Accelerated G(1) phase progression induced by the human T cell leukemia virus type I (HTLV-I) Tax oncoprotein.

Tax, the human T cell leukemia virus type I oncoprotein, plays a crucial role in viral transformation and the development of the virally associated disease adult T cell leukemia. Because oncogenesis involves alterations in cell growth, it is important to examine the effects of Tax on cell cycle progression. Using a synchronized cell system, we have found that Tax expression accelerates G(1) phase progression and S phase entry with concomitant DNA replication. This accelerated progression is accompanied by an earlier onset of cdk2 kinase activity. In contrast to the shortening of G(1) phase, the length of S phase is unaffected by Tax expression. As a result of a more rapid cell cycle progression, cells expressing Tax exhibit faster growth kinetics and display an altered cell cycle distribution. Additionally, the decreased time allowed for growth in the presence of Tax results in a decreased cell size. Tax-associated acceleration of cell cycle progression may play a role in the ability of this viral oncoprotein to mediate cellular transformation and promote the development of human T cell leukemia virus type I-associated diseases.

Tax, the human T cell leukemia virus type I oncoprotein, plays a crucial role in viral transformation and the development of the virally associated disease adult T cell leukemia. Because oncogenesis involves alterations in cell growth, it is important to examine the effects of Tax on cell cycle progression. Using a synchronized cell system, we have found that Tax expression accelerates G 1 phase progression and S phase entry with concomitant DNA replication. This accelerated progression is accompanied by an earlier onset of cdk2 kinase activity. In contrast to the shortening of G 1 phase, the length of S phase is unaffected by Tax expression. As a result of a more rapid cell cycle progression, cells expressing Tax exhibit faster growth kinetics and display an altered cell cycle distribution. Additionally, the decreased time allowed for growth in the presence of Tax results in a decreased cell size. Tax-associated acceleration of cell cycle progression may play a role in the ability of this viral oncoprotein to mediate cellular transformation and promote the development of human T cell leukemia virus type I-associated diseases.
Progression through the cell cycle is controlled by the activity of specific cellular kinases called cyclin-dependent kinases (cdk) 1 associated with their regulatory subunits, cyclins. When active, these kinase complexes phosphorylate a variety of specific cellular substrates that perform functions essential for the process of cellular division. A specific kinase complex regulates each phase of the cell cycle. For example, cdk4/6 associated with a D-type cyclin is required for G 1 phase progression (reviewed in Ref. 1), whereas S phase requires cdk2 associated first with cyclin E (2, 3) and subsequently with cyclin A (4). The ordered, sequential activation and deactivation of these specific cyclin-cdk complexes is essential for the maintenance of proper cell cycle regulation. Aberrant activation of a specific kinase complex can result in dysregulation of the cell cycle possibly leading to erroneous cell cycle progression and improper cellular division (reviewed in Ref. 5).
Because of the important regulatory role these cdk complexes play in normal cell growth, they often serve as targets of oncogenic viruses. Cancer-causing viruses encode one or more viral oncoproteins that, through various mechanisms, disrupt the regulation of cdk kinase complexes. For example, large T antigen of SV40, E1A of adenoviruses, and E7 of papillomaviruses all bind to and inactivate the tumor suppressor protein pRb (6 -8). This inactivation relieves the pRb-mediated repression of the transcription factor E2F, which is then able to activate genes essential for S phase progression, including cyclin E and cyclin A. Therefore, these viral oncoproteins allow for the unregulated expression of the cdk2 kinase regulatory subunits, which ultimately results in the dysregulation of cdk2 kinase activity, thereby altering cell cycle progression.
Human T cell leukemia virus type I (HTLV-I), a human retrovirus known to be the etiologic agent of adult T cell leukemia (ATL) (9), encodes a viral regulatory protein Tax which has been suggested to play a role in HTLV-I mediated transformation. The Tax protein is a transcriptional transactivator that enhances expression of viral as well as select cellular genes. Tax shares many characteristics with classic viral oncoproteins including the ability to immortalize primary cells in culture (10,11), induce tumors in transgenic mice (12,13), functionally inactivate tumor suppressor proteins such as p53 (14 -17) and p16 INK4A (18 -20), and inhibit cellular DNA repair (21,22). However, the effects of the Tax oncoprotein on cell cycle progression remain unclear and, therefore, serve as the focus of this report.
