INTRODUCTION

Top Abstract Introduction Materials and methods Results Discussion References

Human T-cell leukemia virus type 1 (HTLV-1) is the etiologic agent of adult T-cell leukemia, which is characterized by a dysregulated proliferation of T cells (14, 29, 45). HTLV-1 transforms normal T cells in vitro (26, 42). Two types of T-cell lines are transformed by the virus; the first proliferates in an interleukin-2 (IL-2)-dependent manner, while the other proliferates in an IL-2-independent manner (35). The viral nonstructure protein Tax is involved in the IL-2-dependent transformation step, since it transforms normal T cells in the presence of IL-2 (2, 13). However, the exact mechanism through which HTLV-1 transforms normal T cells in an IL-2-independent manner is not well understood.

Tax has multiple functions. For example, it activates the transcription of various cellular genes, such as proto-oncogenes (c-fos, c-jun, fra-1, and c-myc) and genes encoding growth factors (IL-2, IL-6, transforming growth factor-, OX40 ligand, and granulocyte-macrophage colony-stimulating factor) and their receptors (-chain of IL-2 receptor, OX40) (8, 10, 11, 15, 20, 23, 27, 40, 43, 45). Tax also represses the transcription of several cellular genes such as those encoding DNA polymerase  and Bax (5, 16). In addition, it stimulates the activity of several kinase enzymes such as cyclin-dependent kinase (CDK) and protein kinase C (18, 38).

In the activities described above, Tax interacts with various cellular proteins such as the transcription factors SRF (CArG binding protein) and CREB (cyclic AMP-responsive element [CRE]-binding protein), IB (an inhibitor of NF-B/Rel transcription factors), MEKK1 (kinase of IB kinase), and INK-4, an inhibitor of CDK (12, 36, 38, 41, 44, 46). These interactions mediate Tax-dependent activation of transcription via CRE (CREB), CArG (SRF), B element (IB), and CDK kinase (INK-4). Tax also associates with other cellular proteins such as mitotic checkpoint protein MAD1 and cyclin D (17, 28).

In the present study, we investigated whether Tax is involved in IL-2-independent transformation of T cells by HTLV-1. For this purpose, we introduced the tax gene into a mouse IL-2-dependent T-cell line, CTLL-2, to examine the growth property of these cells in the absence of IL-2. There are two well-known Tax mutants, TaxM22 and Tax703(M47). The latter activates NF-B-dependent transcription but not CRE-dependent transcription, whereas the reverse is true for TaxM22. Using these two mutants, we also investigated the involvement of these two pathways in the transformation (IL-2-independent proliferation and colony formation in soft agar) of CTLL-2 cells by Tax.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and methods Results Discussion References

Plasmids. The wild-type tax and tax mutant genes were cloned into pHPr.1-neo, which has a -actin promoter for protein expression in vivo as well as a neomycin resistance gene for G418 selection (21, 22). The Tax mutants TaxM22 (33) and Tax703(M47) (1, 22, 32) have been previously characterized, and their substituted amino acids (position in parentheses) are Thr(130)Leu to SerAla for TaxM22 and Leu(319)Leu to ArgSer for Tax703. HTLV LTR-luc and B-luc are luciferase expression plasmids regulated by the HTLV-1 long terminal repeat (LTR) promoter containing Tax-inducible CREs (21-bp repetitive sequence) and a pentamer of the B element from the IL-2R gene (CAGGGGAATCTC), respectively (37).

Cell culture and cell growth assay. CTLL-2 is a mouse IL-2-dependent T-cell line. TL-Su, HUT102, MT-2, ILT-Hod, ILT-Oot, ILT-MC, ILT-Mat, ILT-Kan2, and ILT-Koy are human HTLV-1-transformed T-cell lines (3, 34). MT-2, HUT102, and TL-Su were cultured in RPMI 1640 supplemented with 10% fetal calf serum (FCS) (RPMI-FCS). ILT cell lines were cultured in RPMI-FCS with 1 nM IL-2. CTLL-2 was cultured in RPMI-FCS with 1 nM IL-2 and 2-mercaptoethanol. To examine cell growth, 10⁵ cells were washed twice with phosphate-buffered saline (PBS) and suspended in RPMI-FCS. They were then cultured in a 24-well at 37°C. The cell number was counted by the trypan blue staining method.

