Cell Cycle-Regulated Transcription by the Human Immunodeficiency Virus Type 1 Tat Transactivator

Abstract

Cyclin-dependent kinases are required for the Tat-dependent transition from abortive to productive elongation. Further, the human immunodeficiency virus type 1 (HIV-1) Vpr protein prevents proliferation of infected cells by arresting them in the G₂ phase of the cell cycle. These findings suggest that the life cycle of the virus may be integrally related to the cell cycle. We now demonstrate by in vitro transcription analysis that Tat-dependent transcription takes place in a cell cycle-dependent manner. Remarkably, Tat activates gene expression in two distinct stages of the cell cycle. Tat-dependent long terminal repeat activation is observed in G₁. This activation is TAR dependent and requires a functional Sp1 binding site. A second phase of transactivation by Tat is observed in G₂ and is TAR independent. This later phase of transcription is enhanced by a natural cell cycle blocker of HIV-1, vpr, which arrests infected cells at the G₂/M boundary. These studies link the HIV-1 Tat protein to cell cycle-specific biological functions.

Progression through the cell cycle requires an ordered array of biochemical interactions, illustrated by the cyclin-Cdk complexes and their Cdk-inhibitory regulators (45, 56, 65). There is increasing evidence that regulation of basal (26, 81, 82, 86) and upstream activator transcription factors (1, 11, 33, 79) also plays an important role in the coordinated cascade of the cell cycle. For example, the activity of the cellular transcription factor E2F is tightly regulated by the Rb protein. In early G₁, the interaction of Rb with E2F inactivates the transcription factor. In late G₁, the cyclin D-Cdk complex phosphorylates Rb, triggering its release from the complex. Activated E2F then increases the transcription of genes which are critical for entry into the S phase, including the family of E2F genes (17, 53, 55, 85).

The human immunodeficiency virus type 1 (HIV-1) Tat protein is required for viral replication and is a potent stimulator of viral transcription (41). Following activation of quiescent viral promoters by mitogenic stimulation, presumably resulting in an increase in transcription factors such as NF-κB, Tat stimulates viral transcription through a unique TAR RNA enhancer (5–8, 14, 16, 18, 19, 22, 27, 29, 36, 41–44, 47, 50, 51, 58, 74, 75, 83, 90). In addition, Tat has been reported to activate a number of cellular promoters (9, 10, 62, 64, 77, 80). In carrying out its transcriptional activation, Tat interacts with a number of cellular transcription factors, including TFIID (TBP) (39, 40, 78), TFIID-associated TAFs (13), TFIIH (57), Tat-associated protein (TAP) (87), Tat-associated kinase (TAK/pTEFb) (20, 21, 30–32, 48, 89, 92), TTK (54), Sp1 (25), and SF-1 (91), as well as TAR RNA (84). It has also been reported that Tat regulates the binding of RNA polymerase II to the TAR RNA (84).

Interaction of Tat with TFIID and TFIIH, both of which may be involved in cell cycle regulation (2, 69–72, 76, 86), prompted us to ask whether Tat transactivation is regulated during the cell cycle. The hypothesis that Tat's physical and functional interaction with a number of basal and upstream factors which may be cell cycle regulated is of further interest in view of the fact that HIV-1 encodes a conserved gene, vpr, that blocks infected cells at the G₂/M boundary by inhibiting p34cdc2 activation. This block may be important for efficient HIV-1 replication, since vpr mutants slow viral replication (4, 37, 59, 60). In this report, we present evidence that Tat activates transcription at two distinct phases of the cell cycle, G₁ and G₂. Moreover, G₁ Tat transactivation is TAR dependent, whereas G₂ Tat transactivation is TAR independent. We hypothesize that G₁ transcription is important for viral mRNA and genomic RNA synthesis, whereas G₂ transcription may activate some of the cellular genes (coding for cytokines) important for activation of neighboring cells subsequent to virus infection.

MATERIALS AND METHODS

In vitro transcription.

