CTGC motifs within the HIV core promoter specify Tat-responsive pre-initiation complexes
© Wilhelm et al.; licensee BioMed Central Ltd. 2012
Received: 30 January 2012
Accepted: 26 July 2012
Published: 26 July 2012
Skip to main content
© Wilhelm et al.; licensee BioMed Central Ltd. 2012
Received: 30 January 2012
Accepted: 26 July 2012
Published: 26 July 2012
HIV latency is an obstacle for the eradication of HIV from infected individuals. Stable post-integration latency is controlled principally at the level of transcription. The HIV trans-activating protein, Tat, plays a key function in enhancing HIV transcriptional elongation. The HIV core promoter is specifically required for Tat-mediated trans-activation of HIV transcription. In addition, the HIV core promoter has been shown to be a potential anti-HIV drug target. Despite the pivotal role of the HIV core promoter in the control of HIV gene expression, the molecular mechanisms that couple Tat function specifically to the HIV core promoter remain unknown.
Using electrophoretic mobility shift assays (EMSAs), the TATA box and adjacent sequences of HIV essential for Tat trans-activation were shown to form specific complexes with nuclear extracts from peripheral blood mononuclear cells, as well as from HeLa cells. These complexes, termed pre-initiation complexes of HIV (PICH), were distinct in composition and DNA binding specificity from those of prototypical eukaryotic TATA box regions such as Adenovirus major late promoter (AdMLP) or the hsp70 promoter. PICH contained basal transcription factors including TATA-binding protein and TFIIA. A mutational analysis revealed that CTGC motifs flanking the HIV TATA box are required for Tat trans-activation in living cells and correct PICH formation in vitro. The binding of known core promoter binding proteins AP-4 and USF-1 was found to be dispensable for Tat function. TAR RNA prevented stable binding of PICH-2, a complex that contains the general transcription factor TFIIA, to the HIV core promoter. The impact of TAR on PICH-2 specifically required its bulge sequence that is also known to interact with Tat.
Our data reveal that CTGC DNA motifs flanking the HIV TATA box are required for correct formation of specific pre-initiation complexes in vitro and that these motifs are also required for Tat trans-activation in living cells. The impact of TAR RNA on PICH-2 stability provides a mechanistic link by which pre-initiation complex dynamics could be coupled to the formation of the nascent transcript by the elongating transcription complex. Together, these findings shed new light on the mechanisms by which the HIV core promoter specifically responds to Tat to activate HIV gene expression.
Latency of HIV contributes to viral persistence despite current antiviral therapies, as well as to immune evasion [1–4]. Post-integration latency is the most long-lived source of latent HIV where the host cell genome harbors a functional provirus that is transcriptionally silent [4, 5]. The HIV promoter region, located within the 5’ long terminal repeat (LTR) of the integrated viral genome, pirates the host cell RNA polymerase II (Pol II) machinery to initiate viral transcription. Transcriptional interference, chromatin structure and modification (e.g. acetylation/deacetylation of histones), DNA methylation, limitation of host factors (e.g. P-TEFb), and activation by specific host cell factors (e.g. NF-κB, NFAT1) can all contribute to the activation of HIV transcription [1, 3, 4]. The core promoter is ultimately the gateway for all signals activating HIV transcription, since it nucleates the formation of Pol II pre-initiation complex (PIC) which is a rate-limiting step in the initiation of transcription .
Once HIV transcription occurs, the HIV Tat gene product can be expressed and plays an important role in the exit from latency by driving a feed-forward loop to fully activate viral transcription . Tat acts by physically interacting with a stem-loop structure of the nascent 5’ HIV RNA termed TAR. Tat binds to TAR in conjunction with an essential cellular cofactor termed positive transcription elongation factor b (P-TEFb), composed of CDK9 and Cyclin T1 [8–10]. Recently published data suggest that nascent TAR RNA can displace promoter-bound transcription complexes containing an inhibitory small non-coding RNA (7SK) that sequesters P-TEFb in an inactive form . Once in the active, 7SK free form, P-TEFb is recruited by Tat into a complex with other cellular elongation factors and co-factors termed the super elongation complex (SEC) [12, 13]. Tat acts within SEC on Pol II, in part via phosphorylation of the c-terminal domain (CTD) of Pol II, to potently increase elongation rates of HIV transcription [14, 15]. The HIV trans-activator Tat thus drives an amplification loop during the reactivation of latent HIV.