Infection with HTLV-I has been associated with changes in cellular proliferation including growth factor independence (23) and resistance to growth inhibitory signals (20). Although these findings demonstrate that HTLV-I is capable of modulating cell growth, the viral protein mediating these effects has not been identified. The viral regulatory protein Tax has been shown to inhibit negative cell cycle regulators such as p53 (14 -17), p16 INK4A (18 -20), and p27 Kip1 (24) and stimulate positive cell cycle regulators such as cdk4/6 (25, 26), D-type cyclins (26 -28), and E2F (29) (reviewed in Ref. 30). Given the ability of Tax to alter the activities of such key regulatory molecules, it is important to examine the effects of Tax on cell cycle progression.
Despite the fact that Tax affects positive and negative cell growth regulators, little work has been reported on the effect of these activities on cell cycle progression. This study was aimed at examining the effects of Tax on cell cycle progression. Using a synchronized cell system, we found that Tax expression was associated with an accelerated G 1 phase progression with an earlier onset of cdk2 kinase activity and with earlier S phase entry. In contrast, progression through S phase was unaffected by Tax expression. Also, Tax expression resulted in an accelerated growth rate causing an alteration in the cell cycle distribution pattern. Consistent with this reduced time allowed for growth during the cell cycle, cells expressing Tax exhibited a decrease in cell size. These results suggest that Tax expression accelerates cell cycle progression, which may play a role in the ability of Tax to mediate HTLV-I induced cellular transformation and disease.

EXPERIMENTAL PROCEDURES
Cell Lines-CREF-neo and CREF-Tax cells were previously described (31). Cells were maintained in Dulbecco's modified Eagle's serum supplemented with 10% fetal bovine serum.
Cell Synchronization-CREF cells were plated at a density of 0.5 ϫ 10 6 cells/100-mm dish and allowed to reach 100% confluence. Cells were maintained at 100% confluence for 48 h. To release cells from the cell cycle arrest, they were split 1:12 into fresh 100-mm dishes.
Propidium Iodide (PI) Staining-One million cells were resuspended in 0.9% NaCl, fixed in 95% ethanol, and incubated at room temperature for 30 min followed by storage at 4°C. For staining, the cell pellet was resuspended in 0.5 ml of 50 g/ml propidium iodide (PI) (Sigma) plus 100 l of 1 mg/ml RNase A (Sigma) and incubated at 37°C for 30 min. Cell cycle distribution was analyzed using flow cytometry (Epic Profile, Coulter, CO). The proportion of cells in each phase of the cell cycle was determined using ModFit (Verity).
BrdUrd Incorporation-Cells were synchronized and released as described above. Twelve h-postrelease media was spiked with 10 M BrdUrd (Sigma). At the indicated times postrelease, one million cells were collected by trypsinization, fixed in 3.7% formaldehyde, and stored at 4°C until staining. Incorporated BrdUrd was analyzed by staining cells with a FITC-conjugated anti-BrdUrd antibody (Becton Dickinson) according to the manufacturer's protocols.
Cell Extracts-For kinase assays, cells were lysed directly on 100-mm plates in 1 ml of EBC-D buffer (50 mM Tris, pH 8.0, 120 mM NaCl, 0.5% Nonidet P-40, 5 mM dithiothreitol) with protease inhibitors and incubated on ice for 30 min. Extracts were clarified at 4°C, and protein concentrations were determined by Bradford analysis (Bio-Rad). Extracts were stored at Ϫ80°C. Nuclear extracts were prepared for EMSA as previously described (32).
