Human T-Cell Leukemia Virus Type 1 Tax Enhances Serum Response Factor DNA Binding and Alters Site Selection

Recombinant protein purification.Tax protein was purified as previously described (34). Briefly, Escherichia coli cells transformed with the pX5 plasmid encoding Tax were pelleted, sonicated, centrifuged to remove insoluble material, and precipitated sequentially with ammonium sulfate. Glutathione S-transferase (GST-SRF) (13) was purified according to the manufacturer's protocol (Amersham GE Healthcare). Briefly, GST-SRF fusion protein was expressed in E. coli, bacterial pellets were lysed in 100 mM Tris-HCl, pH 7.5, 200 mM KCl, 10 mM dithiothreitol, 10 mM MgCl₂, and then GST-SRF protein was purified in batch using glutathione beads. GST-SRF protein was eluted with glutathione elution buffer (25 mM reduced glutathione, 150 mM KCl, 50 mM Tris-HCl, pH 8.0).

Cell lines.Jurkat, CEM, C81-66 (48), and HuT102 (45) cell lines were all maintained in RPMI medium supplemented with 10% fetal bovine serum. MS-9 cells (51), provided by David Derse (National Cancer Institute, Fredrick, MD), were maintained as others with the addition of 100 U/ml interleukin-2 (NCI Preclinical Repository).

GST pull down.Purified Tax and GST-SRF proteins were tumbled with glutathione beads for 1 h. Beads were washed five times, and complexes were separated by 8% sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis. SRF was detected by Western blotting using an SRF-specific antibody (sc-335; Santa Cruz Biotechnology). Tax was detected using polyclonal rabbit anti-Tax serum.

Plasmids and oligonucleotides.The sequences of the c-fos, Nur77, and viral SRE sense-strand oligonucleotides used in gel shift assays were 5′-GATCAGATCCAGGATGTCCATATTAGGACATCTGT-3′, 5′-GATCCCGCCGGAACCGCGCCGCCCCCCGCGCCCTTGTATGGCCAAAC-3′, and 5′-GATCCGGAGACCTCCGGGAAGCCACCGGAACCACCCATTTCCTCCCCATGTTTGTCAAGCC-3′, respectively. The primers used to amplify the c-fos promoter in chromatin immunoprecipitation (ChIP) assays were 5′-ACCCTCGGTGTTGGCTG-3′ and 5′-TCCTAATCTCGTGAGCATTTCG-3′. The 5′ and 3′ PCR primers used to amplify the c-fos cDNA were 5′-TCGGGCTTCAACGCAGACTACG-3′ and 5′-TGACCGTGGGAATGAAGTTGGC-3′, respectively. Amplification of the Nur77 cDNA used the 5′ and 3′ primers 5′-TTGGCACCCACTTCTCCACACC-3′ and 5′-GCACTGTGCGCTTGAAGAAGCC-3′, respectively.

EMSAs.Double-stranded oligonucleotides were labeled with [α-³²P]dCTP using Klenow enzyme (New England Biolabs). Electrophoretic mobility shift assays (EMSAs) using purified proteins included 1× EMSA buffer (20 mM HEPES, pH 7.4, 100 mM KCl, 0.5 mM EDTA, 1 mM dithiothreitol, 0.1 mM phenylmethylsulfonyl fluoride, 10% glycerol), 1 μg sheared salmon sperm DNA (Sigma), 0.02 μg to 1 μg SRF, 2 μg Tax, and 0.5 nM labeled probe in a 20-μl total reaction volume. These reactions were incubated at room temperature for 20 to 30 min. Complexes were resolved on a 4% nondenaturing polyacrylamide gel in 0.6× Tris-borate-EDTA. Antibodies (2 μg per 20-μl reaction mixture) used in supershift EMSA experiments (anti-SRF [sc-335; Santa Cruz Biotechnology], monoclonal antibody Tab170 specific for Tax [NIH AIDS Research and Reference Reagent Program], and anti-AP-2 [Santa Cruz Biotechnology SC25343]) were added 15 min after initiating the protein and DNA incubation. Reaction mixtures were incubated at room temperature for 30 min after adding antibody.