TASHET plays a crucial role in the control of HIV gene expression and latency; yet, the cellular complexes that impart functional specificity upon it are completely unknown. The identification of specific cellular TASHET-binding complexes could potentially reveal drug targets for therapies to flush out, or definitively silence, latent virus in HIV infected individuals. Given the deepening realization that core promoter binding complexes are highly diverse [6, 30–35], together with recent advances in the experimental detection of endogenous pre-initiation complexes , we have revisited TASHET function. Here, we report the results of experiments designed to answer three critical but unresolved questions: 1) Are cellular pre-initiation complexes that recognize TASHET distinct from those binding the other eukaryotic TATA elements such as the prototypical Adenovirus major late promoter (AdMLP)? 2) Precisely what cis-acting DNA sequences are required for the formation of Tat-responsive pre-initiation complexes? 3) What physical interactions account for the functional dependence of the Tat/TAR axis on TASHET?
The availability of extensive sequencing of HIV promoters from clinical isolates provides a wealth of information about the conservation of nucleotides that are important for viral replication. To establish the sequence variation within the HIV core promoter (Figure 1A and 1B), we aligned HIV promoter sequences from the Los Alamos HIV Sequence Database (http://www.hiv.lanl.gov). We focused on nucleotide conservation corresponding to the TATA box and adjacent sequences of HIV essential for Tat trans-activation (TASHET). Nucleotides −33 to −16 with respect to the transcriptional start site displayed mostly high levels of conservation while nucleotides flanking this region displayed decreasing conservation. The conservation of most nucleotides from −33 to −16 corresponds well with the core promoter sequences defined by several independent mutational studies to be essential for Tat trans-activation [16–21, 37]. Two nucleotides within TASHET that display notably high variability include the previously noted position −26 (T or A) within the TATA box [38, 39] and position −18 (T or C) within an E-box element immediately downstream (Figure 1C). The conservation of TASHET sequences from naturally occurring HIV isolates guided our design of mutations to identify functionally important base-pairs throughout this study.
The HeLa cell model has been employed as a valuable model to identify cellular transcription factors that control HIV transcription such as SP1 , NF-κB , and P-TEFb . To examine the suitability of the HeLa cell system for the study of cellular complexes binding to TASHET, we isolated nuclear extracts from activated peripheral blood mononuclear cells (PBMC) from healthy donors as these contain the physiological targets for HIV infection CD4+ lymphocytes and monocytes. EMSA assays were performed with PBMC nuclear extracts and revealed complexes of indistinguishable mobility and DNA binding specificity compared to those from HeLa cell nuclear extracts (Figure 3, panel B, lanes 1–6 versus panel A lanes 1–6). The similarities between TASHET-binding complexes from HeLa cells and PBMC support the suitability of HeLa cells as a model system for the study of host cell TASHET-binding complexes.
The presence of general Pol II transcription factors within PICH cannot by itself account for TASHET's specific capacity to respond to Tat [16–21, 37]. Given reports that the E-box immediately downstream of the TATA box may be important in the response to Tat , we tested for the presence of two bHLH factors, AP4  and USF-1 , that have been reported to bind to this E-box. Antibodies directed against AP4 resulted in the loss of a minor TASHET-interacting complex (Figure 4A, lane 9 versus lane 7 see also Additional file 1, panel B), in agreement with previous findings of AP4 interacting with the 3’ E-box . The addition of USF-1 antibodies resulted in the loss of a complex that binds specifically to the TASHET (Figure 4A, lane 8 versus lane 7). Thus PBMC nuclear extracts contain the bHLH factors USF-1 and AP4 both of which are able to bind to TASHET in EMSA.
