Logo of molcellbPermissionsJournals.ASM.orgJournalMCB ArticleJournal InfoAuthorsReviewers
Mol Cell Biol. 2004 Jan; 24(1): 389–397.
PMCID: PMC303370

Recruitment of SWI/SNF to the Human Immunodeficiency Virus Type 1 Promoter


Following human immunodeficiency virus type 1 (HIV-1) integration into the host cell's genome, the 5′ long terminal repeat (LTR) is packaged into a highly specific chromatin structure comprised of an array of nucleosomes positioned with respect to important DNA sequence elements that regulate the transcriptional activity of the provirus. While several host cell factors have been shown to be important for chromatin remodeling and/or basal transcription, no specific mechanism that relieves the transcriptional repression imposed by nuc-1, a positioned nucleosome that impedes the start site of transcription, has been found. Since phorbol esters cause the rapid disruption of nuc-1 and markedly stimulate HIV-1 transcription, we looked for protein factors that associate with this region of the HIV-1 promoter in a phorbol-ester-dependent manner. We report here that ATF-3, JunB, and BRG-1 (the ATPase subunit of the 2-MDa human chromatin remodeling machine SWI/SNF) are recruited to the 3′ boundary of nuc-1 following phorbol myristate acetate stimulation in Jurkat T cells. Analysis of the recruitment of BRG-1 in nuclear extracts prepared from Jurkat T cells and reconstitution of an in vitro system with purified components demonstrate that ATF-3 is responsible for targeting human SWI/SNF (hSWI/SNF) to the HIV-1 promoter. Importantly, this recruitment of hSWI/SNF required HMGA1 proteins. Further support for this conclusion comes from immunoprecipitation experiments showing that BRG-1 and ATF-3 can exist together in the same complex. Although ATF-3 clearly plays a role in the specific targeting of BRG-1 to the HIV-1 promoter, the maintenance of a stable association between BRG-1 and chromatin appears to be dependent upon histone acetylation. By adding BRG-1 back into a BRG-1-deficient cell line (C33A cells), we demonstrate that trichostatin A strongly induces the 5′-LTR-driven reporter transcription in a manner that is dependent upon BRG-1 recruitment.

Recently, it has become clear that histones, and the specific chromatin structures that they define, are major contributors to the tightly regulated gene expression that is required for cells to grow, differentiate, and respond to environmental signals. Cellular mechanisms which modify chromatin structure to assist in both the activation and the repression of transcription are in place. While two main chromatin-modifying activities, ATP-dependent chromatin remodeling and histone acetylation, have been the focus of much attention (27, 44), the way these activities are targeted to specific promoters and how they act in concert with transcription factors, other nonhistone proteins, and histone modifications to manipulate chromatin structure remain unclear.

The cellular requirement to alter the chromatin structure in distinct ways is emphasized by the existence of several different ATP-dependent chromatin remodeling activities. The 2-MDa human SWI/SNF (hSWI/SNF) multisubunit complex is one such chromatin remodeling machine that, like in other members of this family (ISWI and Mi-2), utilizes the energy from ATP hydrolysis to disrupt nucleosomes. How SWI/SNF disrupts histone-DNA contacts to facilitate transcription factor binding is not clear, but it is believed to either twist the DNA helix in a screw-like manner or propagate a DNA bulge around the nucleosome surface (11, 13, 15, 19, 26).

One mechanism by which SWI/SNF is targeted to specific promoters is through interactions with gene-specific activators (49). Among the factors that can interact with SWI/SNF are MyoD (6), c-Myc (5), and C/EBPβ (28), as well as the ligand-activated nuclear hormone receptors retinoic acid receptor/retinoid X receptor (7), the glucocorticoid receptor (14, 30), and the estrogen receptor (8, 23). In addition, SWI/SNF itself may also contribute to its own recruitment since several subunits have functional DNA-binding and chromatin-binding domains (4, 22, 45).

In some instances, promoters may require histone acetyltransferase activities for activation in addition to ATP-dependent remodelers. Several yeast genetic studies have indicated a functional link between the SAGA acetyltransferase complex and SWI/SNF (31, 34). It is far from understood, however, how the activities of these two complexes are coordinated. Recently, a direct link was established when it was demonstrated that SWI/SNF bound more stably to an acetylated nucleosomal array in vitro. The bromodomain, found in BRG-1 as well as several chromatin-binding proteins, including several members of the HAT family, is an acetylated lysine residue binding motif (12, 22, 48).

Many of the insights into how chromatin contributes to gene regulation have come from studying the strategies employed by integrated viruses, which use host cell machineries to express their own genomes. Following infection of a host cell, human immunodeficiency virus type 1 (HIV-1) becomes integrated into the host's genome as chromatin. Verdin and colleagues have shown that nuc-1, which in the transcriptionally repressed virus is positioned near the start site of viral transcription, is rapidly disrupted upon treatment of latently infected cells with either tumor necrosis factor alpha (TNF-α) or phorbol myristate acetate (PMA), both of which stimulate AP1 and ATF/CREB activity (43). Three AP1 sites within nuc-1 have been described as being critical for viral transcription and replication (10, 32, 35, 40) (Fig. (Fig.1A).1A). Significantly, long terminal repeat (LTR)-driven transcription can also be induced by treatment of cells with inhibitors of histone deacetylases, such as sodium butyrate (29), trapoxin, and trichostatin A (TSA) (41). In vitro, HIV-1 transcription was stimulated when preassembled nucleosomal templates were first acetylated by histone acetyltransferase activities that modified either histone H3 or H4. Taken together, these findings not only show a direct role for chromatin in regulating HIV-1 expression but also argue that histone acetylation is necessary for high levels of HIV-1 transcription (38). Clearly, understanding how nuc-1 is disrupted and the role of histone acetylation in this process will be essential to understanding how HIV-1 expression is regulated.

FIG. 1.
ATF-3 and JunB are recruited to nuc-1 DNA in response to PMA. (A) Schematic representation of the HIV-1 5′ LTR showing transcription factor-specific recognition sites for several constitutive and inducible transcription factors. Importantly, the ...