SELEX.Specific SRF binding sequences were selected as previously described (46). Briefly, 1 μg of purified SRF alone, or together with 2 μg of Tax, was incubated with 0.4 ng of ³²P-labeled randomized oligonucleotide (5′-GTCAGGATCCGTTCAGCTGTCG(N)₁₆GAGGCAGTGCAATCTAGACTGC-3′), 50 μg/ml bovine serum albumin, and 200 ng of poly(dI-dC) for 30 min on ice. The complexes were then incubated with protein A beads (Upstate Biotechnology) that had been prebound with SRF-specific antibody (sc-335; Santa Cruz), or Tax-specific monoclonal antibody (Tab170; NIH AIDS Research and Reference Reagent Program) overnight with tumbling at 4°C. After washing three times with wash buffer (20 mM HEPES pH 7.9, 100 mM KCl, 0.2 mM EDTA, 0.2 mM EGTA, and 20% glycerol), DNA was eluted in 200 μl of recovery buffer (50 mM Tris-HCl pH 8.0, 100 mM sodium acetate, 5 mM EDTA, and 0.5% SDS) for 1 h at 45°C, phenol extracted, chloroform extracted, and then ethanol precipitated. Recovered probe was measured by Cerenkov counting in a scintillation counter, and 1 pg of recovered probe (compared to 1 pg of original systemic evolution of ligands by exponential enrichment [SELEX] input) was amplified by PCR using the primers 5′-GTCAGGATCCGTTCAGCTGTCG-3′ and 5′-GCAGTCTAGATTGCACTGCCTC-3′. The PCR products were purified from an 8% nondenaturing polyacrylamide gel. The amount of purified DNA was determined by Cerenkov counting using 1 pg of input DNA for comparison. The selection process was repeated for a total of five rounds. Selected DNA was blunt end cloned into pBlueScriptII (Stratagene) for sequencing. The ability of selected probes to bind SRF was verified by EMSA for each round of selection.

Chromatin isolation.Chromatin was isolated as previously described (36) with minor modifications. Briefly, 2.5 × 10⁶ cells were pelleted, washed two times with phosphate-buffered saline, resuspended in 1 ml buffer A (10 mM HEPES pH 7.9, 10 mM KCl, 1.5 mM MgCl₂, 0.34 M sucrose, 10% glycerol, and a Complete Mini EDTA-free protease inhibitor tablet [Roche]), Triton X-100 was added to 0.1%, and cells were incubated for 5 min on ice. Nuclei were collected by low-speed centrifugation, washed with buffer A, and pelleted as before. Pelleted nuclei were lysed in 200 μl buffer B (3 mM EDTA, 0.2 mM EGTA, and a Complete Mini EDTA-free protease inhibitor tablet [Roche]) for 30 min on ice. The insoluble chromatin fraction was collected by low-speed centrifugation and washed once in buffer B. The final chromatin pellet was resuspended in SDS loading dye and sonicated for 15 seconds. Samples were separated by 8% SDS-polyacrylamide gel electrophoresis and transferred to nitrocellulose for Western blotting. An antibody specific for ORC2 (BD Pharmingen) was used as a chromatin marker to distinguish the insoluble chromatin fraction from the soluble nuclear fraction.

ChIPs.ChIP assays were performed as previously described (58). Briefly, lysates from 5 × 10⁷ Jurkat, C81-66, or HuT102 formaldehyde cross-linked cells were precleared with protein G beads (Upstate Biotechnology). Beads were prebound with an SRF-specific antibody (sc-335; Santa Cruz Biotechnology). Prebound beads were incubated with precleared lysates in immunoprecipitation (IP) buffer (20 mM Tris-HCl, pH 8.0, 150 mM NaCl, 2 mM EDTA, 1% Triton X-100, 1 μg/ml leupeptin, 1 μg/ml aprotinin, 1 μg/ml pepstatin, 1 mM phenylmethylsulfonyl fluoride) (9). Precipitated beads were washed sequentially with IP buffer once, IP buffer with 0.5 M NaCl twice, LiCl wash buffer (20 mM Tris-HCl, pH 8.0, 250 mM LiCl, 2 mM EDTA, 0.5% Nonidet P-40) twice, and Tris-EDTA once (6, 7, 9). Elution buffer (0.1 M NaHCO₃, 1% SDS, 0.5 M NaCl) was added to the precipitated DNA-protein complexes, and the cross-links were reversed at 65°C overnight. Recovered DNA was analyzed by real-time PCR with primers specific for the c-fos promoter that spanned the SRF binding site.

mRNA isolation.mRNA was harvested using TRIzol reagent according to the manufacturer's protocol (Invitrogen). Briefly, 5 × 10⁶ cells were pelleted and resuspended in 1 ml of TRIzol. RNA was recovered by chloroform extraction and ethanol precipitation. After DNase treatment, avian myeloblastosis virus reverse transcriptase (Invitrogen) was used to produce cDNA from 1 μg of mRNA.