To test the effects of the mutations that altered PICH formation in vitro on Tat trans-activation in living cells, we again employed co-transfection of reporter constructs carrying the full length HIV LTR driving luciferase with a Tat expression vector in HeLa cells. Co-transfection of a construct bearing point mutations within the single 5’ CTGC increased basal HIV LTR-directed transcription (Figure 8C, mutation CTGC5’) without significantly increasing Tat trans-activation (Figure 8D). This result mirrors the previously reported effects of mutations immediately 5' of the TATA box . Mutation of the two 3’ CTGC motifs resulted in a significant drop in Tat trans-activation (Figure 8D, mutation CTGC3’) without significantly changing basal transcription (Figure 8C). A construct bearing mutations in all three flanking CTGC motifs resulted in highly impaired Tat trans-activation (Figure 8D, mutation CTGC5’3’), yet only slightly decreased basal HIV LTR-driven transcription (Figure 8C). To rule out indirect effects on Tat trans-activation by grossly altered transcriptional start sites, the start sites of transcription for mutated core promoters were analyzed by primer extension in transfected HeLa cells. The major start site of transcription was found to be unchanged with all mutations used (Additional file 5). We concluded that the CTGC DNA motifs flanking the TATA box are essential for Tat trans-activation of the HIV promoter.
To define the impact of TASHET point mutations on proviral HIV transcription in PBMCs, we isolated the cells and activated them with mitogens IL-2 and PHA before infection with viral particles bearing point mutations in the 3’ LTR that are subsequently copied into the 5’ LTR by HIV reverse transcriptase before integration into the host cell genome (Figure 9A). Luciferase activity was measured to monitor HIV gene expression forty-eight hours post infection. An HIV genome with a mutated TATA box served as a negative control and resulted in very low levels of luciferase activity (Figure 9B, CATAKO). Point mutations that increase USF-1 binding (Figure 6) decreased HIV expression to approximately 70% of wild type levels (Figure 9B, USF+). Point mutations that prevent USF-1 binding but retain PICH-2 binding (Figure 6) resulted in Tat-activated HIV gene expression that was measurably higher than that of wild type promoter (Figure 9B, USF1KO). We conclude that in the single round infection assay in PBMC USF-1 binding is not essential for activated HIV gene expression. Mutation of the 5’ CTGC motif alone had little effect on HIV gene expression and even slightly increased luciferase activity in infected PBMC (Figure 9B, CTGC5’). Mutations within the two 3’ CTGC motifs reduced HIV gene expression by approximately one half (Figure 9B, CTGC3’). Importantly, the point mutation of the three flanking CTGC motifs reduced HIV expression levels strongly showing that, in the proviral chromatin context and in primary PBMC, these motifs are essential for activated HIV gene expression.
The data we present here show for the first time that the TATA box of HIV and adjacent sequences of HIV essential for Tat trans activation (TASHET) is recognized by cellular pre-initiation complexes (PICHs) that: 1) are distinct from canonical PIC that recognize the model AdMLP, 2) require the flanking CTGC motifs for their accurate formation, and 3) include PICH-2 whose stable binding to TASHET is disrupted by HIV TAR RNA. TASHET has been shown to play a pivotal role in governing Tat-activated HIV transcription in cell culture [16–21, 37] and by logical extension is thought to impact the dynamics of latency in vivo. Moreover, previous studies have provided a proof-of-principle that DNA sequences within the TASHET element of the HIV core promoter can be specifically targeted by polyamides to modulate HIV transcription [29, 53, 54]. The host cell PICH complexes identified here are of importance as an essential step in the activation of HIV transcription, and also as a source of potential new drug targets for the eradication of HIV from infected individuals.
The functional diversity of core promoters and of the transcription factors that they bind is increasingly recognized as an important contributor to genomic regulation [6, 30–35]. Of the known consensus core promoter elements (DPE, DCE, BRE, XCPE1, etc.), the HIV promoter contains only an identified TATA box  and a non-classical initiator (Inr) element . In contrast to TASHET [16–21], the HIV Inr can be replaced by a heterologous AdMLP Inr for Tat-responsive transcription . The data presented here excluded an essential role for USF-1 binding to the 3' E box in Tat trans-activation (Figure 6). Two other CTGC motifs were found in upstream HIV LTR motifs termed RBEIV and RBEIII elements , but were not required for Tat trans-activation . The 3’ CANNTG E-box is embedded within the 3’ palindromic GCAGCTGC motif (Figure 6A). The 3’ E-box displays relatively low sequence conservation [22, 39] (Figure 1C). Our data could help explain why a GCAGCCGC variant is frequently found in natural HIV isolates (Figure 1C), yet weakens the consensus E-box GCAGCTGC, if positive selection pressure to maintain PICH-2 - CTGC contacts can prevail over selection to maintain USF-1/AP4 - E-box interactions in HIV infected individuals.