HeLa whole-cell extract (25 to 50 μg total) (49) was added to reaction mixtures containing either long terminal repeat (LTR)-TAR⁺ template, LTR-TAT−template, or a construct with Sp1 deleted (41, 66) linearized with EcoRI (100 ng). In vitro transcription reactions were incubated for 1 h at 30°C and contained the nucleoside triphosphates ATP, GTP, and CTP at a final concentration of 50 μM and [³²P]UTP (20 μCi; 400 Ci/mmol; Amersham) in buffer D (10 mM HEPES [pH 7.9], 50 mM KCl, 0.5 mM EDTA, 1.5 mM dithiothreitol, 6.25 mM MgCl₂, and 8.5% glycerol). Transcription reactions were terminated by the addition of 20 mM Tris-HCl (pH 7.8), 150 mM NaCl, and 0.2% sodium dodecyl sulfate (SDS). The quenched reactions were extracted with equal volumes of phenol-chloroform and precipitated with 2.5 volumes of ethanol and 0.1 volume of 3.0 M sodium acetate. Following centrifugation, the RNA pellets were resuspended in 12 μl of formamide denaturation mix containing xylene cyanol and bromophenol blue, heated at 90°C for 3 min, and electrophoresed at 400 V in a 4% polyacrylamide (19:1 acrylamide-bisacrylamide) gel containing 7 M urea (prerun at 200 V for 30 min) in 1× Tris-borate-EDTA. The gels were analyzed with the Molecular Dynamics PhosphorImager screen. Radioactivity was quantitated with the ImageQuant program.

Generation of epitope-tagged Tat-expressing cell line.

HeLa CD4⁺ cells (a generous gift of Bruce Chesebro, National Institutes of Health, Rocky Mountain Laboratories, Hamilton, Mont.) were used for transfection with either an epitope-tagged (the influenza epitope at the C terminus of Tat 1-86) plasmid or the parental vector pCEP4. Following transfection, cells were selected under 200 μg of hygromycin/μl. Hygromycin-resistant lines established from single-cell clones were maintained for up to 6 months with continuous passage and used to make extracts for in vitro transcription analysis.

Transfections and CAT assays.

Wild-type and TAR mutant constructs were electroporated into pCEP4 and etat cells as described previously (38). Extracts were prepared 18 h later for chloramphenicol acutyltransferase (CAT) assay. The cells were harvested, washed once with phosphate-buffered saline (PBS) without Ca²⁺ and Mg²⁺, pelleted, and resuspended in 150 μl of 0.25 M Tris (pH 7.8). The cells were freeze-thawed three times with vortexing and then incubated for 5 min at 68°C followed by centrifugation. The supernatants were transferred to 1.5-ml Eppendorf tubes. After one final spin, the supernatant was again transferred to 1.5-ml Eppendorf tubes, and the protein concentration was determined. CAT assays were performed with 2 μg of protein according to the method of Gorman et al. (25).

For vpr experiments, both etat and control cells (3 × 10⁷/time point) were transfected with HIV-1-LTR-vpr (15 μg) (Mike Emerman, Fred Hutchinson Cancer Research Center) and pSV2-neo (5 μg). Zero-hour samples were processed immediately after electroporation for cell sorting and in vitro transcription analysis. Twenty-four-hour samples were electroporated with vpr and neo plasmids, followed by selection with hygromycin (200 μg/ml) and G418 (100 μg/ml). Two sets of templates (100 ng each), EcoRI-linearized HIV-1 wild type and TM26 (66) and HIV-1 wild type and Sp1⁻ (41) were used for in vitro transcription assays (50 μg of total cellular protein).

Immunoprecipitation assays.

Immunoprecipitations were performed as described previously (39). Cellular protein (100 μg) was mixed with monoclonal antibody (2.5 μg) for 2 h at 4°C. Protein A + G agarose beads (5 μl; Calbiochem, Inc.) were added and incubated at 4°C for another 2 h. The immunoprecipitated complex was then spun down and washed with buffer D containing 500 mM KCl (three times; 1 ml each). Where indicated, immunoprecipitated proteins were eluted with the influenza virus tag peptide (88). The supernatant was retained for in vitro transcription analysis.

Cell cycle analysis.

The etat or control cells were either blocked with hydroxyurea for 18 h or blocked with hydroxyurea (2 mM final concentration), washed, and released for 1 h, followed by addition of nocodazole (50 ng/ml) for 14 h. Following the block, the cells were washed with PBS (2×) and released with complete medium. Samples were collected every 3 h, and the cells were used to make whole-cell extracts (5 × 10⁷ cells/time point) for in vitro transcription or Western blot analysis or processed for fluorescence-activated cell sorter (FACS) sorting. Single-color flow cytometric analysis of DNA content was performed on both etat and control cell lines. The cells were washed with PBS, and approximately 2 × 10⁶ were fixed by the addition of 500 μl of 70% ethanol. The cell pellets were washed with PBS (three times; 10 ml each time) and incubated in 1 ml of PBS containing 150 μg of RNase A (Sigma)/ml and 20 μg of propidium iodide (Sigma)/ml at 37°C for 30 min. The stained cells were analyzed for red fluorescence (FL2) on a FACScan (Becton Dickinson), and the distribution of cells in the G₁, S, and G₂/M phases of the cell cycle was calculated from the resulting DNA histogram with Cell FIT software, based on a rectangular S-phase model (FAST Systems, Inc., Gaithersburg, Md.).