Real-time PCR.Real-time PCR for cDNA analysis was performed using Platinum SYBER Green qPCR SuperMix-UDG (Invitrogen). Reaction mixtures containing 1× PCR Master Mix buffer, specific primers, and 0.1 μg of cDNA were amplified using a Roto-Gene real-time PCR machine (Corbett Research). Real-time PCR analysis of ChIP samples was performed using SYBR Green JumpStart Taq ReadyMix (Sigma) according to the manufacturer's protocol. ChIP data were analyzed using the following equation: (c-fos_IP/c-fos_input)/(p53_IP/p53_input). All PCRs were performed in duplicate, and each experiment was repeated three times.

Tax and SRF interact.Previous studies demonstrated that Tax activates the expression of SRF-responsive genes and that Tax and SRF interact (23, 52). GST pull-down assays with recombinant purified Tax and GST-tagged SRF were used to confirm the direct interaction of Tax and SRF (Fig. 1A). Tax bound to SRF-loaded beads but not to beads in the absence of SRF. GST-SRF was observed as a doublet, which is likely due to a cleavage of the GST tag from SRF. The bottom band of the doublet ran near the expected size of the SRF protein (62 kDa), while the top band ran near the expected size of the fusion protein (90 kDa).

• Open in new tab
• Download powerpoint

FIG. 1.

Direct interaction of Tax and SRF. Western blots for SRF (top) and Tax (bottom) from GST pull-down assays. Controls include beads incubated with GST-SRF alone (lane 2), beads with Tax alone (lane 3), and beads only (lane 4).

If the interaction of Tax with SRF is important for activation of SRF-responsive genes, the SRF-Tax complex must be recruited to DNA. To determine whether Tax and SRF form a complex on DNA, SRE probes containing SRF binding sites from three different promoters (Fig. 2A) were used in supershift analyses. The c-fos SRE contains a canonical CArG box with the sequence CC(A/T)₆GG. We recently established that SRF binds to a noncanonical CArG box (vSRE) that resides within Tax-responsive region 2 of the HTLV-1 LTR (58). The vSRE is 80% identical to a canonical SRE, with a G in the A/T-rich region and a T in the final G position. Since the Nur77 promoter contains a functional CArG box with 90% identity to the c-fos CArG box sequence and has the same G mismatch that exists in the A/T-rich region of the viral SRE, the Nur77 promoter SRE was also used as a probe.

• Open in new tab
• Download powerpoint

FIG. 2.

Tax and SRF form a complex on DNA. (A) Nucleotide sequences of the viral, c-fos, and Nur77 CArG boxes. The EMSA was conducted using ³²P-labeled c-fos (B), Nur77 (C), or viral (D) SRE probes in the presence of GST-SRF alone (lane 3), Tax alone (lane 2), or GST-SRF and Tax (lane 4). Specific antibodies to Tax, SRF, and AP-2 (lanes 5, 6, and 7, respectively) were incubated with the indicated reaction mixtures. The location of supershifted bands is indicated on the right.

Incubation of GST-SRF with each of the labeled probes resulted in formation of a complex (Fig. 2B, C, and D, lane 3), while Tax alone did not form a complex with any of the labeled probes (Fig. 2B, C, and D, lane 2). If Tax could interact with SRF on DNA, we expected that the addition of Tax to the SRF gel shift reaction might reduce the mobility of the complex in the gel. Instead, incubation of Tax and SRF with the labeled probes produced complexes that migrated similar to the complex formed with SRF alone (Fig. 2B, C, and D, lane 4). However, addition of antibodies specific for Tax (lanes 5) or SRF (lane 6) further reduced the migration of the complex (Fig. 2B, C, and D), while a nonspecific antibody did not (Fig. 2B, C, and D, lane 7). Also, no shifted bands were detected when Tax- or SRF-specific antibodies were incubated with the probe alone (data not shown). These results suggest that Tax and SRF are both present in the complex. We also observed that the supershifted bands containing SRF antibody migrated more slowly than the supershifted bands containing Tax antibody. This is likely due to antibody differences, since the SRF antibody is purified rabbit immunoglobulin G while the Tax antibody is mouse ascites fluid.