We postulate two mechanisms to account for the specific functional and DNA binding properties displayed by the PICH described herein. First, the known Pol II general transcription factors (GTFs) could have unique affinities or conformations when bound to TASHET that confer functional specificity. There are precedents for core promoter specific GTF function, for example TFIIA, that we have found as a component of PICH-2, has been shown to have positive or negative effects on transcription depending on the core promoter sequence . A second alternative is that unknown accessory (non-GTF) host cell proteins could bind to TASHET to confer Tat-responsiveness. The two possible mechanisms are not mutually exclusive and could contribute together to PICH specificity. Certain TBP-associated factors (TAFs) can recognize core promoter elements [32, 60]. Our supershift analysis suggests that stable PICH formation on TASHET requires at least some core TAFs. The PICH appear to bind less tightly to TASHET compared to canonical PIC to the AdMLP, since antibodies to TFIID subunits disrupt PICH – TASHET interactions more readily than PIC – AdMLP interactions in EMSA (Additional file 1). These results are compatible with chromatin immunoprecipitation (ChIP) results obtained from Tat-expressing cells showing that TAF occupancy of the HIV core promoter is lower than that of TBP when compared to an adenovirus E1b core promoter . Nevertheless, both TFIID  and TAFs  have been shown to bind the HIV core promoter in vitro. The ratios of TBP/TFIID occupancy measured by ChIP must be interpreted in light of the fact that association of TBP with TFIID is highly dynamic both in vitro and in living cells on model promoters .
Cellular complexes have been previously reported to bind TASHET [37, 57, 66], but their identity has remained enigmatic due to their enormous size, subunit complexity, and their dynamic nature on core promoters. The minimal set of classical GTFs required for PIC formation and transcription is considered to be 70 polypeptides , and proteomic studies of yeast PIC composition imply that hundreds of proteins are involved . TASHET’s key role in HIV transcription was discovered more than two decades ago , yet the complexes recognizing it have remained out of reach. Based on the GTFs they contain, PICH-1 (e.g. TBP, TAFs), PICH-2 (TBP, TAFs, TFIIA) and PICH-3 (TBP, TAFs, Pol II) likely correspond to intermediates in the PIC assembly pathway . The availability of a tractable EMSA to detect PICH, the demonstration that PICH contain classical GTFs yet are distinct from canonical PIC, together with the definition of the nucleotides essential for their formation, opens the door for molecular genetic, biochemical and proteomic dissection of PICH composition.
Our data show that TAR RNA, via its bulge structure, can prevent stable association of PICH-2 with TASHET in EMSA (Figure 10). In principle TAR could act by blocking association of PICH-2 with TASHET or by accelerating the dissociation of PICH-2 from TASHET. We favour a role for TAR in the dissociation from TASHET because PICH-2 binding correlates positively with Tat-responsiveness (Figure 6, and data not shown), and because TAR decoys do not inhibit PIC formation in vitro. We propose a dynamic model in which cellular PICH specifically recognize TASHET via its CTGC motifs. Nascent TAR transcribed by Pol II facilitates the departure of PICH-2, in turn accelerating the next cycle of PIC formation (Figure 11B). This model is compatible with the model recently proposed by D’Orso et al. including the expulsion of 7SK snRNP from the HIV promoter by TAR . A dynamic model is also compatible with the observation that point mutations to the CTGC motifs result in aberrant but stable aPIC formation and a failure to respond to Tat (Figure 8). The presence of AP4 in aPIC was unexpected since the 3’ E-box is destroyed by these mutations. A plausible explanation is that AP4 may be recruited into non-productive PICH via protein-protein interactions. The dynamic model we propose for the impact of both PICH association and disassociation on Tat trans-activation is consistent with previous genetic observations that appropriate PIC destabilization is necessary for activated transcription in yeast  and recent reports showing that transcription factor dynamics are important in transcriptional regulation [77, 78].