RESULTS

Construction of epitope-tagged Tat cell line.

HeLa cells (12) were transfected with either the backbone control plasmid (PCEP4) or a hemagglutinin epitope-tagged Tat plasmid (etat/PCEP4) at the C terminus. HeLa cells were chosen for these studies because Tat transactivation can be reproduced in vitro and in vivo with these cells. T-lymphocyte cell lines, which were also considered, could not be developed, presumably due to the apoptotic effect of Tat in T cells (52, 63, 67). Two HeLa cell lines, containing either the control or etat plasmid, were successfully selected by single-cell dilution. Both cell types were selected and maintained under 200 μg of hygromycin/ml. The control PCEP4 HeLa line is designated the “control” cell line, and the etat/PCEP4 HeLa line is designated “etat” throughout the text.

To determine if the endogenous Tat was active, etat and control cells were transfected with the wild-type HIV-1 LTR CAT construct (pLTR-CAT) or a TAR mutant, pLTR TM26-CAT (66). As a further control for transfection efficiency, the cells were also transfected with pSV2-CAT. At 18 h posttransfection, cell extracts were prepared and assayed for CAT activity with 2 μg of protein. The results of this study demonstrate that the endogenous Tat protein is functional in the etat cell line. The pLTR-CAT, but not the pLTR TM26-CAT, promoter is activated in the etat cell line (Fig. 1A, lanes 2, 3, 5, and 6). A 25- to 50-fold increase in HIV-1 LTR promoter activity is routinely observed in the etat cell line. Importantly, transcription from the simian virus 40 promoter as assayed by pSV2-CAT activity was similar in both cell lines (Fig. 1A, lanes 1 and 4). These studies demonstrated that the endogenous Tat protein was able to transactivate the wild-type HIV-1 LTR in the etat cells. It is important to point out that five independent clones each of the PCEP4 and etat cell lines were tested. All PCEP4 and etat clones had similar growth properties, Tat expression, and Tat activities. One clone of PCEP4 and etat were selected for the subsequent experiments.

FIG. 1.

Functional assay of HIV-1 etat and control cell lines. (A) Transient assays. Five micrograms of wild-type HIV-1 LTR (LTR-TAR⁺) (LTR-CAT), a TAR mutant (LTR-TAR⁻) (LTR TM26-CAT), or control pSV2-CAT was electroporated into each cell type (38), and CAT assays were performed with extracts prepared 24 h posttransfection. (B) In vitro transcription. Whole-cell extracts from the control or etat cell line were used for in vitro transcription assay of either HIV-1 LTR-TAT⁺ (WTLTR) or HIV-1 LTR-TAT⁻ (TM26) templates linearized with EcoRI (40). Twenty-five micrograms of extract was used. Arrow, RNA transcript. (C) (Top) Immunodepletion of etat cell extracts with 12CA5 or Tab172 antibody. Pellets from the immunoprecipitates (lanes 1 and 2) or depleted extracts (lanes 3 and 4) were analyzed by Western blotting with polyclonal immunoglobulin G purified α-Tat antibody. (Bottom) Supernatants from extracts (25 μg) immunodepleted with either 12CA5 (lanes 1 and 3) or Tab172 (lanes 2 and 4) antibody (2.5 μg) were used for in vitro transcription of the HIV-1 LTR-TAR⁺ template. (D) Supernatants from extracts (25 μg) immunodepleted with either 12CA5 or Tab172 antibody (2.5 μg) were used for in vitro transcription of the AdML template.

As further controls for the experiment, we selected transfected cells by using an affinity antibody to interleukin-2R. The number of cells expressing interleukin-2R were similar in the etat and pCEP cell lines, suggesting a similar transfection efficiency. Moreover, the CAT assay results presented in Fig. 1A were reproduced. The wild-type HIV LTR, but not the TM26 mutant, was specifically transactivated in the eTat cell line.

In vitro transcription from epitope-tagged cell line extracts.

