266:22634–22644. 14. Chan, W., G. Klock, and H. Bernard. 1989. Progesterone and glucocorticoid
VOL. 77, 2003
15. 16. 17. 18.
19.
20.
21. 22. 23. 24. 25. 26. 27. 28. 29.
30.
31. 32. 33.
HPV31 URR DIFFERENTIATION-DEPENDENT DEX INDUCTION
response elements occur in the long control region of several human papillomaviruses involved in anogenital neoplasia. J. Virol. 63:3261–3269. Eckert, R., J. Crish, and N. Robinson. 1997. The epidermal keratinocyte as a model for the study of gene regulation and cell differentiation. Physiol. Rev. 77:397–424. Flores, E. R., and P. F. Lambert. 1997. Evidence for a switch in the mode of human papillomavirus type 16 DNA replication during the viral life cycle. J. Virol. 71:7167–7179. Frattini, M. G., H. B. Lim, and L. A. Laimins. 1996. In vitro synthesis of oncogenic human papillomaviruses requires episomal genomes for differentiation-dependent late expression. Proc. Natl. Acad. Sci. USA 93:3062–3067. Garlatti, M., M. Daheshia, E. Slater, J. Bouguet, J. Hanoune, M. Beato, and R. Barouki. 1994. A functional glucocorticoid-responsive unit composed of two overlapping inactive receptor-binding sites: evidence for formation of a receptor tetramer. Mol. Cell. Biol. 14:8007–8017. Gloss, B., H. U. Bernard, K. Seedorf, and G. Klock. 1987. The upstream regulatory region of the human papilloma virus-16 contains an E2 proteinindependent enhancer which is specific for cervical carcinoma cells and regulated by glucocorticoid hormones. EMBO J. 6:3735–3743. Grassmann, K., B. Rapp, H. Mascjek, K. Petry, and T. Ifner. 1996. Identification of a differentiation-inducible promoter in the E7 open reading frame of human papillomavirus type 16 (HPV-16) in raft cultures of a new cell line containing high copy numbers of episomal HPV-16 DNA. J. Virol. 70:2339– 2349. Howley, P. 1996. Papillomavirinae: the viruses and their replication, p. 2045– 2076. In B. N. Fields, D. M. Knipe, and P. M. Howley (ed.), Fields virology, 3rd ed., vol. 2. Lippincott-Raven Publishers, Philadelphia, Pa. Hummel, M., J. B. Hudson, and L. A. Laimins. 1992. Differentiation-induced and constitutive transcription of human papillomavirus type 31b in cell lines containing viral episomes. J. Virol. 66:6070–6080. Hummel, M., H. Lim, and L. Laimins. 1995. Human papillomavirus type 31b late gene expression is regulated through protein kinase C-mediated changes in RNA processing. J. Virol. 69:3381–3388. Jeon, S., B. L. Allen-Hoffmann, and P. F. Lambert. 1995. Integration of human papillomavirus type 16 into the human genome correlates with a selective growth advantage of cells. J. Virol. 69:2989–2997. Jonat, C., H. Rahmsdorf, K. Park, A. Cato, S. Gebel, H. Ponta, and P. Herrlich. 1990. Antitumor promotion and antiinflammation: down-modulation of AP1 (Fos/Jun) activity by glucocorticoid hormone. Cell 62:1189–1204. Kyo, S., A. Tam, and L. A. Laimins. 1995. Transcriptional activity of human papillomavirus type 31b enhancer is regulated through synergistic interaction of AP1 with two novel cellular factors. Virology 211:184–197. Lambert, P. F. 1991. Papillomavirus DNA replication. J. Virol. 65:3417– 3420. Li, Q., and O. Wrange. 1995. Accessibility of a glucocorticoid response element in a nucleosome depends on its rotational positioning. Mol. Cell. Biol. 15:4375–4384. Li, S., S. Labrecque, M. C. Gauzzi, A. R. Cuddihy, A. H. Wong, S. Pellegrini, G. J. Matlashewski, and A. E. Koromilas. 1999. The human papilloma virus (HPV)-18 E6 oncoprotein physically associates with Tyk2 and impairs JakSTAT activation by interferon-␣. Oncogene 18:5727–5737. Medina-Martinez, O., N. Morales-Peza, M. Yaniv, A. Garcia-Carranca, and F. Thierry. 1996. A single element mediates glucocorticoid hormone response of HPV18 with no functional interactions with AP1 or hbrm. Virology 217:392–396. Meyers, C. 1996. Organotypic (raft) epithelial tissue culture system for the differentiation-dependent replication of papillomavirus. Methods Cell Sci. 18:1–10. Meyers, C., M. Frattini, J. Hudson, and L. Laimins. 1992. Biosynthesis of human papillomavirus from a continuous cell line upon epithelial differentiation. Science 257:971–973. Meyers, C., T. Mayer, and M. Ozbun. 1997. Synthesis of infectious human
34.