Our results shed new light on the mechanisms that control HIV gene expression, but also have broader implications for the combinatorial control of cellular gene expression. Biologically, the sequence of core promoters has been shown to dictate: 1) the differential response to activators , 2) the response to cell-type specific enhancers , and 3) the alternative splicing of the transcribed pre-mRNA . The composition of core promoter-binding complexes is increasingly recognized as being very heterogeneous [31–33, 82]. In addition, the range of core promoter cis-acting DNA sequences is also known to be highly complex [6, 60]. To our knowledge, to date no biochemical data had been reported showing that distinct core promoters bind PIC that are functionally distinct, yet contain common PIC components such as TBP and Pol II. The data presented here link a specific biological outcome (Tat trans-activation), to a core promoter sequence (the CTGC motifs) and also to a specialized core promoter-binding complex (PICH-2). The broader mechanistic implication is that core promoter complexes, despite the fact that they contain common classical GTFs, can confer very distinct transcriptional responses.
The regulation of HIV transcription dictates viral latency versus active replication. The viral Tat trans-activating protein is essential for activated HIV transcription. A long-standing unanswered question concerning the mechanism of Tat trans-activation is: why can the HIV core promoter TATA box region (TASHET) not be functionally replaced by heterologous TATA elements? We report here that CTGC DNA motifs in the HIV core promoter are essential for the formation of specialized Tat-responsive pre-initiation complexes (PICH). PICH-2 contains the general transcription factor TFIIA and its stable association with TASHET is prevented by the presence of HIV TAR RNA. The detection of Tat-responsive PICH binding complexes provides an essential step forward towards the full elucidation of the mechanisms underlying activated HIV transcription, and paves the way for the identification of new molecular targets for therapies to eradicate latent HIV.
Antibodies used in this study were raised against: AP-4 (sc-18593), Med6 (sc-9434), TFIIA-γ (sc-5316), USF-1 (sc-229), all from Santa Cruz, CA; Pol II (8WG16) from Covance (Emeryville, CA); SMARCA3 (BL825) from Bethyl Laboratories (Montgomery, TX); TFII-I (#4562) from Cell Signaling (Beverly, MA). Monoclonal antibodies against TFIID subunits were generous gifts from Dr. Laszlo Tora and have been described [83–85] : TBP : TBP-1 (2 C1), TBP-2 (4 C2), TBP-3 (3 G3) ; TAF4 (20TA) ; TAF5 (1TA) ; TAF6 (25TA) ; TAF10 (2B11) ; TAF15 (2B10).
pCMV-Tat has been described . To obtain the negative control lacking Tat coding sequence, pCMV-Tat was linearized with XhoI and SalI restriction enzymes and blunted with the Klenow enzyme before religation. pHIV-RL is based on the pHRL-null vector (Promega #E6231) whose XbaI and SphI sites were eliminated by restriction enzymes, Klenow digestion and religation. The HIV-1 LTR of pU3S , derived from pHIVSCAT , was cloned into the XhoI and HindIII sites of the previously modified pHRL-null. The resulting plasmid was digested with XhoI and KpnI and treated with Klenow enzyme before religation to eliminate the 5' XbaI site of the LTR. All the Renilla luciferase mutant constructs were based on pHIV-RL, by replacing the wild-type core promoter sequence between the XbaI and SphI sites with 35 bp synthetic oligonucleotides bearing the various mutations. Key mutations have been cloned in pNL4-3-LucE- (kindly provided by Michel J. Tremblay, Université Laval, Québec) for pseudotyped virus production. To facilitate the mutation, the XhoI-NcoI portion of pNL4-3-LucE- has been cloned into phRLnull and directed mutagenesis was performed by PCR as previously described . The XhoI-NcoI portion was then reintroduced into pNL4-3-LucE- by standard subcloning. The envelope encoding plasmid pCMV-VSV-G was also a gift of Dr. M.J. Tremblay. Mutated TAR RNAs based on previously reported mutations [49–51] were cloned into the SphI and SacI sites of the pHIV-RL backbone using synthetic oligonucleotides bearing mutations. In the ΔTAR mutant, 30 nucleotides were deleted in TAR sequence by SacI-HindIII digestion, Klenow blunting and religation.