We next prepared whole-cell extracts from control and etat cells for in vitro transcription assays. pLTR-CAT and pLTR TM26-CAT reporter plasmids were digested with the restriction enzyme EcoRI to generate a template which would yield an RNA runoff transcript of 330 bases (40). Equivalent amounts of the templates were added to in vitro transcription reaction mixtures. Extracts were carefully titrated to optimize for LTR transcription. Similar to the results obtained in vivo, the wild-type HIV-1 LTR, but not the TAR⁻ template, was transactivated in the in vitro transcription assay with etat extracts (Fig. 1B, lanes 1 and 2). The level of transcription from the wild-type LTR template was approximately 10-fold above that observed with the TM26 TAR mutant template. In contrast, the levels of transcription from the two templates were equivalent in the control cell extracts (Fig. 1B, lanes 3 and 4). Moreover, the levels of transcription observed in the “nontransactivated” samples were equivalent, demonstrating that the basal levels of transcription were similar in the two extracts (Fig. 1B, compare lanes 1, 3, and 4). Similar results were obtained with HIV wild-type, TAR mutant, and AdML G-free cassette templates (data not shown). Control transcription reactions which included a Pol II (α-amanitin) or Pol III (tagetoxin) inhibitor demonstrate that the transcription is Pol II dependent.

To demonstrate that the increase in HIV-1 LTR transcription was in fact due to Tat, etat extracts were cleared by using the monoclonal antibodies 12CA5 (anti-epitope-tagged Tat) and Tab172 (anti-Tax control). Western blot analysis of the immunoprecipitates from the clearing experiments demonstrated that Tat was specifically immunoprecipitated with the anti-epitope antibody (12CA5) but not the control antibody (Fig. 1C, top, lanes 1 and 2). Further, Western blot analysis of the extracts from the clearing experiments demonstrated that Tat was, in fact, depleted from the extract (Fig. 1C, lanes 3 and 4). When the extracts were used for transcription assays with the HIV template, the 12CA5 antibody cleared transactivation activity whereas the control monoclonal antibody failed to inhibit transcription (Fig. 1C, bottom, lanes 1 and 2). These results suggest that transactivation is a direct effect of Tat, since depletion of Tat from the extract abolishes transcription. It was important to demonstrate that the decrease in transcription was not due to clearance of important transcription factors from the extract. A control experiment, utilizing the AdML template, is presented in Fig. 1D. The results of this experiment demonstrate that while immunoprecipitation of Tat from the extract decreases HIV transcription, the level of AdML transcription is unchanged. These experiments demonstrated that the Tat protein expressed in the etat cells was active and that extracts prepared from dividing cell populations transactivate the HIV-1 template. Similar to in vivo studies, in vitro transactivation is dependent upon the HIV-1 TAR RNA enhancer. These studies demonstrate that our transcription extracts correctly reflect in vivo Tat transactivation.

TAR-dependent Tat transactivation occurs during the G₁ phase of the cell cycle.

To examine the profile of HIV transcription and Tat transactivation during the cell cycle, we performed experiments with the reversible G₁/S cell cycle blocker hydroxyurea to synchronize the cells (34, 61). Following an 18-h block, the cells were washed and placed in fresh medium, and samples were collected every 3 h. An aliquot of the cells was analyzed by flow cytometry (Fig. 2A). The results of this analysis demonstrate that hydroxyurea effectively blocked etat and control cells at the G₁/S border (Fig. 2A, zero-hour profile). Upon release, there was a synchronous progression of the cells through S (Fig. 2A, 3-h profile), G₂/M (Fig. 2A, 6- and 9-h profiles), and into G₁ (Fig. 2A, 12- to 24-h profiles). A graphic presentation of the cell cycle block and progression is presented in Fig. 2B. The percentages of cells in G₁, S, and G₂/M are plotted at each time point.

FIG. 2.