In Vitro Kinase Assay-Twenty g (cdk2) or 200 g (cyclin E or A) of extract was used per immunoprecipitation reaction using 2 g (cdk2), 4 g (cyclin E), or 6 g (cyclin A) of antibody. Immune complexes were isolated using Protein G Plus/A-agarose beads (Oncogene). Beads were washed five times with 1 ml of EBC-D buffer plus 0.03% SDS and once with 0.5 ml of TKB (50 mM Tris, pH 7.4, 10 mM MgCl 2 ). Washed beads were incubated with 25 l of kinase reaction mix (25 l of TKB, 2.5 mM MnCl 2 , 5 M ATP, 1.25 g of histone H1, 5 Ci [␥-32 P]ATP) for 30 min at room temperature. Proteins were resolved on 10% SDS-polyacryl-amide gel electrophoresis at 40 V for 15 h. Gels were fixed, dried, and exposed to film. Kinase activity was quantitated using a PhosphorImager. Results are presented as activity relative to the highest kinase activity for that specific assay.
EMSA-Electrophoretic Mobility Shift Assays (EMSA) were carried out as previously described (33) with a few variations. The DNA probe was a double stranded oligonucleotide (5Ј-CTAGATCTAGTTT-TCGCGCTTAAATTTGA-3Ј) containing the distal E2F binding site from the adenovirus type 5 E2a promoter (34). Each E2F binding reaction contained 7 g of nuclear extract, 10 l of DNA binding reaction mix (20 mM HEPES, pH 7.9, 40 mM KCl, 6 mM MgCl 2 , 1 mM EGTA, 1 mM dithiothreitol, 0.1% Nonidet P-40, 10% glycerol, 15 g of bovine serum albumin, 1 g of sonicated herring sperm DNA) and 5 ng of 32 P-labeled DNA probe. The binding reaction was incubated at room temperature for 20 min then resolved on a 4% polyacrylamide gel (with 5% glycerol) in TBE (50 mM Tris borate, 1 mM EGTA) at 4°C for 4 h at 150 V. For competition reactions, 500 ng of double-stranded unlabeled oligonucleotide of either the specific competitor, E2Fwt (sequence shown above), or the nonspecific competitor, E2Fmut, with the following sequence (5Ј-CTAGATCTAGTTTTCGATATTAAATTTGA-3Ј) was used. For supershift analyses, 0.5 l of the specified antibody was used in each binding reaction.

Tax Expression Accelerates Entry into S Phase-
The ability of classic oncoproteins to stimulate cell growth is believed to be important for the transformation processes mediated by these proteins. To examine the effects of a viral oncoprotein, HTLV-1 Tax, on cell cycle progression, a synchronized cell system was used to monitor progression through a single cell division cycle. In this system, an established rat embryo fibroblast cell line (CREF) stably expressing Tax (CREF-Tax) or a control gene (CREF-neo) were arrested by contact inhibition for 48 h then released into the cell cycle by replating at a lower density. Synchronization by cell contact inhibition typically results in cell cycle arrest and accumulation of cells in a quiescent, or G 0 , state. When CREF-neo and CREF-Tax cells were synchronized by contact inhibition for 48 h followed by flow cytometric analysis, almost 90% of the cells contained a 2N DNA content indicative of a cell in G 0 /G 1 phase (Fig. 1A).
To determine the effects of Tax on G 1 phase progression, CREF-neo or CREF-Tax cells were released from the cell cycle block and monitored for cell cycle progression. At various times postrelease, cells were fixed and stained with the DNA binding dye PI, and the cell cycle distribution was analyzed by flow cytometry. Analysis of S phase entry demonstrated that upon release from a G 0 block, CREF-Tax cells enter S phase ϳ4 h earlier than CREF-neo cells (Fig. 1B).
S phase, as assessed by PI staining in the above experiment, is defined as an increase in the relative amount of DNA. To verify that the cells were indeed in S phase, DNA replication, the hallmark of S phase, was analyzed by measuring BrdUrd incorporation. CREF-neo and CREF-Tax cells were synchronized by contact inhibition as described above and released in the presence of BrdUrd. At various times postrelease, cells were fixed and the amount of incorporated BrdUrd was determined by flow cytometry following staining with a FITC-conjugated anti-BrdUrd antibody. In agreement with the PI staining results (Fig. 1B), CREF-Tax cells displayed a 4-h earlier onset of DNA replication compared with CREF-neo cells (Fig. 1C) (13 h postrelease compared with 17 h). As expected, the onset of DNA replication as measured by BrdUrd staining precedes the detection of cells containing increased DNA content as measured by PI staining. Together, these results suggest that Tax expression promotes an accelerated progression of cells into a productive S phase.