In order to understand how nuc-1 is disrupted, our strategy has been to identify the cellular factors that interact with the nuc-1 region. Here, we show that ATF-3 and JunB are recruited to this nucleosomal DNA in a PMA-dependent manner. Importantly, the inducible binding of ATF-3/JunB correlates with the recruitment of the hSWI/SNF chromatin remodeling complex containing BRG-1. Using purified components, we confirm that ATF-3 can recruit hSWI/SNF to the HIV-1 promoter but only in the presence of HMGA1. In addition, we used a BRG-1-deficient cell line to show that the targeted recruitment of BRG-1 is stabilized by histone acetylation and that this recruitment is accompanied by a dramatic increase in LTR-driven reporter expression in stably transfected cell lines.


Cell culture, transfection, and nuclear extract preparation.

The BRG-1/hBrm-deficient cervical carcinoma cell line C33A was maintained under standard conditions in Dulbecco's modified Eagle's medium (Life Technologies). Jurkat T cells were similarly maintained in RPMI 1640 (Life Technologies). All culture media were supplemented with 10% fetal bovine serum (CSL), antibiotics (60 μg of benzylpenicillin [CSL] per ml and 100 μg of streptomycin [Sigma] per ml), and 2 mM l-glutamine (CSL). C33A cells were manipulated following treatment with 2% trypsin in phosphate-buffered saline (PBS). All cells were induced in 30 nM PMA (Sigma) for 2 h and/or 500 nM TSA (Sigma) for 12 h unless otherwise indicated.

Transfection was performed by using calcium phosphate, and cells were treated with 40 mM chloroquine for 3 h. Cells were induced and harvested 72 h after transfection, and nuclear extracts were prepared as previously described (21). Recombinant His-tagged ATF-1 and ATF-3 were purified with Ni-nitrilotriacetic acid resin (QIAGEN) according to the method of Thanos and Maniatis (39). Recombinant HMGA1a and HMGA1b were expressed and purified according to the method described by Reeves and Nissen (33). hSWI/SNF was purified from Flag-tagged Ini1 HeLa cells as described previously (37).

Immobilized-template assays.

Two hundred micrograms of Dynabeads M280 streptavidin (Dynal) was prepared as described in the manufacturer's instructions, concentrated over a magnet (Dynal), and resuspended in 20 μl of buffer T (10 mM Tris [pH 7.5], 1 mM EDTA, 1 M NaCl) containing 10 pmol of biotinylated oligonucleotide probe (see Fig. Fig.11 for DNA sequences). For nucleosome reconstitution, a 220-bp probe was generated (nucleotides −30 to 190) by using a PCR mixture that was both biotinylated and fluorescently labeled. Primers were biotin-5′-TCAGATCCTGCATATAAGCAGC-3′ and 4,7,2′,7′-tetra-chloro-6-carboxyfluorescein-5′-GCTTTCAGGTCCCTGTTCGG-3′. After gentle agitation at room temperature for 1 h, beads were washed several times in buffer T to remove the unbound probe. Bead-coupled probes were equilibrated in buffer R (10 mM Tris [pH 7.5], 1 mM MgCl2 0.1% NP-40, 1 mM EDTA, 10 mM dithiothreitol [DTT], 5% glycerol, 1% sucrose, 60 mM KCl, 12 mM HEPES [pH 7.9], 0.3 mg of bovine serum albumin [BSA] per ml) for 30 min, concentrated and resuspended in buffer R containing 250 μg of nuclear protein and 40 ng of poly(dG-dC) per μl in a final volume of 120 μl, and agitated gently for 30 min at room temperature.

Assays with purified and recombinant proteins and complexes were performed under conditions identical to those for nuclear extracts. Twenty picomoles of each recombinant component (ATF-1, ATF-3, HMGA1a, and HMGA1b) and 0.5 pmol of purified hSWI/SNF complex were used in a 64-μl reaction mixture containing 4 ng of poly(dG-dC) per μl. All binding reaction mixtures were then washed three times in buffer R containing 10 ng of poly(dG-dC) per μl to reduce nonspecific protein binding before the beads were again concentrated and resuspended in sodium dodecyl sulfate loading dye, loaded on a 5 to 20% sodium dodecyl sulfate-polyacrylamide gel electrophoresis gradient gel (Gradipore), and electrophoresed at 15 V/cm for approximately 1.5 h before being analyzed by Western blotting as previously described (21).


Brg-1, JunB, or ATF-3 was immunoprecipitated by incubating 250 μg of PMA-induced Jurkat nuclear extract with either 2 μg of anti-BRG-1 antibodies (a kind gift from R. Kingston), 5 μg of JunB, or 5 μg of ATF-3 antibodies (Santa Cruz Biotechnologies), respectively, and incubated at 4°C overnight with gentle mixing. Protein A-Sepharose beads (Amersham Pharmacia) were swelled and blocked by agitation with three changes of PBS supplemented with 10 mg of BSA (Sigma) per ml. After the final wash, protein A beads were resuspended in an equal packed volume with buffer D (100 mM KCl, 20% glycerol, 0.2 mM EDTA, 20 mM HEPES [pH 7.9], 1 mM DTT). Extract-antibody mixtures were used to gently resuspend the beads, and the resulting mix was agitated at 4°C for 2 h. Immunoprecipitates were collected by centrifugation for 10 min at 4°C, washed thoroughly in buffer D, and subjected to Western analysis.

Nucleosome assembly.

Chicken long chromatin was prepared as described previously (9). Acetylated HeLa chromatin was prepared according to the method of Ausio and van Holde (2). Nucleosomes were assembled onto probe DNA by using a salt gradient dialysis procedure as previously described (42).

Chromatin immunoprecipitation assays.