Cell cycle analysis and in vitro transcription of etat and control cell lines following hydroxyurea or hydroxyurea-nocodazole block. Log-phase growing cells were blocked with hydroxyurea for 18 h (2 mM) and released by removing the inhibitor and adding fresh medium. Samples were collected every 3 h, and the cells were used to make whole-cell extracts for in vitro transcription or processed for FACS analysis. (A) Cell cycle analysis of HIV-1 transcription. The cells were removed from the medium at each time point, washed with PBS without Mg²⁺ or Ca²⁺, fixed with 70% ethanol, and stained with propidium iodide followed by cell-sorting analysis on a Coulter EPICS cell analyzer. The FACS profile for each time point is presented. Un, untreated. (B) (Top) Graphic presentation of the cell cycle block and progression. The percentage of cells in G₁, S, or G₂/M is plotted as percent total at each time point. (Middle) Whole-cell extracts were made from 5 × 10⁷ cells/time point, and 25 μg of total protein was used for in vitro transcription from both the LTR-TAT⁺ and LTR-TAT⁻ (100 ng) templates. (Bottom) Cell extract (100 μg) from each time point was immunoprecipitated and separated on an SDS-polyacrylamide gel electrophoresis gel, and Western blot analysis was performed with anti-Tat antibody. (C) Cell cycle analysis of AdML transcription. The AdML template (250 μg) was incubated with 25 μg of the same cell cycle extracts described above. Following a 1-h incubation, labeled transcripts were purified and analyzed as described above. (D) Cell cycle analysis of HIV-1 transcription following hydroxyurea-nocodazole treatment. Cells were blocked with hydroxyurea for 18 h, washed, and released for 1 h followed by addition of nocodazole (50 ng/ml) for 14 h. Following the block, the cells were washed and released. Samples were collected at each time point and processed for in vitro transcription analysis.

FIG. 3.

Analysis of HIV-1 transcription at G₂ and M phases of cell cycle. (A) etat or control cells were blocked with hydroxyurea and released for 6 h to obtain G₂-phase cells or double-blocked with hydroxyurea and nocodazole to obtain M-phase cells. The cells were then processed for in vitro transcription analysis. (B) In vitro transcription following immunodepletion of Tat protein from etat G₂ extracts. G₂-phase extract was treated with either 12CA5 (anti-epitope) or Tab172 (anti-Tax control) antibody followed by in vitro transcription from the supernatant. α-amanitin was used at 1.0 μg/ml to inhibit Pol II-specific transcription.

Whole-cell extracts were made from 5 × 10⁷ cells/time point and assayed for activity by in vitro transcription with both wild-type and TAR mutant templates. Transcription from control cell extracts showed a low level of transcription from both the wild-type and the mutant templates throughout the S and G₂/M phases of the cell cycle. At approximately 18 h after release, corresponding roughly to early to mid-G₁ phase of the cell cycle, a transient but significant 10-fold increase in TAR-independent basal HIV-1 transcription was observed (Fig. 2B, right).

When in vitro transcription was performed with the etat extracts, a distinct profile was observed. We consistently observed that cells at late G₁/early S (21 h) showed 10-fold-higher levels of transcriptional activity on the wild-type LTR template (LTR-TAR⁺) than on the TAR mutant template (LTR-TAR⁻) (Fig. 2B, left). Of interest, the timing of the Tat-dependent transcription was routinely observed at a slightly later phase of G₁ (18 versus 21 h). The fact that the timing of basal (in control cells) and Tat-transactivated (in etat cells) transcription are distinct may suggest that Tat switches the HIV-1 transcription phase. This point is under investigation. Unexpectedly, we also observed a distinct peak of Tat-dependent activation of both the LTR-TAT⁺ and LTR-TAT⁻ templates in the 6-h G₂ extracts. The Tat-dependent G₂ transcription was increased 10-fold over background.

Western blot analysis of Tat protein demonstrated that the protein was relatively constant throughout the cell cycle. Quantitative analysis of the Western blot determined that there was less than a twofold variation in Tat protein. These results suggest that the transcriptional levels in 6- and 21-h extracts were not simply due to changes in Tat protein levels (Fig. 2B, Tat immunoprecipitation/Western blot [IP/WB]). Moreover, the peaks in HIV-1 transcription are not due to variation in the overall transcription activity. Utilizing the same extracts, transcription assays with the AdML promoter were performed. The results of these transcription assays demonstrated that there was a fairly consistent level of transcription throughout the cell cycle (Fig. 2C). Quantitative analysis of the transcription assays demonstrated that there was less than a twofold difference in AdML transcription across the cell cycle.

In a separate series of experiments, we have tested the transcriptional activity of extracts from unselected cells synchronized at different stages of the cell cycle, supplemented with exogenous Tat. As reported by ourselves and numerous other groups, the addition of Tat to unsynchronized G₁ cell extracts results in an increase in transcription (8, 40, 42, 44, 48, 50, 51, 92). This transactivation, which is TAR dependent, is reproduced in the etat extracts from synchronized G₁ cells. In contrast, the addition of exogenous Tat to extracts from cells synchronized in G₂ does not result in activation of transcription. The potential for in vivo protein-protein interactions and posttranslational modifications that are necessary for Tat transactivation are evidently not totally reproduced when Tat is simply added to the G₂ extracts.