Tax Expression Accelerates Onset of cdk2 Kinase Activity-The accelerated entry of Tax-expressing cells into S phase may result from stimulation of the G 1 /S phase transition, an event dependent upon kinase activity of the cyclin E-cdk2 protein complex. Aberrant activation of this kinase activity stimulates the G 1 /S transition (35). To determine whether Tax expression dysregulates cdk2 kinase activity during cell cycle progression, this activity was examined in synchronized CREF-neo and CREF-Tax cells. At various times postrelease from the contactinhibited arrest, whole cell extracts were prepared. Cdk2 was immunoprecipitated from these extracts and used in an in vitro kinase assay with histone H1 as a substrate. As shown in Fig.  2, induction of cdk2 kinase activity is first detectable at 12 h postrelease in CREF-Tax cells, whereas the first detectable induction of this activity in the absence of Tax does not occur until 16 h postrelease. Both the time of induction and the 4-h difference in the onset of this kinase activity between Taxexpressing and non-expressing cells correlates with the difference in S phase entry demonstrated in Fig. 1, B and C.
Because cdk2 is found sequentially in complex with cyclin E, regulating the G 1 /S transition, and with cyclin A, regulating S phase progression, it was important to determine whether the cdk2 activity observed in Fig. 2 could be attributed to one or both of these cyclin complexes. When cyclin E-and cyclin A-associated cdk2 kinase activities were analyzed separately using an in vitro kinase assay following cyclin-specific immunoprecipitation, results similar to those using total cdk2 activity were observed (Fig. 3). Cyclin E-cdk2 kinase activity (Fig. 3A), the activity required for the G 1 /S phase transition, was induced earlier than cyclin A-cdk2 kinase activity (Fig. 3B), the activity required for S phase progression in both CREF-Tax and CREFneo cell lines. Additionally, cyclin E-cdk2 and cyclin A-cdk2 kinase activities were apparent ϳ4 h earlier in cells expressing Tax. These results are similar to those observed for the induction of total cdk2 kinase activity (Fig. 2) as well as for entry into S phase (Fig. 1, B and C). The protein levels of cdk2, cyclin E, and cyclin A were determined by Western blotting and were induced at expected times during the cell cycle (data not shown). These data suggest that Tax expression results in an earlier onset of both cyclin E-and cyclin A-associated cdk2 kinase activity. This dysregulated activation of cdk2 may be responsible for the earlier entry into S phase of cells expressing Tax.
Tax Expression Does Not Affect the Length of S Phase-The data described above suggest that Tax expression causes a shortening of G 1 phase resulting in an accelerated entry into S phase. To determine whether these effects are specific to this phase of the cell cycle or whether other phases may be affected in the presence of Tax, progression through S phase was measured by completion of the phase and entry into G 2 /M phase. At various time points postrelease from the cell cycle arrest, cells were fixed and stained with PI, and their cell cycle distribution was analyzed by flow cytometry (Fig. 4). Cells expressing Tax complete S phase and enter G 2 /M phase ϳ4 h earlier than control cells (19 and 23 h postrelease, respectively). This difference in phase transition is the same as that seen with entry into S phase, suggesting that the time required for a cell to proceed through and complete S phase is unaffected by Tax expression.
Tax Does Not Affect Entry into or Exit from G 0 -The above results suggest that Tax expression stimulates G 1 phase progression. However, it is possible that Tax affects some aspect of cell synchronization, thereby contributing to the earlier onset of cdk2 activity and S phase entry. For example, if Tax-expressing cells arrested at the beginning of G 1 phase in response to contact inhibition, they would still appear to be synchronized in a G 0 /G 1 peak as we observed by flow cytometry (as shown in Fig. 1A), but they would gain a time advantage over non-Taxexpressing cells for progression through G 1 phase upon release from the cell cycle arrest. Alternatively, Tax-expressing cells may arrest in G 0 but may enter the cell cycle more rapidly, such that the differences in cell cycle progression map to the G 0 /G 1 phase transition rather than to G 1 phase progression. To address these questions, a quiescence, or G 0 phase-specific, marker was analyzed.