Incubation of Tax and SRF together resulted in an increased intensity of the gel shift band (Fig. 2 B, C, and D, compare lanes 4 and 3). The increased intensity of the SRF-Tax complex was most readily observed when the c-fos and Nur77 SREs were used as probes. Therefore, we hypothesized that Tax enhances the binding of SRF to target DNA binding sites, specifically, cellular SREs.

Tax enhances the binding of SRF to DNA.Previous studies demonstrated that Tax can increase the affinity of cellular transcription factors, including ATF/CREB family members, Sp1, NF-κB, Fos-Jun heterodimer, and GAL4, for their target elements (3, 16, 18). In Fig. 2 we observed an increased intensity of the SRF complex in the presence of Tax and, thus, the ability of Tax to enhance the binding of SRF to various SREs was examined. Quantitative EMSAs were performed using the c-fos, Nur77, and viral SRE sequences described above. Increasing amounts of purified GST-SRF were incubated with each ³²P-labeled probe alone or in the presence of a constant amount of purified Tax. Incubation of increasing SRF concentrations with the probe resulted in increased binding to all three sequences (Fig. 3 A, B, and C, lanes 2 through 8). However, intensity of the bands was noticeably reduced on vSRE, which varies from the canonical SRF binding site by 20%. In the presence of Tax, complexes were visible at lower SRF concentrations (Fig. 3 A, B, and C; for example, compare lanes 5 and 12). Since quantitation of these results demonstrated significantly more SRF bound to each element in the presence of Tax (Fig. 3 D, E, and F), we conclude that Tax enhances the binding of SRF to all three SREs. The ability of Tax to enhance the DNA binding affinity of SRF on all three probes suggests that even SREs that diverge from the canonical CArG box may be better utilized in the presence of Tax. Tax has been reported to enhance the binding affinity of other transcription factors, including ATF/CREB family members, Sp1, Fos-Jun heterodimer, and GAL4 (3, 16, 18). Thus, enhancement of transcription factor DNA binding affinity appears to be a common mechanism utilized by Tax to activate transcription, even when the target DNA sequence is nonconsensus.

• Open in new tab
• Download powerpoint

FIG. 3.

Tax enhances the binding of SRF to SREs. ³²P-labeled c-fos (A), Nur77 (B), or viral (C) SRE probes were incubated with increasing amounts of GST-SRF alone (lanes 2 thru 8) or increasing GST-SRF with a constant amount of Tax (lanes 9 thru 15). Samples were analyzed by EMSA (left) and quantitated by phosphorimaging (right). Error bars represent the standard deviations from three independent experiments.

SRF tolerates more divergent sequences in the presence of Tax.The ability of Tax to enhance SRF binding affinity appeared to depend on the sequence of the DNA target site, prompting us to determine whether Tax could influence the target DNA sequence to which SRF binds. The SELEX method was used to identify DNA binding sequences preferred by SRF in the absence or presence of Tax. SRF was incubated alone, or in the presence of Tax, with a double-stranded 60-mer selection oligonucleotide containing 16 central randomized nucleotides flanked by 22 bp of defined sequence that were used for PCR amplification. DNA-protein complexes were allowed to form on this oligonucleotide, and the complexes were then immunoprecipitated with antibodies specific to either SRF or Tax. Precipitated DNA was eluted from the beads and amplified by PCR using primers specific for the conserved ends of the randomized selection oligonucleotide. The PCR product was subjected to another round of selection until five sequential rounds of selection had been performed. To ensure that sequences selected by the Tax:SRF complex bound SRF and Tax together and not SRF only, an antibody to Tax rather than SRF was used to retrieve the complexes.