Chromatin immunoprecipitation was performed according to the method of Weinmann et al. (46), with minor alterations. Briefly, formaldehyde was added at a final concentration of 1% directly to cell culture media of C33A cells at approximately 90% confluence (107 cells). Fixation proceeded at room temperature for 10 min and was stopped by the addition of glycine to a final concentration of 0.125 M. The C33A cells were treated with 20% trypsin in PBS, scraped from the flask, collected and washed by centrifugation with warm media, and rinsed in cold PBS. Cell nuclei were isolated, lysed, and sonicated as described previously (46). Chromatin solutions were precleared with the addition of 50 μl of protein A-Sepharose slurry (equal volumes of wet-packed protein A-Sepharose beads [Pharmacia] and a solution containing 1 mM EDTA, 50 mM Tris-HCl [pH 8.0], 0.05% Na-azide, 200 μg of sonicated salmon sperm DNA per ml, and 500 μg of BSA per ml) for 1 h at 4°C. Protein A-Sepharose was pelleted at 3,000 rpm in an Eppendorf microcentrifuge for 10 min at 4°C. The precleared chromatin supernatant from cells was incubated with 1.5 μg of affinity-purified rabbit anti-BRG-1 polyclonal antibody or no antibody and rotated at 4°C for 12 to 18 h. Antibody-nuclear lysates were rotated with 70 μl of protein A-Sepharose slurry for 2 h at 4°C. Immunoprecipitation of antibody-protein complexes was performed, and BRG-1-associated DNA sequences were purified (46). Semiquantitative multiplex PCR was used to analyze purified DNA. The primers used to detect nuc-1 DNA are as listed above. For the control, the forward and reverse primers were 5′-CAACCCGGTAAGACACGACT-3′ and 5′-CTACCAGCGGTGGTTTGTTT-3′, respectively.

Luciferase assays.

Cells were harvested by centrifugation, resuspended, and lysed in 200 μl of cell lysis buffer (0.1 M K2HPO4 [pH 7.8], 1 mM EDTA [pH 8.0], 0.2 mM DTT, 1% [vol/vol] Triton X-100). Cell debris was pelleted by centrifugation, and the total protein concentration of the supernatant was determined with the Bio-Rad protein assay and normalized for total protein. Typically, 50 or 100 μg of protein was loaded onto 96-well plates combined with 200 μl of assay buffer (100 mM K2HPO4 [pH 7.8], 2 mM DTT, 10 mM MgSO4, 320 mM coenzyme A, 500 mM ATP). To begin the light reaction, 40 μl of d-luciferin (in 5 mM K2HPO4 buffer) was added to each well. The light reaction was read on a Top counter (Canberra Packard).


Inducible binding of ATF-3/JunB to a site located at the 3′ boundary of nuc-1.

An important first step in identifying the mechanisms of chromatin remodeling at the HIV-1 promoter is to identify the protein factors that interact with important DNA regulatory elements in response to phorbol esters. Figure Figure1A1A highlights the approximate location of nuc-1, which is positioned immediately downstream of the transcription start site and three key AP1 binding sites (sites AP1-1, -2, and -3) that reside within this nucleosomal DNA.

Previously, we employed an immobilized-template assay using a biotin-tagged oligonucleotide probe and magnetic beads (coupled with Western blot analysis) to isolate and identify transcription factors from Jurkat T-cell nuclear extracts that interact with site AP1-3. In this initial screen, we identified ATF-3 as an inducible transcription factor that interacted with this site (21) (Fig. (Fig.1B).1B). Given that leucine zipper (bZip) transcription factors often have dimerization partners, we used this assay to identify potential partners for ATF-3 by using a battery of antibodies raised against various bZip proteins (Fig. (Fig.1B,1B, lanes 3 and 4). To complement this analysis and to determine the specificity of bZip transcription factor binding, we also identified bZip transcription factors capable of binding to site AP1-1 (Fig. (Fig.1B,1B, lanes 5 and 6).

Figure Figure1B1B clearly shows that ATF-1, ATF-2, CREB-1, c-Fos, and JunD are all present in nuclear extracts but show little or no affinity for site AP1-3. On the other hand, fra-1 and fra-2 bound to probe AP1-3 irrespective of PMA treatment. Dependent upon PMA, ATF-4 appears to bind slightly better to this site even though it is not induced. Importantly, in addition to ATF-3, JunB binds specifically to site AP1-3 in response to PMA (Fig. (Fig.1B,1B, lanes 3 and 4). Interestingly, all of the bZip transcription factors that were not significantly induced by PMA (ATF-1, ATF-2, ATF-4, fra-1, fra-2, and JunD) were capable of associating with the AP1-1 probe (Fig. (Fig.1B,1B, lanes 5 and 6). We conclude that both JunB and ATF-3 are inducible factors that interact with site AP1-3 located at the boundary of nuc-1. In addition, we have also identified several bZip transcription factors that may play a role in maintaining the basal transcription of HIV-1 and/or may act as dimerization partners for inducible factors to allow a rapid transcriptional response (see below).

BRG-1 is recruited to site AP1-3 in response to PMA.

Given the nucleosomal location of these AP1 binding sites (Fig. (Fig.1A)1A) and the fact that Widlak et al. (47) have shown an ATP dependence of chromatin remodeling in this region in vitro, we wanted to investigate the possibility that a chromatin-modifying activity is recruited to nuc-1. The immobilized-template assay was repeated by using antibodies raised against the two known ATPase subunits of the hSWI/uSNF chromatin remodeling complex, namely, BRG-1 and hBrm. BRG-1 and hBrm exist only as part of distinct SWI/SNF complexes (27). Western blot analyses revealed that while neither hBrm nor BRG-1 was significantly induced by PMA (Fig. (Fig.2A,2A, lanes 1 and 2), BRG-1, but not hBrm, interacted with site AP1-3 but only following PMA stimulation (Fig. (Fig.2A,2A, lanes 3 and 4). On the other hand, no recruitment of BRG-1 to site AP1-1 was observed (Fig. (Fig.2A,2A, lanes 5 and 6).