When normal cells are cycling, or non-quiescent, several E2F protein complexes, such as those containing p107, pRb, or "free" E2F (i.e. E2F not complexed with a pocket protein), are present, but the p130-E2F complex is absent. Upon induction of quiescence, the p130-E2F complex forms, and the p107-and pRb-containing complexes as well as free E2F complexes dissociate (36 -38). Therefore, the presence of p130-E2F complexes serves as a marker of quiescent cells, whereas the appearance of free E2F or p107-E2F complexes is indicative of cells proceeding through the cell cycle.
To verify that the contact-inhibited cells were arrested in a G 0 state EMSAs were performed to analyze the E2F complexes present in the synchronized cells (Fig. 5). Nuclear extracts from asynchronously growing CREF-neo and CREF-Tax cells contained p107-associated E2F complexes as well as free E2F complexes but lacked detectable p130-containing E2F complexes. However, synchronization of both cell lines by contact inhibition resulted in reduction of the p107-E2F and free E2F complexes with the concomitant appearance of a p130-E2F complex. The specificity of these complexes was demonstrated by competition with wild type and mutant E2F oligonucleotides. The identities of the p107 and p130-E2F complexes were confirmed by antibody supershifts. These results demonstrate that the synchronized cells are arrested in G 0 and that Tax expression does not alter the phase in which the cells are arrested by contact inhibition.
Because the duration of synchronization may affect a cell's ability to be released from cell cycle arrest, it was important to determine whether CREF-neo and CREF-Tax cells were synchronized at similar rates in response to contact inhibition and were maintained in G 0 for the same length of time. To address this question, equivalent numbers of cells were plated on 100-mm dishes. Cell number and cell cycle distribution was analyzed by PI staining every 12 h until the cells had been maintained at 100% confluence for 48 h. CREF-neo and CREF-Tax cells reach their respective maximum cell number at the same time (72 h postplating) and at the same rate as indicated by the slope of the curve (Fig. 6A). The slightly higher number of Tax-expressing cells at confluence is likely due to their reduced size (discussed later) and ability to grow to a higher density 2 than cells without Tax. Examination of the cell cycle distribution during synchronization demonstrated that the rate at which CREF-neo and CREF-Tax cells accumulate with a G 0 /G 1 content of DNA was similar and correlated with the maintenance of constant cell number following confluence at 72 h postplating (Fig. 6B).
To further investigate the rate at which CREF-neo and CREF-Tax cells synchronize in response to contact inhibition, nuclear extracts were prepared at intervals following confluence, and the various E2F complexes were analyzed by EMSA (Fig. 6C). The p107-E2F complex as well as free E2F was present in cells with and without Tax until 12 h after the cells had reached 100% confluence. At 24 h postconfluence, both CREF-neo and CREF-Tax cells appear to have lost p107-E2F complexes and free E2F. In their place, p130-E2F complexes were observed in both cell types at 24 h postconfluence and persisted for the remainder of the experiment. Specificity of these complexes was demonstrated by competition analysis and the positions of p107, p130, and free E2F complexes were confirmed by supershift (data not shown). Using this analysis, there was no detectable difference in the rate at which CREFneo and CREF-Tax cells become synchronized in a G 0 state.
Several lines of evidence suggest that Tax facilitates the release of cells from a cell cycle arrest (20,25) were synchronized in G 0 by contact inhibition and released as described above. At 2-h intervals following release, nuclear extracts were prepared, and the E2F complexes present in these cells were analyzed by EMSA (Fig. 6D). At the time of release (0 h), both cell lines contained p130-E2F complexes as shown in Fig. 4. At 4 h postrelease, free E2F complexes indicative of G 1 phase began to accumulate. The composition of these complexes was verified by supershift analysis using specific antibodies (data not shown). An equivalent rate of free E2F accumulation was observed between CREF-neo and CREF-Tax cells. The slightly higher levels of E2F seen here in CREF-Tax cells at 4 h was not observed reproducibly. These results suggest Tax expression does not alter the ability of CREF cells to exit a quiescent state nor does it affect the rate at which cells enter G 1 phase of the cell cycle.