The 21-h extract showed specific Tat activation on the LTR-TAR⁺ promoter, suggesting that Tat transactivation occurred during the G₁ phase of the cell cycle. To confirm this observation, hydroxyurea-synchronized cells were released and treated with the mitotic blocker nocodazole, followed by release at various time points. Extracts were made from the control and etat cell lines at 0, 2, 4, 6, and 8 h post-nocodazole release for in vitro transcription. The results of such an experiment are shown in Fig. 2D. Low levels of HIV-1 transcription were observed during the early G₁ phase (Fig. 2D, 0-, and 2-h samples). Tat transactivation of the LTR-TAR⁺, but not the LTR-TAR⁻, increased transcription 9- to 10-fold at 4 h after nocodozole release. The level of Tat transactivation decreased to fivefold in the later stages of G₁/early S (Fig. 2D, 6 and 8 h).

TAR-independent Tat transactivation occurs at the G₂ phase of the cell cycle.

FACS analysis measures the DNA content of cells as determined by propidium iodide incorporation and scores cells of G₂ or M as a mixed G₂/M population. We were interested in determining if G₂- or M-phase cells were responsible for the TAR-independent transcription effect seen in the etat line (Fig. 2A, 6 h). To distinguish these cell populations, the cells were blocked with hydroxyurea and released for 6 h to obtain primarily G₂ extracts (73, 81). To obtain M-phase extracts, hydroxyurea-blocked cells were released and subsequently blocked with nocodazole for 18 h. G₂ extracts from etat cells showed high transcriptional activity compared to M-phase or control cell extracts (Fig. 3A, lanes 1 to 4). Consistent with the results presented in Fig. 2A, Tat transactivation (10- to 20-fold) at the G₂ phase was TAR independent (Fig. 3A, lane 1, upper and lower gels). Antibody clearing experiments demonstrated that the G₂/M transcription activity was Tat dependent in the etat cell line (Figure 3B). The epitope-directed 12CA5, but not the control Tab172, antibody decreased transcriptional activity approximately 11-fold (Fig. 3B, lanes 1 and 2). We further show that the G₂-phase transcription is Pol II transcription, since it is sensitive to low levels of α-amanitin (Fig. 3B, lane 3). The low transcriptional activity of the M-phase extracts is in agreement with published reports showing the downregulation of Pol II transcription during the M phase (68).

G₁/S Tat transactivation, but not G₂ transactivation, is Sp1 dependent.

We next compared in vitro transcription of HIV-1 wild-type and Sp1⁻ deletion mutant templates (73). Consistent with the results presented above, the level of HIV-1 transcription was significantly higher in the extracts containing Tat (Fig. 4, compare lanes 2 and 6 and lanes 4 and 8). Interestingly, the requirement for Sp1 was dependent upon the stage of the cell cycle (Fig. 4, lanes 1 and 3). G₁ activation (21-h extracts) required Sp1 binding sites (Fig. 4, lanes 1 and 2), whereas G₂ extracts (6-h) support activated transcription in the absence of Sp1 sites (Fig. 4, lanes 3 and 4). Western blot analysis of the pCEP and etat extracts demonstrated that the levels of Sp1 were similar for an asynchronous population of cells which are primarily in G₁ (21-h) (Fig. 4B, lanes 1 and 5). In the etat cells, a slight decrease in the level of Sp1 was observed as the cells progressed from G₁/S (0 h) through S (3 h) and into G₂ (6 h) (Fig. 4B). The change in Sp1 protein levels (twofold) was, however, less dramatic than the change in Sp1-dependent transcription activity, suggesting a fundamental change in the transcription program by Tat. It will be of interest to determine which upstream factors regulate HIV-1 transcription during the G₂ phase of the cell cycle. Of interest, the Sp1-del-LTR plasmid has all three Sp1 sites deleted, moving the NF-κB sites adjacent to the TATA box. The basal activity of the promoter in the PCEP4 cells correlates with in vivo transfection assays (41).

FIG. 4.