FIG. 2.
BRG-1 is specifically recruited to an AP1 binding site located at the 3′ boundary of nuc-1. The immobilized-template assay was repeated as described in the legend of Fig. Fig.1,1, and the Western blot was probed with antibodies generated ...

Next, we employed the BRG-1-deficient cervical carcinoma cell line C33A to address the question of whether BRG-1 expressed from a transfected expression construct might also associate with site AP1-3 in vitro. JunB and ATF-3 were induced by PMA in this cell line (data not shown). C33A cells were transiently transfected with either a BRG-1 eukaryotic expression vector or a vector-alone plasmid. Figure Figure2B2B clearly shows that exogenously supplied BRG-1 is expressed in C33A cells (compare lane 1 with lanes 2 and 3) and that it can be recruited to site AP1-3 in a PMA-dependent manner (lanes 4 and 5).

Recruitment of BRG-1 correlates with ATF-3/JunB binding.

Given that the recruitment of BRG-1 to a DNA site located at the edge of nuc-1 is PMA dependent, we investigated the possibility that ATF-3/JunB may be responsible for this targeted recruitment. To support this hypothesis, we investigated whether ATF-3, JunB, and BRG-1 are recruited to site AP1-3 within similar time frames following PMA stimulation. Figure Figure3A3A clearly shows that within a 2-h time course following PMA treatment of cells, both ATF-3 and BRG-1 shared a parallel biphasic pattern of recruitment, with maximal binding of both of these factors occurring by 60 min. Interestingly, the first peak in the biphasic binding pattern of ATF-3 occurred within 10 min. This result probably reflects the binding of ATF-3 already present in the nucleus prior to PMA stimulation. Potentially, at this time point, ATF-3 may heterodimerize with a basal bZip family member. The second peak of ATF-3 binding, after 60 min, reflects the binding of newly synthesized protein. JunB, while not recruited in a similar biphasic pattern, also reaches its maximal binding after 1 h of PMA stimulation.

FIG. 3.
BRG-1 is recruited to site AP1-3 by ATF-3/JunB. (A) Nuclear extracts were prepared from unstimulated Jurkat T cells (NE) or cells stimulated with PMA (PMA-NE) for 10, 20, 30, 60, and 120 min. Proteins that bound to site AP1-3 at each time point were purified ...

Next, we mutated the second half of binding site AP1-3 to prevent both JunB and ATF-3 from binding (CTTTTAGTCAGTC to CTTTTAGAGTGTC) and repeated the immobilized-template assay. Importantly, this mutagenesis not only prevented the interaction between ATF-3/JunB and site AP1-3 but also inhibited the recruitment of BRG-1 (Fig. (Fig.3B3B).

Having determined that the major peak of ATF-3/JunB and BRG-1 binding to site AP1-3 occurred at the same time, we wanted to examine whether these proteins might interact prior to their association with template DNA. To address this, we immunoprecipitated nuclear extracts prepared from Jurkat T cells, following stimulation with PMA for 2 h, with antibodies raised against ATF-3, JunB, or, as a control, histone H2A. After high-stringency washing, a Western blot was carried out by using BRG-1 antibodies. As can be seen in Fig. Fig.4A,4A, BRG-1 coimmunoprecipitates with both JunB and ATF-3 (lanes 4 and 5). Consistent with this observation, both JunB and ATF-3 are recovered when nuclear extracts are immunoprecipitated with BRG-1 antibodies (Fig. (Fig.4B4B).

FIG. 4.
ATF-3/JunB and BRG-1 interact together prior to their association with template DNA. (A) PMA-induced nuclear extract (250 μg) was immunoprecipitated with 5 μg of ATF-3, JunB, or H2A antibodies. The results of a Western blot assay used ...

In summary, these experiments have shown that (i) ATF-3 and BRG-1 are recruited to site AP1-3 with similar kinetics; (ii) maximal DNA binding of ATF-3, JunB, and BRG-1 occurred by the 60-min time point; (iii) DNA mutations that inhibit ATF-3 and JunB binding also prevent the recruitment of BRG-1; and (iv) coimmunoprecipitation experiments indicate that BRG-1 can interact with ATF-3 and JunB prior to their association with site AP1-3. Based on these findings, we suggest that ATF-3 and/or JunB is involved in targeting BRG-1 to a site located at the 3′ boundary of nuc-1.

ATF-3 and HMGA1 are sufficient to recruit hSWI/SNF.

To test directly whether ATF-3 is able to specifically recruit hSWI/SNF to site AP1-3, the immobilized-template assay was repeated with hSWI/SNF purified from Flag-tagged Ini1 HeLa cells (37) and recombinant His-tagged ATF-3 and ATF-1 (ATF-1 served as a negative control since it does not bind to site AP1-3 in nuclear extracts [Fig. [Fig.1B]).1B]). Under the conditions of the assay, no binding of ATF-1 or ATF-3 to site AP1-3 was detected (Fig. (Fig.5,5, lanes 2 and 3, respectively). Not surprisingly, recruitment of hSWI/SNF was also not observed (lanes 8 and 9, respectively).

FIG. 5.
ATF-3 and the HMGA1 proteins recruit SWI/SNF to site AP1-3. Purified recombinant ATF-3 or ATF-1 (as a control) in the presence or absence of HMGA1 proteins were assayed by the immobilized-template assay for their ability to differentially recruit purified ...

Previously, Henderson et al. reported the importance of HMGA1 proteins in enhancing the binding of ATF-3 to site AP1-3 in Jurkat T cells induced with PMA (21). Figure Figure55 confirms this observation, and interestingly, while equal concentrations of unmodified HMGA1a and HMGA1b stimulated ATF-3 binding, only HMGA1a enhanced ATF-1 binding (it is worth noting that in gel shift competition assays, ATF-1 binds poorly to site AP1-3 when ATF-3 is present [data not shown]). Most significantly, in the presence of HMGA1 proteins, ATF-3, but not ATF-1, can recruit hSWI/SNF to site AP1-3 (compare lanes 13 and 11 with lanes 12 and 10). The binding of ATF-3 is also enhanced in the presence of hSWI/SNF, perhaps indicating the formation of a stable enhanceosome-like complex between HMGA1 proteins, ATF-3, and hSWI/SNF. We conclude that ATF-3 is able to specifically recruit hSWI/SNF but only in the presence of HMGA1 proteins. We were unable to examine the contribution of JunB to hSWI/SNF recruitment because it could not be expressed as a recombinant protein.