Tax Expression Affects Cell Growth Kinetics-All of the results described above are consistent with an effect of Tax on cell cycle progression during G 1 phase progression or at the G 1 /S phase transition. The ability of Tax to stimulate cell cycle progression would also likely stimulate cell proliferation resulting in accelerated cell growth. To examine whether Tax expression affected the length of the complete cell division cycle, cell proliferation and growth kinetics of CREF-neo and CREF-Tax cells were analyzed. Equivalent numbers of cells were plated in 150-mm dishes, and the total cell number was determined every 24 h for 6 days, during which time cells were maintained at subconfluence. Over the course of the experiment, CREF-Tax cells outgrew CREF-neo cells by ϳ1.6-fold (Fig. 7). This correlates with an approximate cell division time of 16 h for CREF-Tax and 19 h for CREF-neo cells and suggests that Tax, like other oncoproteins, is able to stimulate cell growth. These effects have been reproduced in p53 null mouse embryo fibroblasts and SaOS2 cells, both of which stably express Tax (data not shown).
For a cell to divide more rapidly some alteration(s) of the cell cycle must occur, resulting in the shortening of one or more of the phases of the cell cycle such that the time required for completion of the entire cell division process is decreased. If such a change has taken place, it may be detectable as a difference in the proportion of cells in each phase of the cell cycle. With the observation that CREF-Tax cells divide more rapidly than control cells, the ability of Tax expression to affect changes in the cell cycle distribution pattern of asynchronous cells was examined. Asynchronously growing CREF-neo and CREF-Tax cells were fixed and stained with PI, and their cell cycle distribution was analyzed by flow cytometry. The percentage of cells in each phase of the cell cycle as determined by relative DNA content demonstrated that Tax expression is associated with changes in the cell cycle profile of CREF cells (Fig. 8). Although these differences are small, there is a reproducible decrease in the percentage of cells with a G 0 /G 1 (2N) content of DNA in the presence of Tax. These results are consistent with Tax affecting cell proliferation by facilitating a shortening of G 1 phase and confirm that the accelerated cell cycle progression stimulated by Tax is not simply an effect of cell synchronization.
Tax Expression Affects Cell Size-As a consequence of Tax stimulating cell cycle progression, cells have less time to pro- FIG. 6. Tax expression does not affect the rate of synchronization or entry into the cell cycle. One half-million CREF-neo and CREF-Tax cells were plated in 100-mm dishes. Every 12 h, cells were counted (A) and fixed. The cell cycle distribution was analyzed by flow cytometry (B). C, CREF-neo (left) and CREF-Tax (right) cells were allowed to reach confluence. At the indicated time points post-confluence, nuclear extracts were prepared, and the E2F complexes present were analyzed by EMSA. D, CREF-neo and CREF-Tax cells were synchronized by contact inhibition for 48 h then released by replating at a lower density. At the indicated times postrelease, nuclear extracts were prepared, and the E2F complexes were analyzed by EMSA. The position of E2F complexes is shown. ceed through G 1 phase, one of the two growth phases of the cell cycle. In other systems, a shortened G 1 phase results in a decreased cell size because of reduced time allowed for growth (39,40). To determine whether Tax-mediated acceleration of cell cycle progression correlates with an effect on cell growth, the size of CREF-Tax and CREF-neo cells was examined by flow cytometry (Fig. 9). In the presence of Tax, CREF cells exhibit an approximate 12% reduction in cell size. This effect was also observed in SaOS2 cells stably expressing Tax (data not shown). Reduced cell size in the presence of Tax expression is consistent with the acceleration of G 1 phase progression described above. Indeed, additional flow cytometry studies demonstrated a 10% reduction in cell size in the G1 population and a recovery of cell size in subsequent cell cycle phases (data not shown).