Using this approach, a progressively greater portion of DNAs recovered after each round of selection should bind to SRF or Tax:SRF complexes. To test this prediction, DNA samples from each round of selection were analyzed by EMSA. SRF binding to sequences selected by GST-SRF alone was detected as early as round two (Fig. 4A, lane 8), increased slightly in round three, and remained relatively constant throughout subsequent rounds of selection. SRF binding to sequences selected by Tax and SRF were detected by round four and increased slightly after five rounds of selection (Fig. 4A, lanes 5 and 6). GST-SRF binding to DNAs selected in the negative control reactions was not observed in any round of SELEX (Fig. 4B).

• Open in new tab
• Download powerpoint

FIG. 4.

Enhanced SRF binding in later rounds of SELEX. (A) Enrichment of SRF binding sequences was examined by incubating GST-SRF with ³²P-labeled input probe for the SELEX experiment (lane 1) or with DNA recovered from each round of SELEX using either GST-SRF plus Tax with a Tax-specific antibody (lanes 2 thru 6) or GST-SRF with an SRF-specific antibody (lanes 7 thru 11). The resulting samples were analyzed by EMSA. (B) As negative controls, recovered DNA from each round of SELEX was incubated with Tax alone, along with a Tax-specific antibody (lanes 1 thru 5) or GST-SRF plus Tax in the absence of antibody (lanes 6 thru 10). The resulting samples were analyzed by EMSA.

After five rounds of selection, the recovered DNAs were cloned and sequenced. The sequences from 54 clones resulting from the binding of SRF alone and 42 clones resulting from the binding of Tax and SRF were analyzed. Selected sequences were categorized based on their divergence from the consensus CArG box sequence (Table 1 and Fig. 5). Using an antibody to SRF in the presence of GST-SRF alone, 40 of the 54 sequenced clones (74.1%) corresponded to a consensus CArG box. Sequences similar to those selected by GST-SRF were also reported previously in a SELEX study of SRF (46). Since we showed that Tax and SRF associate cooperatively with SREs and that Tax does not bind to DNA specifically, an antibody to Tax was used to select Tax:SRF:DNA complexes. In contrast to SRF alone, when a Tax-specific antibody was used to recover Tax and GST-SRF complexes, only 17 of the 42 sequenced clones (40.5%) contained a consensus CArG box sequence. In the presence of SRF and Tax, more sequences (59.5%) diverged from the canonical CArG box than in the presence of SRF alone (25.9%). To ensure that sequences selected in reactions containing SRF and Tax did not result from nonspecific Tax binding directly to DNA, we sequenced DNAs recovered from negative control reactions that included Tax alone. No DNAs capable of forming gel shift complexes with SRF were detected in these Tax-selected reactions (Fig. 4B), and DNA sequencing failed to identify CArG box sequences in theses samples (data not shown). CArG box sequences were also not detected in a second negative control reaction containing GST-SRF and Tax in the absence of a specific antibody (data not shown). These results demonstrate that in the presence of Tax, SRF can bind more divergent CArG box sequences than the absence of Tax.

• Open in new tab
• Download powerpoint

FIG. 5.

Tax affects SRF binding site selection. Sequences for each selected CArG box category are represented as percentages of the total sequences analyzed for SRF only (white bars) or SRF plus Tax (gray bars). Error bars represent confidence intervals.

Tax recruits SRF to chromatin in cells.The ability of Tax to promote SRF binding to divergent DNA sequences suggested that SRF would be recruited to an expanded set of cellular promoters in Tax-expressing cells. We tested this prediction by asking whether more SRF was associated with chromatin in Tax-expressing cells than in non-Tax-expressing cells. The insoluble chromatin fractions from Tax-negative cells (CEM) and Tax-positive cells (HuT102 and MS9) were collected and analyzed by Western blotting. Both of the Tax-positive cell lines tested contained higher levels of SRF in the chromatin fraction than did the Tax-negative cell line (Fig. 6A). The absence of chromatin contamination in the soluble fraction of the nuclear lysate was demonstrated by probing the blot with an antibody against origin of replication complex 2 (ORC2), which is known to be associated with chromatin (36) (Fig. 6B). As a control, Western blot assays for total SRF protein expression and actin as a loading control from the three cell lines were performed (Fig. 6C and D). These results demonstrated that more SRF is associated with chromatin in Tax-expressing cells, which is likely due to SRF recruitment to an expanded promoter set.