The binding of BRG-1 to nuc-1 in vitro and in vivo is stabilized by histone acetylation.

We next investigated whether BRG-1 might be recruited to template DNA assembled into a nucleosome. To do this, we generated a 220-bp biotinylated DNA fragment that contained the entire nuc-1 region. Surprisingly, whereas ATF-3/JunB could bind to the longer, naked DNA probe, the affinity of BRG-1 for the longer DNA template was markedly reduced compared to that for the shorter DNA probe (Fig. (Fig.6A,6A, lanes 3 and 4). The reason for this reduction in affinity is unclear, but it is possible that nonspecific DNA-binding proteins present in the crude nuclear extract bind more efficiently to the longer 220-bp probe, thus preventing or disrupting the association of BRG-1 with site AP1-3.

FIG. 6.
Histone acetylation facilitates BRG-1 recruitment and stimulates HIV-1 reporter expression. (A) Unacetylated or acetylated nucleosomes were assembled onto a 220-bp DNA template that contained the nuc-1 region and incubated with 250 μg of induced ...

The 220-bp DNA probe was assembled into acetylated and unacetylated nucleosomes by using standard salt dialysis procedures, and the immobilized-template assay was repeated. When Jurkat T-cell nuclear extracts were incubated with an unmodified or an acetylated nucleosome, no apparent binding of JunB or ATF-3 to site AP1-3 was detected (Fig. (Fig.6A,6A, lanes 5 to 8). However, since the concentration of such transcription factors is low in the crude nuclear extract, a reduced level of ATF-3/JunB binding to nucleosomal templates may be below the limits of detection of the Western assay. Strikingly, BRG-1, while not binding to an unmodified nucleosome, is capable of interacting with an acetylated nucleosome. However, this binding was nontargeted because it was independent of a PMA stimulus (Fig. (Fig.6A,6A, compare lane 7 with lane 8). Therefore, histone acetylation plays a general role in stabilizing nucleosome-SWI/SNF interactions.

Next, we investigated whether BRG-1 is recruited to the HIV-1 promoter in vivo and whether this recruitment is also enhanced by histone acetylation. To do this, we transfected a BRG-1 eukaryotic expression construct into C33A cells that had an HIV-1 reporter construct stably integrated into its genome. Cells were either left untreated, stimulated with PMA, treated with the histone deacetylase inhibitor TSA, or exposed to both PMA and TSA. Following harvesting, a chromatin immunoprecipitation assay was performed with antibodies raised against BRG-1. Pulled-down DNA was analyzed by using semiquantitative PCR with primers specific for nuc-1. Figure Figure6B6B shows that BRG-1 does interact with the nuc-1 region in vivo and that this interaction is dependent upon PMA stimulation (compare lane 5 with lane 4). Importantly, and in agreement with the results of our in vitro nucleosome recruitment assay (Fig. (Fig.6A),6A), enhanced levels of histone acetylation caused by TSA treatment significantly promote the specific recruitment of BRG-1 to nuc-1 in the presence of PMA and, interestingly, also in the absence of PMA (compare lanes 7 and 6 with lane 4). Taking these findings into consideration (Fig. 6A and B), we conclude that histone acetylation plays an important role in stabilizing the interaction between hSWI/SNF and HIV-1 chromatin.

Finally, since BRG-1 is recruited to the nuc-1 region of the HIV-1 promoter in response to TSA treatment alone, we investigated whether this recruitment stimulates transcription. C33A cells containing a stably integrated HIV-1 luciferase reporter construct were transiently transfected with the BRG-1 eukaryotic expression construct or a vector-alone plasmid construct. Figure Figure6C6C shows that without BRG-1, PMA stimulation of C33A cells caused an increase in luciferase expression of about ninefold. Surprisingly, PMA-induced expression was not enhanced with BRG-1 even though BRG-1 was recruited to nuc-1 (Fig. 6A and B, respectively). This indicates that a factor required for high levels of HIV-1 reporter expression is limiting in C33A cells and, moreover, that hSWI/SNF recruitment is not sufficient for high levels of expression. We propose that this limiting factor may be a histone acetyltransferase because 5′-LTR-driven transcriptional activation of the HIV-1 reporter construct in the presence of BRG-1 is enhanced 43-fold in response to TSA alone and 54-fold when the cells received both PMA and TSA (compared to 16-fold and 22-fold, respectively, in the absence of BRG-1). Taken together, these results suggest that histone acetylation is required for the stable binding and transcriptional activity of hSWI/SNF.


The aim of this study was to determine how nuc-1 is remodeled by determining the nature of factors that are recruited in response to phorbol esters. In addition, we wanted to begin to investigate the role of histone acetylation in this process.

Although the three downstream AP1 transcription factor binding sites have been shown to play a role in HIV-1 transcription by relieving repressive chromatin structures (10, 40), the mechanisms by which they specifically manipulate nuc-1 have not been addressed. We have found that the ATPase subunit of the hSWI/SNF chromatin remodeling complex BRG-1, but not hBrm, is specifically recruited to one of these AP1 sites located at the edge of nuc-1 (site AP1-3) following PMA stimulation. While BRG-1 appears to be a constitutive factor in Jurkat T cells, its PMA-dependent recruitment implies that it is reliant on PMA-inducible factors. We propose that one of these factors is ATF-3 (with JunB as a possible dimerization partner). In support of this notion, several other studies have indicated that SWI/SNF can be recruited to various promoters through specific interactions with other bZip transcription factors (1, 3, 24, 28) or members of the closely related helix-loop-helix family (5, 6).