35. 36. 37.
38.
39. 40.
41.
42. 43.
44. 45. 46.
47.
48. 49. 50.
10983
papillomavirus type 18 in differentiating epithelium transfected with viral DNA. J. Virol. 71:7381–7386. Mukhopadhyay, S. S., S. L. Wyszomierski, R. M. Gronostajski, and J. M. Rosen. 2001. Differential interactions of specific nuclear factor I isoforms with the glucocorticoid receptor and STAT5 in the cooperative regulation of WAP gene transcription. Mol. Cell. Biol. 21:6859–6869. Ozbun, M., and C. Meyers. 1997. Characterization of late gene transcripts expressed during vegetative replication of human papillomavirus type 31b. J. Virol. 71:5161–5172. Ozbun, M., and C. Meyers. 1998. Temporal usage of multiple promoters during the life cycle of human papillomavirus type 31b. J. Virol. 72:2715– 2722. Petz, L. N., A. M. Nardulli, J. Kim, K. B. Horwitz, L. P. Freedman, and D. J. Shapiro. 1997. DNA bending is induced by binding of the glucocorticoid receptor DNA binding domain and progesterone receptors to their response element. J. Steroid Biochem. Mol. Biol. 60:31–41. Ruesch, M., F. Stubenrauch, and L. Laimins. 1998. Activation of papillomavirus late gene transcription and genome amplification upon differentiation in semisolid medium is coincident with expression of involucrin and transglutaminase but not keratin-10. J. Virol. 72:5016–5024. Schule, R., P. Rangarajan, S. Kliewer, L. J. Ransone, J. Bolado, N. Yang, I. M. Verma, and R. M. Evans. 1990. Functional antagonism between oncoprotein c-Jun and the glucocorticoid receptor. Cell 62:1217–1226. Sen, E., J. L. Bromberg-White, and C. Meyers. 2002. Genetic analysis of cis regulatory elements within the 5⬘ region of the human papillomavirus type 31 upstream regulatory region during different stages of the viral life cycle. J. Virol. 76:4798–4809. Stoler, M. H., S. M. Wolinsky, A. Whitbeck, T. R. Broker, and L. T. Chow. 1989. Differentiation-linked human papillomavirus types 6 and 11 transcription in genital condylomata revealed by in situ hybridization with messagespecific RNA probes. Virology 172:331–340. Strahle, U., G. Klock, and G. Schutz. 1987. A DNA sequence of 15 base pairs is sufficient to mediate both glucocorticoid and progesterone induction of gene expression. Proc. Natl. Acad. Sci. USA 84:7871–7875. Takeda, T., H. Kurachi, T. Yamamoto, Y. Nishio, Y. Nakatsuji, K. Morishige, A. Miyake, and Y. Murata. 1998. Crosstalk between the interleukin-6 (IL-6)-JAK-STAT and the glucocorticoid-nuclear receptor pathway: synergistic activation of IL-6 response element by IL-6 and glucocorticoid. J. Endocrinol. 159:323–330. Truss, M., and M. Beato. 1993. Steroid hormone receptors: interaction with deoxyribonucleic acid and transcription factors. Endocr. Rev. 14:459–479. Ustav, M., and A. Stenlund. 1991. Transient replication of BPV-1 requires two viral polypeptides encoded by the E1 and E2 open reading frames. EMBO J. 10:449–457. Wyszomierski, S. L., and J. M. Rosen. 2001. Cooperative effects of STAT5 (signal transducer and activator of transcription 5) and C/EBP (CCAAT/ enhancer-binding protein-) on beta-casein gene transcription are mediated by the glucocorticoid receptor. Mol. Endocrinol. 15:228–240. Yang-Yen, H. F., J. C. Chambard, Y. L. Sun, T. Smeal, T. J. Schmidt, J. Drouin, and M. Karin. 1990. Transcriptional interference between c-Jun and the glucocorticoid receptor: mutual inhibition of DNA binding due to direct protein-protein interaction. Cell 62:1205–1215. Zhao, W., L. T. Chow, and T. R. Broker. 1997. Transcription activities of human papillomavirus type 11 E6 promoter-proximal elements in raft and submerged cultures of foreskin keratinocytes. J. Virol. 71:8832–8840. Zhu, Q., and J. P. Dudley. 2002. CDP binding to multiple sites in the mouse mammary tumor virus long terminal repeat suppresses basal and glucocorticoid-induced transcription. J. Virol. 76:2168–2179. Zwerschke, W., S. Joswig, and P. Jansen-Durr. 1996. Identification of domains required for transcriptional activation and protein dimerization in the human papillomavirus type-16 E7 protein. Oncogene 12:213–220.