Establishment of CTLL-2 cell lines expressing Tax. To establish CTLL-2 cell lines expressing Tax proteins, CTLL-2 cells were washed twice with PBS and once with K-PBS (30.8 mM NaCl, 120.7 mM KCl, 8.1 mM Na₂HPO₄, 1.46 mM KH₂PO₄), and suspended in K-PBS. Cells (10⁷) in K-PBS were mixed with 30 µg of Tax or Tax mutant expression plasmid in an electroporation cuvette, incubated on ice for 10 min, and then pulsed with a Electroporator (Bio-Rad) at 320 V and 950 µF. In the next step, the cells were placed on ice for 10 min, addition of FCS was added (to a final concentration of 10%), and the cells were further incubated at room temperature for 10 min. Then the cells were seeded in 24-well plates and cultured in RPMI-FCS containing IL-2. At 24 h after electroporation, G418 (0.4 mg/ml) was added, and the cells were cultured for about 4 to 6 weeks. G418-resistant cells were screened for the expression of Tax protein by Western blot analysis, and cells expressing Tax were further cloned by the limiting-dilution method.

Western blotting. Cell lysates prepared from CTLL-2 or HTLV-1-transformed T-cell lines were resolved by electrophoresis on 10% polyacrylamide gels and then transferred to polyvinylidene difluoride membranes (Amersham). The blots were incubated with a mouse monoclonal anti-Tax antibody (Taxy-8) (39) and then washed and incubated with anti-mouse immunoglobulin conjugated with horseradish peroxidase. Proteins recognized by the antibodies were visualized with the enhanced chemiluminescence Western blotting detection system (Amersham).

Luciferase assay. CTLL-2 cells were washed twice with PBS and suspended in RPMI 1640 medium. The cells were mixed with luciferase plasmid (10 µg) and Tax expression plasmid (20 µg), and DEAE dextran (500 µg/ml) was added. They were then incubated at 37°C for 30 min with occasional mixing. After the cells were washed with RPMI 1640, they were cultured for 48 h, a cell lysate was prepared, and the luciferase activity in the lysate was determined with a luminometer.

EMSA. For the preparation of nuclear extract, 10⁷ cells were washed with PBS containing 1 mM Na₃VO₄ and 5 mM NaF. The cells were then treated with 0.2% Nonidet P-40 in lysis buffer containing 20 mM HEPES (pH 7.9), 20 mM NaF, 1 mM Na₃VO₄, 1 mM EDTA, 1 mM EGTA, 1 mM dithiothreitol, 0.5 mM phenylmethylsulfonyl fluoride, 1 µg of leupeptin per ml, and 1 µg of aprotinin per ml. After centrifugation, the pellets were further treated at 4°C for 30 min with lysis buffer supplemented with 420 mM NaCl and 20% glycerol and then subjected to centrifugation. The resulting supernatant was used as the nuclear extract in an electrophoretic mobility shift assay (EMSA). Nuclear extract (10 µg) was preincubated for 15 min on ice with 1 µg of poly(dI-dC) in 20 µl of a binding buffer containing 13 mM HEPES (pH 7.9), 65 mM NaCl, 0.15 mM EDTA, 8% glycerol, and 1 mM dithiothreitol. Approximately 1 ng of labeled oligonucleotide was added to the reaction mixture, which was then further incubated for 15 min at 25°C. The formed complexes were separated from the unbound probe by electrophoresis in a 5% polyacrylamide gel containing 0.5× TBE, and 2.5% glycerol, and the gel was dried and exposed to X-ray film. A double-stranded synthetic oligonucleotide corresponding to the NF-B binding site (top strand, AGCTTTGGGAAATTCCTCGGGTGGTAC) from the interferon gene was labeled with [-³²P]ATP by using polynucleotide kinase and was used as the B-site probe.