The recruitment of chromatin remodeling activities by inducible transcription factors may be a common mechanism for achieving the promoter-specific targeting of these complexes. We have shown that the interaction between BRG-1 and site AP1-3 is dependent on an intact AP1 binding sequence and, moreover, that it shares a biphasic pattern of recruitment concomitant with that of ATF-3. Importantly, both ATF-3 and BRG-1 associate with their target DNA within 10 min of PMA stimulation, which is in agreement with the finding that BRG-1 becomes tightly associated within the nucleus within 10 min of stimulation (50). This association provides a mechanistic basis by which nuc-1 can be rapidly disrupted, allowing viral transcription to proceed within 20 min of cellular stimuli (43). The second-highest peak of BRG-1 recruitment at 60 min provides a mechanism by which nuc-1 can be maintained in a disrupted state to sustain subsequent rounds of transcription. It is known that ATF-3 can heterodimerize with JunB (16), and since we see maximal DNA binding of both ATF-3 and JunB within 60 min of PMA treatment, this suggests that these factors may dimerize. The results of our immunoprecipitation experiments also support this proposal since BRG-1 can interact with both ATF-3 and JunB prior to their interaction with nuc-1 DNA. Finally, using purified components, we present direct evidence that ATF-3 (but not ATF-1) can recruit hSWI/SNF to site AP1-3 in the presence of HMGA1 proteins. Taken together, these findings suggest that hSWI/SNF, ATF-3, and JunB may be recruited together to form a stable complex with HMGA1 at the 3′ boundary of nuc-1.

In addition to confirming our previous observation that HMGA1a enhances the binding of ATF-3 to site AP1-3 (21), we also show here that this interaction is required for the recruitment of hSWI/SNF. We postulate that, in a process analogous to the that of the formation of the beta interferon gene enhanceosome (1), the assembly of ATF-3/JunB/HMGA1 (and other factors) into a stable complex at the edge of nuc-1 is the first step that directs the staged recruitment of chromatin-modifying enzymes that ultimately leads to the disruption of nuc-1.

Although we have provided strong evidence suggesting that ATF-3 is directly responsible for recruiting hSWI/SNF to the HIV-1 5′ LTR, both our in vitro and our in vivo chromatin recruitment assays indicate that bZip transcription factors alone may not be sufficient to efficiently recruit or stably retain hSWI/SNF on a chromatin template. We show here that, both in vivo and in vitro, histone acetylation facilitates the interaction of BRG-1 with chromatin. These findings both confirm and extend the results of Hassan et al. (17). In their studies, histones that had been acetylated in vitro were able to retain Saccharomyces cerevisiae SWI/SNF on promoter-proximal nucleosomes even after the sequence-specific transcription factors that initially recruited SWI/SNF were dissociated from the nucleosomal template. It was subsequently demonstrated that the bromodomain of yeast Swi2/Snf2, an acetyl-lysine-binding motif, is responsible for its ability to interact with acetylated chromatin (18). It is worth pointing out that a single bromodomain displays a relatively low affinity for acetylated histone tails (22, 25). We propose that ATF-3 and HMGA1 confer the specificity of recruiting hSWI/SNF to nuc-1 and that histone acetylation helps to stabilize the interaction with chromatin. It is also likely that other subunits of SWI/SNF, which possess DNA- or chromatin-binding motifs, contribute to keeping SWI/SNF bound to chromatin.

Most interestingly, in C33A cells, PMA-induced recruitment of hSWI/SNF did not activate HIV-1 transcription. However, in combination with TSA, a dramatic increase in transcription was observed. This novel observation suggests that histone acetylation is also required for the transcriptional activity of hSWI/SNF. It was recently shown that HDAC1 is displaced from nuc-1, with a corresponding increase in histone acetylation, upon TSA treatment (20). Previously, Van Lint et al. (41) and Sheridan et al. (36) showed that TSA causes the disruption of nuc-1, with a concomitant activation of HIV-1 transcription. Here, we have demonstrated that TSA can recruit BRG-1 to the HIV-1 promoter. Taken together, these findings strongly support our hypothesis that nuc-1 acetylation facilitates the recruitment of SWI/SNF and that this remodeling complex is subsequently responsible for the disruption of nuc-1. In the future, it will be important to establish the specific histone acetyltransferase that acetylates nuc-1.


Special thanks go to Bob Kingston for his continued support and for providing us with purified hSWI/SNF and BRG-1 antibodies.