HeLa cells (ATCC # CCL-2) were grown in DMEM supplemented with 2.5% FBS and 2.5% NBCS (Wisent). HEK293 cells (ATCC #CRL-1573) were grown in DMEM containing 10% FBS. Peripheral blood mononuclear cells (PBMC) were isolated from healthy donors by lymphocyte separation medium (Wisent) according to manufacturer’s instructions and stimulated 3 days with PHA-L (1 μg/ml, L2769 Sigma) and IL-2 (30 U/ml, Sigma or F081 Bioshop) before viral infection or nuclear extracts preparation. Nuclei were prepared from activated PBMC  and HeLa cells  as previously described. PBMC and HeLa cell nuclei were then used to prepare nuclear extracts according to the protocol of Dignam et al. .
To generate TAR transcripts, the sequence was first amplified by PCR using a sense primer containing a T3-polymerase recognition site. Three 50 μl reactions were prepared, each containing 200 ng of matrix plasmid, 1 mM of each dNTPs, 400 nM of sense and reverse primer, 2.5 U of Pfu turbo enzyme in 1x reaction buffer (NEB, Ipswich MA). PCR products were pooled and purified by phenol-chloroform extraction and ethanol precipitation. The obtained template was reverse transcribed for 1 to 2 h at 37 °C in a mix containing 0.5 mM of each rATP, rCTP and rUTP, 0.1 mM rGTP, 0.2 mM Cap analog, 12.5 pmol (10 μCi) of α-32P UTP, 2 μl RNA guard, 34U of T3 RNA-polymerase in 100 μl of 1X transcription buffer (40 mM Tris–HCl pH7.9, 6 mM MgCl2, 2 mM spermidine, 10 mM DTT). The transcription product was mixed with formamide dye (1 mg/ml of each bromophenol blue and xylene cyanole, 10 mM EDTA in deionized formamide), boiled for 90 seconds and immediately chilled on ice, then loaded on a pre-equilibrated denaturing 8 M urea- 4.75% polyacrylamide gel in TBE 1X and run for 45 min at 300 V. Full length TAR transcripts were identified by autoradiography and then excised from the gel for extraction with 300 μl crunch solution (300 mM Na Acetate, 0.2% SDS) on a rocking table twice for 20 min. Pooled supernatants were phenol-chloroform extracted twice, ethanol precipitated, resuspended in water and the radioactive counts were measured for the calculation of RNA concentration.
The specific EMSA protocol used in this study has been described in detail . Briefly, 20 μg nuclear extracts were mixed together with 4μg of acetylated BSA (Promega #R3961), 2 μg of Poly(dI-dC)·Poly(dI-dC) (Sigma) and 1.8pmol of nonspecific double-stranded oligonucleotide (sense : GATCCGGAGTACTTCAAGAACG; reverse : GATCCGTTCTTGAAGTACTCCG) in a final volume of 20 μl of binding buffer (20 mM Hepes, 5 mM MgCl2, 8% glycerol, 100 mM KCl), for 5 minutes on ice. For competition assays, TAR RNA or 200pmol (unless otherwise specified) of unlabelled competitor oligonucleotide were added to the reaction. 1pmol of labelled doubled-stranded promoter (35 bp) were added to reaction for 15 minutes at room temperature. For supershift assays, 1 to 4 μl of specific antibody were added to the sample, and the reaction was continued on ice for 1 additional hour. Samples were then loaded on a native 4.5% polyacrylamide gel and complexes separated for 3.5 hours at 150 Volts. For RNase H directed degradation of 7SK RNA within the nuclear extracts used in EMSA, nuclear extracts were first incubated with anti-sense oligonucleotide and 10 U of RNase H (NEB) for 1 h at 30°C in binding buffer. The remaining constituents were then added as indicated above (BSA, Poly(dI-dC)·Poly(dI-dC), non specific oligonucleotide and finally radiolabelled oligonucleotide) and the reaction was allowed to continue as described above. The anti-sense oligonucleotides for 7SK RNase H directed digestion have been described .