DISCUSSION
This study examined the effect of a viral transforming protein on cell cycle progression. The results are summarized in a schematic diagram (Fig. 10). Upon analysis of cell cycle progression and phase transitions using a synchronized cell system, we found that Tax expression accelerated progression through G 1 phase resulting in an earlier entry into S phase with concurrent DNA replication. These synchronized cells also displayed an earlier onset of cdk2 kinase activity, an aberration that may be responsible for the accelerated G 1 /S phase transition. We demonstrated that Tax expression resulted in an accelerated growth rate in rat embryo fibroblasts as well as a reduction in cell size. These results support the idea that the viral oncoprotein, Tax, accelerates cell cycle progression. This ability to promote cellular proliferation may play an important role in the virally associated transformation mediated by Tax.
Previous reports have suggested a role for Tax in the stimulation of cell cycle progression (20,25). These studies have examined the ability of Tax to promote entry into the cell cycle from a quiescent or arrested state such that in the absence of Tax expression proliferation was inhibited. However, when Tax was expressed, cells were stimulated to enter the cell cycle. The results suggest that Tax can stimulate an early event resulting in cell cycle entry, but they do not address directly the effects of Tax expression on the progression of cells that are actively proceeding through the cell cycle.
The results presented here demonstrate that Tax expression causes an accelerated entry into S phase following the similar entry into G 1 of cells with and without Tax expression. Because there is no difference in cell cycle entry upon release from the synchronization block (as shown in Fig. 6D), it appears that Tax exerts its effect at a later point during G 1 phase. There are two possible explanations for such an effect. First, Tax may accelerate the progression through a complete G 1 phase such that the cell reaches the G 1 /S phase transition earlier. Alternatively, Tax may stimulate the G 1 /S phase transition such that the cell is induced to enter S phase before the completion of G 1 phase. Further studies will distinguish between these two potential explanations. Our results definitively demonstrate the ability of Tax to accelerate cell cycle progression in actively dividing cells and show that this effect is mediated at a point between entry into the cell cycle at G 1 phase and the G 1 /S phase transition. The accelerated cell cycle progression reported here is not a consequence of the synchronization process because asynchronized cells showed a shorter cell division time, an altered cell cycle distribution, and a smaller size corresponding well with accelerated progression following synchronization.
Previous reports have shown that Tax can stimulate cdk4/6 kinase activity (25, 26), which is necessary for proper G 1 phase progression. This increased cdk4/6 activity may play a role in the accelerated progression through G 1 phase observed here. However, we cannot overlook the possibility that Tax may independently stimulate cyclin E-cdk2 kinase activity with concomitant induction of the G 1 /S phase transition regardless of G 1 phase completion. Determining which kinase activity affected by Tax facilitates the observed acceleration of cell cycle progression will be an important next step in understanding this activity.
Infection with the transforming retrovirus HTLV-I has been associated with changes in cellular growth (20,23). The results presented in this study demonstrate that the HTLV-I regula- Upon release from a G 0 cell cycle arrest induced by contact inhibition, CREF-neo and CREF-Tax cells enter G 1 phase at the same rate. CREF-Tax cells display a shortened G1 phase with an earlier onset of cdk2 kinase activity. Tax-expressing cells also display an earlier entry into G 2 /M suggesting that the length of S phase is similar in Tax (ϩ) and (-) cells. The complete cell division time for Tax-expressing cells is shorter than non-Tax-expressing cells suggesting that the length of G 2 /M is similar in both cell types. The ability of Tax to cause an accelerated cell cycle progression may play a role in HTLV-I-mediated transformation induced by Tax. tory protein Tax can induce alterations in cellular proliferation and, therefore, may be the viral protein responsible for the effects observed with HTLV-I infection. The ability of HTLV-I Tax to alter cellular proliferation likely plays an important role in viral-mediated transformation. This effect of Tax is similar to that described for other viral oncoproteins such as large T antigen of SV40 and E1A of adenovirus. The finding that the retroviral oncoprotein Tax shares a common mechanism of stimulating cellular growth with DNA tumor virus-transforming proteins suggests the importance of dysregulating the cell division cycle in viral-mediated cellular transformation. A greater understanding of the mechanism(s) underlying the ability of these viral oncoproteins to induce cell growth will provide insight into the regulation of cell cycle progression and cellular transformation and facilitate our understanding of mechanisms underlying the development of both virally and non-virally associated cancers.