1. Agalioti, T., S. Lomvardas, B. Parekh, J. Yie, T. Maniatis, and D. Thanos. 2000. Ordered recruitment of chromatin modifying and general transcription factors to the IFN-beta promoter. Cell 103:667-678. [PubMed]
2. Ausio, J., and K. E. van Holde. 1986. Histone hyperacetylation: its effects on nucleosome conformation and stability. Biochemistry 25:1421-1428. [PubMed]
3. Bouallaga, I., S. Massicard, M. Yaniv, and F. Thierry. 2000. An enhanceosome containing the Jun B/Fra-2 heterodimer and the HMG-I(Y) architectural protein controls HPV 18 transcription. EMBO Rep. 1:422-427. [PMC free article] [PubMed]
4. Bourachot, B., M. Yaniv, and C. Muchardt. 1999. The activity of mammalian brm/SNF2α is dependent on a high-mobility-group protein I/Y-like DNA binding domain. Mol. Cell. Biol. 19:3931-3939. [PMC free article] [PubMed]
5. Cheng, S. W., K. P. Davies, E. Yung, R. J. Beltran, J. Yu, and G. V. Kalpana. 1999. c-MYC interacts with INI1/hSNF5 and requires the SWI/SNF complex for transactivation function. Nat. Genet. 22:102-105. [PubMed]
6. de la Serna, I. L., K. A. Carlson, and A. N. Imbalzano. 2001. Mammalian SWI/SNF complexes promote MyoD-mediated muscle differentiation. Nat. Genet. 27:187-190. [PubMed]
7. Dilworth, F. J., C. Fromental-Ramain, K. Yamamoto, and P. Chambon. 2000. ATP-driven chromatin remodeling activity and histone acetyltransferases act sequentially during transactivation by RAR/RXR in vitro. Mol. Cell 6:1049-1058. [PubMed]
8. DiRenzo, J., Y. Shang, M. Phelan, S. Sif, M. Myers, R. Kingston, and M. Brown. 2000. BRG-1 is recruited to estrogen-responsive promoters and cooperates with factors involved in histone acetylation. Mol. Cell. Biol. 20:7541-7549. [PMC free article] [PubMed]
9. Drew, H. R., and C. R. Calladine. 1987. Sequence-specific positioning of core histones on an 860 base-pair DNA. Experiment and theory. J. Mol. Biol. 195:143-173. [PubMed]
10. El Kharroubi, A., and M. A. Martin. 1996. cis-acting sequences located downstream of the human immunodeficiency virus type 1 promoter affect its chromatin structure and transcriptional activity. Mol. Cell. Biol. 16:2958-2966. [PMC free article] [PubMed]
11. Fan, H. Y., X. He, R. E. Kingston, and G. J. Narlikar. 2003. Distinct strategies to make nucleosomal DNA accessible. Mol. Cell 11:1311-1322. [PubMed]
12. Filetici, P., P. Ornaghi, and P. Ballario. 2001. The bromodomain: a chromatin browser? Front. Biosci. 6:D866-D876. [PubMed]
13. Flaus, A., and T. Owen-Hughes. 2001. Mechanisms for ATP-dependent chromatin remodelling. Curr. Opin. Genet. Dev. 11:148-154. [PubMed]
14. Fryer, C. J., and T. K. Archer. 1998. Chromatin remodelling by the glucocorticoid receptor requires the BRG1 complex. Nature 393:88-91. [PubMed]
15. Gavin, I., P. J. Horn, and C. L. Peterson. 2001. SWI/SNF chromatin remodeling requires changes in DNA topology. Mol. Cell 7:97-104. [PubMed]
16. Hai, T., and T. Curran. 1991. Cross-family dimerization of transcription factors Fos/Jun and ATF/CREB alters DNA binding specificity. Proc. Natl. Acad. Sci. USA 88:3720-3724. [PMC free article] [PubMed]
17. Hassan, A. H., K. E. Neely, and J. L. Workman. 2001. Histone acetyltransferase complexes stabilize swi/snf binding to promoter nucleosomes. Cell 104:817-827. [PubMed]
18. Hassan, A. H., P. Prochasson, K. E. Neely, S. C. Galasinski, M. Chandy, M. J. Carrozza, and J. L. Workman. 2002. Function and selectivity of bromodomains in anchoring chromatin-modifying complexes to promoter nucleosomes. Cell 111:369-379. [PubMed]
19. Havas, K., A. Flaus, M. Phelan, R. Kingston, P. A. Wade, D. M. Lilley, and T. Owen-Hughes. 2000. Generation of superhelical torsion by ATP-dependent chromatin remodeling activities. Cell 103:1133-1142. [PubMed]
20. He, G., and D. M. Margolis. 2002. Counterregulation of chromatin deacetylation and histone deacetylase occupancy at the integrated promoter of human immunodeficiency virus type 1 (HIV-1) by the HIV-1 repressor YY1 and HIV-1 activator Tat. Mol. Cell. Biol. 22:2965-2973. [PMC free article] [PubMed]
21. Henderson, A., M. Bunce, N. Siddon, R. Reeves, and D. J. Tremethick. 2000. High-mobility-group protein I can modulate binding of transcription factors to the U5 region of the human immunodeficiency virus type 1 proviral promoter. J. Virol. 74:10523-10534. [PMC free article] [PubMed]
22. Horn, P. J., and C. L. Peterson. 2001. The bromodomain: a regulator of ATP-dependent chromatin remodeling? Front. Biosci. 6:D1019-D1023. [PubMed]
23. Ichinose, H., J. M. Garnier, P. Chambon, and R. Losson. 1997. Ligand-dependent interaction between the estrogen receptor and the human homologues of SWI2/SNF2. Gene 188:95-100. [PubMed]
24. Ito, T., M. Yamauchi, M. Nishina, N. Yamamichi, T. Mizutani, M. Ui, M. Murakami, and H. Iba. 2001. Identification of SWI. SNF complex subunit BAF60a as a determinant of the transactivation potential of Fos/Jun dimers. J. Biol. Chem. 276:2852-2857. [PubMed]
25. Jacobson, R. H., A. G. Ladurner, D. S. King, and R. Tjian. 2000. Structure and function of a human TAFII250 double bromodomain module. Science 288:1422-1425. [PubMed]
26. Kassabov, S. R., B. Zhang, J. Persinger, and B. Bartholomew. 2003. SWI/SNF unwraps, slides, and rewraps the nucleosome. Mol. Cell 11:391-403. [PubMed]
27. Kingston, R. E., and G. J. Narlikar. 1999. ATP-dependent remodeling and acetylation as regulators of chromatin fluidity. Genes Dev. 13:2339-2352. [PubMed]
28. Kowenz-Leutz, E., and A. Leutz. 1999. A C/EBP beta isoform recruits the SWI/SNF complex to activate myeloid genes. Mol. Cell 4:735-743. [PubMed]