To check for proper and specific RNA degradation, the EMSA reaction was doubled, and RNAs were recovered from one half of the resulting volume by trizol extraction. The samples were DNase I treated (Promega) for 30 min at 37°C. RNAs were reverse transcribed by MMuLV-RT (Roche) using random hexamer-primers according to manufacturer’s recommendations. PCR was performed on 1/10 of the obtained cDNA to amplify 7SK-snRNA : forward GGATGTGAGGCGATCTGGC ; reverse : AAAAGAAAGGCAGACTGCCAC; or U6snRNA : forward CTCGCTTCGGCAGCACATATAC ; reverse GGAACGCTTCACGAATTTGCGTG.
For Renilla Luciferase reporter assays, HeLa cells were transfected in 96 well plates with DMRIE-C (Invitrogen,) according to the manufacturer’s recommendations. 12000 cells were seeded into 96 well plates the day before transfection in complete medium. 50 ng of pHIV-RL construct, together with 15 ng of either pCMV-Tat or the empty control vector were transfected, complexed with 0.2 μl of DMRIE-C transfection reagent in a final volume of 100 μl of OptiMEM (Invitrogen). Two days later, the Renilla Luciferase assay system (# E2810, Promega, Madison, WI) was used according to the manufacturer’s instructions to lyse and measure luminescence on an automated luminometer (BMG Labtech, Ortenberg, Germany).
HeLa cells were transfected in 6 well plates with DMRIE-C (Invitrogen,) according to the manufacturer’s recommendations. 5 X 105 cells were seeded per well the day before transfection in complete medium. 1.25 μg of pHIV-RL construct, together with 250 ng of either pCMV-Tat or the empty control vector were transfected, complexed with 4 μl of DMRIE-C transfection reagent in a final volume of 1.5 ml OptiMEM (Invitrogen). RNAs were extracted with Trizol (Invitrogen) reagent after 24 h. 5 μg of total RNA was used in primer extension reactions for 1 h at 42 °C in a buffer containing 1.25 mM Tris pH 8.0, 1.75 mM KCl, 5 mM DTT, 10 mM MgCl2, 125 μM of each dNTP, 25 μg/ml Actinomycin D, 5U of AMV Reverse-Transcriptase (Roche) and 20 ng of a PNK radiolabelled reverse primer specific for TAR (5’ GCTTTATTGAGGCTTAAGCAGTG3’). The ethanol precipitated pellet was resuspended in formamide dye, boiled and loaded on an 8 M urea 9% polyacrylamide denaturing sequencing gel.
The day before transfection, 4 X 106 HEK-293 packaging cells were seeded on 100 mm dishes in 9 ml complete medium. Calcium phosphate transfection was performed by mixing together 1.5 μg pVSV-G and 13.5 μg pNL4-3-LucE- based constructs in 500 μl of 250 mM CaCl2 and 500 μl HBS 2X (280 mM NaCl, 50 mM Hepes, 1.5 mM Na2HPO4, pH7.08). The precipitates were allowed to form for 2 minutes and immediately added drop-wise on top of cell culture medium. Two days later, the virion containing medium was recovered and filtered through a 0.45 μm filter and kept frozen at −80°C. Virion content was estimated using an in house enzyme-linked immunosorbent assay (ELISA) for the viral major core protein p24 that has been previously described .
Oligonucleotide sequences for EMSA and pHIV-RL constructs
Adenovirus major late promoter
Aberrant pre-initiation complex (of RNA polymerase II)
Electrophoretic mobility shift assay
Peripheral blood mononuclear cells
Pre-initiation complex (of RNA polymerase II)
Pre-initiation complex of HIV
RNA polymerase II
Relative light units
Small nuclear ribonucleoprotein complex
TATA binding protein
TATA box of HIV and Adjacent Sequences of HIV Essential for Tat trans-activation
We thank B. Cullen (pCMV-Tat), M. Peterlin (pHIVSCAT), and M. Tremblay (pNL4-3-LucE-, pCMV-VSV-G) for providing plasmids. We thank C. Fortin and P. MacDonald for providing purified PBMC for nuclear extracts, L. Tora for antibodies against TFIID subunits, and A. Leclerc for assistance with illustrations. B. Bell is a member of the FRSQ-funded Centre de recherche clinique Étienne-Le Bel. We gratefully thank an anonymous donor for funding. This work was funded in part through the award of a NSERC Discovery grant to B. Bell.