In vitro transcription of HIV-1 LTR Sp1 deletion mutants. (A) The etat or control cells were blocked with hydroxyurea for 18 h, washed, and released for 6 or 21 h for G₂ and G₁/S extracts, respectively. Following the block, the cells were washed twice with PBS and released with complete medium. Whole-cell extracts were made from 5 × 10⁷ cells/time point. One hundred nanograms of HIV-1 LTR (Wt) or HIV-1 LTR Sp1 deletion mutant (Sp1 del) template linearized with EcoRI (41) was added to the in vitro transcription reaction mixture containing 25 μg of extract. (B) Western blot analysis of Sp1 in pCEP and etat cell extracts. Log-phase growing cells were blocked with hydroxyurea for 18 h (2 mM) and released by removing the inhibitor and adding fresh medium. Samples were collected every 3 h, and the cells were used to make whole-cell extracts for in vitro transcription or Western blot analysis. One hundred micrograms of cell extract from each time point was separated on an SDS-polyacrylamide gel electrophoresis gel, and Western blot analysis was performed with anti-Sp1 anti-TBP antibody. A, asynchronous cell extract, primarily G₁; 0, 3, 6, time (in hours) after release from hydroxyurea block.

vpr cell cycle block induces Tat-dependent and TAR-independent transcription.

It was important to relate these observations on Tat transactivation to events which occur during the normal viral infection. The HIV-1 vpr protein prevents Cdc2 activation, thereby delaying or preventing replication of infected cells at the G₂/M boundary (4, 37, 60). It was of interest, therefore, to examine if vpr-blocked cells exhibited a high level of Tat-dependent and TAR-independent transactivation activity as observed in cells progressing through G₂ (Fig. 2). Control and etat cells were cotransfected with a HIV-1 LTR-driven vpr expression vector (4) and a plasmid carrying the gene for neomycin resistance. Cells were selected by addition of 100 μg of G418/ml to the media. Transfected cells were harvested at zero hour, immediately after transfection, or at 24 h and analyzed by FACS or used for preparation of in vitro transcription extracts. FACS analysis demonstrated that control or etat cells collected immediately after transfection were primarily in G₁ (Fig. 5A, zero-hour etat and control FACS profiles). In contrast, etat cells harvested 24 h after transfection with HIV-1 LTR-driven Vpr expression vector were primarily in G₂ (Fig. 5A, 24-h etat FACS profile). The 24-h sample from control cells transfected with HIV-1 LTR-driven Vpr expression vector did not accumulate in G₂, since there was no activator to increase Vpr expression (Fig. 4, 24-h control FACS profile). Extracts made from 0- and 24-h control or etat transfected cells were used for in vitro transcription of the wild-type HIV-1 promoter, TAR mutant TM26, or an Sp1 deletion mutant. Extracts from the zero-hour etat cells (vpr⁻), which were primarily in G₁, exhibited a requirement for both Sp1 and TAR (Fig. 5B and 5C, compare lanes 1 and 2). Mutations within either the TAR or Sp1 regulatory domain decreased transcription 10- to 14-fold. In contrast, vpr-blocked cells at G₂ support efficient Tat-dependent transcription from the Sp1 and TAR mutant TM26 templates (Fig. 5B and C, compare lanes 3 and 4). These results demonstrate that the G₂-blocked cells support a qualitatively different mode of activation than the G₁ population. The fact that control cells, which do not express high levels of vpr because of the absence of the Tat transactivator to drive LTR-vpr expression, did not lock at G₂/M indicates that vpr is responsible for the observed results.

FIG. 5.

vpr block at G₂ increases Tat-dependent and TAR-independent transcription. etat and control cells (3 × 10⁷ cells/time point) were transfected with 15 μg of the HIV-1-LTR-vpr expression vector and 5 μg of pSV2-neo. Zero-hour samples for cell sorting and in vitro transcription analysis were processed after electroporation. Twenty-four-hour samples were selected with hygromycin (200 μg/ml) and G418 (100 μg/ml) for 24 h. Two sets of templates (100 ng each), EcoRI-linearized HIV-1 wild type (Wt) and TM26 (66) and HIV-1 wild type and Sp1⁻ (41), were used for in vitro transcription assays (25 μg of total cellular protein). IRF, integral red fluorescence.

DISCUSSION

The experiments presented here provide important and novel insight into the nature of HIV-1 Tat transactivation and its regulation during the cell cycle. In particular, the data demonstrate that Tat transactivation through the TAR RNA enhancer element (5, 6, 14–16, 18, 19, 22, 27, 29, 35, 36, 42, 44, 47, 74, 83, 84, 91) occurs distinctly during mid- to late G₁. Tat transactivation during this phase of the cell cycle likely represents a window in which critical transcription factors required for Tat transactivation are present. A second and distinct phase of Tat transactivation, which is not dependent upon the TAR RNA enhancer or the Sp1 binding site, is observed during the G₂ phase of the cell cycle. Tat transactivation at this stage may represent the transactivation of cellular genes which are critical for virus replication (data not shown) (10, 62, 64, 77, 80). Because viral activators are absolutely essential for the virus life cycle, it is plausible to predict that they carry multiple functions to manipulate the host machinery at different stages of the cell cycle. It will be interesting to find out if this dual function of Tat is accomplished by interacting with different transcription factors or kinases at various stages of the cell cycle.