29. Laughlin, M. A., G. Y. Chang, J. W. Oakes, F. Gonzalez-Scarano, and R. J. Pomerantz. 1995. Sodium butyrate stimulation of HIV-1 gene expression: a novel mechanism of induction independent of NF-kappa B. J. Acquir. Immune Defic. Syndr. Hum. Retrovirol. 9:332-339. [PubMed]
30. Östlund Farrants, A.-K., P. Blomquist, H. Kwon, and O. Wrange. 1997. Glucocorticoid receptor-glucocorticoid response element binding stimulates nucleosome disruption by the SWI/SNF complex. Mol. Cell. Biol. 17:895-905. [PMC free article] [PubMed]
31. Pollard, K. J., and C. L. Peterson. 1997. Role for ADA/GCN5 products in antagonizing chromatin-mediated transcriptional repression. Mol. Cell. Biol. 17:6212-6222. [PMC free article] [PubMed]
32. Rabbi, M. F., M. Saifuddin, D. S. Gu, M. F. Kagnoff, and K. A. Roebuck. 1997. U5 region of the human immunodeficiency virus type 1 long terminal repeat contains TRE-like cAMP-responsive elements that bind both AP-1 and CREB/ATF proteins. Virology 233:235-245. [PubMed]
33. Reeves, R., and M. S. Nissen. 1999. Purification and assays for high mobility group HMG-I(Y) protein function. Methods Enzymol. 304:155-188. [PubMed]
34. Roberts, S. M., and F. Winston. 1997. Essential functional interactions of SAGA, a Saccharomyces cerevisiae complex of Spt, Ada, and Gcn5 proteins, with the Snf/Swi and Srb/mediator complexes. Genetics 147:451-465. [PMC free article] [PubMed]
35. Roebuck, K. A., D. S. Gu, and M. F. Kagnoff. 1996. Activating protein-1 cooperates with phorbol ester activation signals to increase HIV-1 expression. AIDS 10:819-826. [PubMed]
36. Sheridan, P. L., T. P. Mayall, E. Verdin, and K. A. Jones. 1997. Histone acetyltransferases regulate HIV-1 enhancer activity in vitro. Genes Dev. 11:3327-3340. [PMC free article] [PubMed]
37. Sif, S., P. T. Stukenberg, M. W. Kirschner, and R. E. Kingston. 1998. Mitotic inactivation of a human SWI/SNF chromatin remodeling complex. Genes Dev. 12:2842-2851. [PMC free article] [PubMed]
38. Steger, D. J., A. Eberharter, S. John, P. A. Grant, and J. L. Workman. 1998. Purified histone acetyltransferase complexes stimulate HIV-1 transcription from preassembled nucleosomal arrays. Proc. Natl. Acad. Sci. USA 95:12924-12929. [PMC free article] [PubMed]
39. Thanos, D., and T. Maniatis. 1996. In vitro assembly of enhancer complexes. Methods Enzymol. 274:162-173. [PubMed]
40. Van Lint, C., C. A. Amella, S. Emiliani, M. John, T. Jie, and E. Verdin. 1997. Transcription factor binding sites downstream of the human immunodeficiency virus type 1 transcription start site are important for virus infectivity. J. Virol. 71:6113-6127. [PMC free article] [PubMed]
41. Van Lint, C., S. Emiliani, M. Ott, and E. Verdin. 1996. Transcriptional activation and chromatin remodeling of the HIV-1 promoter in response to histone acetylation. EMBO J. 15:1112-1120. [PMC free article] [PubMed]
42. Varga-Weisz, P. D., E. J. Bonte, and P. B. Becker. 1999. Analysis of modulators of chromatin structure in Drosophila. Methods Enzymol. 304:742-757. [PubMed]
43. Verdin, E., P. Paras, Jr., and C. Van Lint. 1993. Chromatin disruption in the promoter of human immunodeficiency virus type 1 during transcriptional activation. EMBO J. 12:3249-3259. [PMC free article] [PubMed]
44. Vignali, M., A. H. Hassan, K. E. Neely, and J. L. Workman. 2000. ATP-dependent chromatin-remodeling complexes. Mol. Cell. Biol. 20:1899-1910. [PMC free article] [PubMed]
45. Wang, W., T. Chi, Y. Xue, S. Zhou, A. Kuo, and G. R. Crabtree. 1998. Architectural DNA binding by a high-mobility-group/kinesin-like subunit in mammalian SWI/SNF-related complexes. Proc. Natl. Acad. Sci. USA 95:492-498. [PMC free article] [PubMed]
46. Weinmann, A. S., S. M. Bartley, T. Zhang, M. Q. Zhang, and P. J. Farnham. 2001. Use of chromatin immunoprecipitation to clone novel E2F target promoters. Mol. Cell. Biol. 21:6820-6832. [PMC free article] [PubMed]
47. Widlak, P., R. B. Gaynor, and W. T. Garrard. 1997. In vitro chromatin assembly of the HIV-1 promoter. ATP-dependent polar repositioning of nucleosomes by Sp1 and NFkappaB. J. Biol. Chem. 272:17654-17661. [PubMed]
48. Winston, F., and C. D. Allis. 1999. The bromodomain: a chromatin-targeting module? Nat. Struct. Biol. 6:601-604. [PubMed]
49. Yudkovsky, N., C. Logie, S. Hahn, and C. L. Peterson. 1999. Recruitment of the SWI/SNF chromatin remodeling complex by transcriptional activators. Genes Dev. 13:2369-2374. [PMC free article] [PubMed]
50. Zhao, K., W. Wang, O. J. Rando, Y. Xue, K. Swiderek, A. Kuo, and G. R. Crabtree. 1998. Rapid and phosphoinositol-dependent binding of the SWI/SNF-like BAF complex to chromatin after T lymphocyte receptor signaling. Cell 95:625-636. [PubMed]

Articles from Molecular and Cellular Biology are provided here courtesy of American Society for Microbiology (ASM)
PubReader format: click here to try


Save items

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • Gene
    Gene records that cite the current articles. Citations in Gene are added manually by NCBI or imported from outside public resources.
  • GEO Profiles
    GEO Profiles
    Gene Expression Omnibus (GEO) Profiles of molecular abundance data. The current articles are references on the Gene record associated with the GEO profile.
  • HomoloGene
    HomoloGene clusters of homologous genes and sequences that cite the current articles. These are references on the Gene and sequence records in the HomoloGene entry.
  • MedGen
    Related information in MedGen
  • PubMed
    PubMed citations for these articles
  • Substance
    PubChem chemical substance records that cite the current articles. These references are taken from those provided on submitted PubChem chemical substance records.

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...