• Open in new tab
• Download powerpoint

FIG. 6.

Tax recruits SRF to chromatin. The chromatin fraction (lanes 1 thru 3) or soluble nuclear fraction (lanes 4 thru 6) of HTLV-1-negative cells (CEM) and HTLV-1-positive cells (HuT 102 and MS9) was analyzed by Western blotting for SRF (A) or for ORC2 (B). Whole-cell extracts were analyzed for total SRF (C) and actin (D) expression from CEM (lane 1), HuT 102 (lane 2), and MS9 (lane 3) cells.

Tax recruits SRF to specific promoters.To determine whether the increased association of SRF with chromatin in Tax-expressing cells could be observed on a promoter specifically responsive to SRF and Tax, we performed ChIP assays using an SRF-specific antibody and extracts from Jurkat cells (Tax negative) and C81-66 and HuT102 cells (Tax positive). After eluting the DNA precipitated from each cell line, the relative amount of c-fos promoter recovered in each ChIP was determined by real-time PCR (Fig. 7A). Quantitation was performed by normalizing the amount of c-fos promoter in the immunoprecipitated and input samples from each cell line to the amount of p53 promoter in the same samples. The p53 promoter does not contain an SRF binding site and therefore was used as the normalization standard in each experiment. Three independent experiments were performed, and the corrected amount of c-fos promoter precipitated from Jurkat cell extracts was set to 1 in each experiment (Fig. 7A). The corrected amount of c-fos promoter from each Tax-expressing cell line was normalized to Jurkat cells (Fig. 7A). The amount of c-fos promoter precipitated from Tax-expressing cells was sixfold higher than from Tax-negative cells. The Western blot assays in Fig. 7B and C demonstrate that the three cell lines express similar levels of SRF. Therefore, these results are consistent with our earlier in vitro EMSA findings, indicating that Tax increases the binding affinity of SRF to cellular promoters and providing further in vivo relevance that more SRF binds the c-fos promoter in Tax-expressing cells.

• Open in new tab
• Download powerpoint

FIG. 7.

Tax recruits SRF to specific promoters. (A) Relative levels of c-fos promoter precipitated from an SRF chromatin immunoprecipitation of HTLV-1-infected C81 and HuT 102 cells (gray bars) or uninfected Jurkat cells (white bar). The data were analyzed using the equation (c-fos_IP/c-fos_INPUT)/(p53_IP/p53_INPUT). (B) Western blots of whole-cell extracts from Jurkat, C81, and HuT 102 cells (lanes 1 to 3, respectively) were analyzed for total SRF expression. (C) As a loading control, blots from panel B were analyzed for actin expression.

Tax activates cellular SRF-responsive genes.The ability of Tax to interact with SRF and increase the affinity of SRF binding to promoters such as c-fos and Nur77 suggests that Tax can activate the expression of these genes. Tax has previously been shown to activate the c-fos promoter using transient-transfection assays, but the effects of Tax on the Nur77 promoter have not been previously reported. To determine whether the ability of Tax to enhance SRF binding and increase the diversity of SRF binding sites had functional effects on Tax transactivation, expression of the endogenous c-fos and Nur77 genes was examined in Tax-expressing and nonexpressing cells (Fig. 8). Real-time PCR was used to quantitate c-fos and Nur77 mRNA levels in Tax-expressing cells (HuT102 and MS9) and in non-Tax-expressing cells (Jurkat and CEM). Glyceraldehyde-3-phosphate dehydrogenase expression was determined from each RNA sample and used to normalize relative gene expression between cell lines. Expression of the c-fos and Nur77 transcripts was greater in Tax-expressing lymphocytes than in non-Tax-expressing lymphocytes, suggesting that Tax is a potent activator of these cellular genes.

• Open in new tab
• Download powerpoint

FIG. 8.

Tax activates c-fos and Nur77 RNA expression. Relative levels of c-fos (A) and Nur77 (B) mRNA in Tax-positive (HuT 102 and MS9) and Tax-negative (Jurkat and CEM) cell lines were analyzed by real-time PCR. Each graph represents three independent experiments. RNA levels of each gene in Jurkat cells were set to 1. *, Student t tests showed a significant difference between each Tax-positive and Tax-negative cell line individually, all with P values of <0.05.