It is likely that Tat transactivation at the G₂ phase of the cell cycle plays an important role in virus replication. The HIV-1 protein Vpr inhibits cell growth by inducing a cell cycle arrest at the G₂/M boundary (4, 37, 59, 60). This arrest correlates with inactivation of the cellular Cdc2 kinase. All primate lentiviruses encode Vpr, and recent studies by Planelles et al. (60) suggest that the cell cycle arrest induced by vpr is a property shared by HIV-1 strains with diverse biological phenotypes. Further, deletion studies of HIV-1 vpr result in viruses with slower replication kinetics in lymphoid cells and peripheral blood mononuclear cells and severely attenuated virus replication in macrophages (3, 28). Goh et al. (24) have recently reported that Vpr increases viral gene expression by arresting cells in G₂, which the authors find is an optimal phase for viral transcription. Our results support the observation that cells in G₂ have a high level of Tat-dependent LTR transcription. Importantly, we find that this peak of viral transcription is TAR independent. We think it is also possible that the vpr block accentuates the Tat-dependent and TAR-independent transactivation of cellular genes which are critical for successful virus replication. Along these lines, it has been demonstrated that HIV-1 infection requires a “mitogenically activated” state. We would postulate that one of the targets of Tat is cellular cytokine genes whose products facilitate virus spread to neighboring cells by creating an environment which favors viral infection and replication.

It is becoming increasingly apparent that transcription factors are regulated in a cell cycle-dependent manner. For example, phosphorylation of the carboxy-terminal domain (CTD) of Pol II by cyclin-dependent kinases can result in activation or inactivation of the polymerase. CTD phosphorylation by TFIIH-associated Cdk7, the catalytic subunit of cyclin-activating kinase, results in activation of the elongation phase of Pol II transcription (46). In contrast, phosphorylation of the CTD by Cdc2 inhibited promoter-dependent transcription (23). It is also relevant that TFIIH phosphorylates Cdc2 and Cdk2 (72), suggesting a link between transcription and the cell cycle machinery. TFIID-associated TAFs are also apparently regulated during the cell cycle. It has been reported that a hamster cell line bearing a mutation in CCG1 (TAF 250) was defective in G₁ progression. TAF 250 is a major component of the holo-TFIID complex, acting as a scaffold for other TAFs and transcriptional activators. Since TFIID and TFIIH both play critical roles in HIV-1 Tat transactivation, it will be important to determine how cell cycle regulation of these transcription factors influences interaction with Tat and Tat transactivation. Finally, it has recently been reported that the cell cycle regulator E2F mediates repression of HIV-1 transcription through NF-κB (38). Interestingly, the other component of the E2F complex, Rb, also inhibits HIV-1 transcription through NF-κB (71). Thus, repression of HIV-1 transcription during the S phase may follow dissociation of the E2F-Rb complex at the G₁/S border.

The mechanism of Tat transactivation is unique among the viral activators in that Tat requires an RNA enhancer. From a virus replication standpoint, the use of an RNA enhancer likely provides an advantage for the virus in specifically activating viral transcription. It is possible that Pol II elongation factors are diminished during the G₁ window of HIV-1 transcription. Tat's ability to stimulate transcription elongation may overcome this inhibition. It is also possible that specific kinases important for Tat transactivation are active during the G₁ phase. Our experiments suggest that the key to understanding these RNA enhancer questions will lie in analyzing transcription activation during the G₁ phase of the cell cycle. In addition, our experiments demonstrate that there is a discrete Tat transactivation that occurs during the G₂ phase of the cell cycle. This activation window has been largely missed by previous analysis, since the majority of dividing cells used in either transient assays or preparation of in vitro transcription extracts are derived from populations of cells that are 60 to 80% G₁/S. Our studies, therefore, point to novel and unexpected pathways linking Tat transactivation to the cell cycle. Finally, the concept that a retroviral activator may work at different stages of the cell cycle opens the possibility of a similar phenomenon among all DNA and RNA virus activators.