Analysis of apoptosis. The CTLL-2/Vec, CTLL-2/WT, CTLL-2/703, and CTLL-2/M22 cell lines were washed twice with PBS and then cultured in the absence of IL-2 for 0 to 3 days. The cells were washed twice with cold PBS and suspended in a binding buffer provided by the manufacturer (R & D). They were then mixed with fluorescein-conjugated annexin V and propidium iodine reagent and incubated for 15 min at 20 to 25°C in the dark. After fivefold dilution with the binding buffer, the cells were analyzed for staining for annexin V on a FACScalibur.

Assay of colony formation in soft agar (CFSA). Cells (10³) were suspended in RPMI-FCS containing 0.4% agarose (SeaPlaque; FMC), which was first melted by heating. The suspension (1 ml) was then poured onto a base layer consisting of 2 ml of RPMI 1640 containing 0.53% agarose in a six-well plate. After 2 weeks, the number of colonies (each containing >100 cells) was counted. The colony number was expressed as the percentage of colonies in 10³ inoculated cells.

	RESULTS

Top Abstract Introduction Materials and methods Results Discussion References

Tax converts cell growth of CTLL-2 from IL-2 dependent to IL-2 independent. CTLL-2 is a mouse T-cell line, and the cells are dependent on IL-2 for their proliferation. CTLL-2 cells were transfected with either a Tax-expressing plasmid or a vector plasmid and cultured in the presence of IL-2 and G418 for selection. Selected clones were first examined for the expression of Tax protein by Western blot analysis. Tax protein was detected in three representative clones transfected with the tax plasmid (WT-5, WT-7, and WT-14) but not in those transfected with the vector plasmid (Fig. 1A). Such CTLL-2 clones were then cultured in the absence of IL-2, and cell growth was examined by trypan blue exclusion. Similar to parental CTLL-2 cells, cells transfected with vector plasmid (CTLL-2/Vec) rapidly died in the absence of IL-2 whereas all three CTLL-2 cell lines that expressed Tax showed a continuous proliferation (Fig. 1B). These three clones were maintained in the absence of IL-2 for more than 5 months. Thus, Tax can modulate the growth properties of CTLL-2 cells from being IL-2 dependent to being IL-2 independent. The culture supernatant of these CTLL-2/WT cells did not support the growth of parental CTLL-2 cells in the absence of IL-2 and did not contain detectable levels of IL-2, as confirmed by enzyme-linked immunosorbent assay analysis (data not shown). Thus, IL-2-independent proliferation of CTLL-2/WT cells was not due to IL-2 or other cytokines that might have been induced by Tax.

View larger version (24K): [in this window] [in a new window]   FIG. 1.   Tax induces IL-2-independent proliferation of CTLL-2 cells. (A) Expression of Tax protein in CTLL-2/WT cells. Cell lysates were prepared from CTLL-2 cells transfected either with the vector plasmid (lane 1) or with expression plasmids encoding TaxWT (lanes 2 to 4). The expression of Tax proteins in these lysates was measured by Western blot analysis with anti-Tax antibody (Taxy-8). (B) Cell growth analysis of CTLL-2 cells expressing Tax protein in the absence of IL-2. CTLL-2, CTLL-2/Vec, and three CTLL-2 clones expressing Tax protein were cultured in the absence of IL-2. Cell growth was measured by trypan blue staining.

Function of Tax responsible for IL-2-independent transformation of CTLL-2 cells. Tax activates transcription via the B element and CRE in the HTLV-1 LTR and in cellular genes (45). TaxM22 and Tax703(M47) are substitution mutants of Tax, originally described by Smith et al. (32, 33) (Fig. 2A), which can segregate the activities of Tax to the B element and CRE in various types of cells (1, 22, 32, 33). The wild-type Tax (TaxWT) and the two Tax mutant plasmids were transiently transfected into CTLL-2 cells together with Luc (luciferase) reporters containing either CRE or the B element. Expression of TaxWT stimulated Luc activity from two reporters with either CRE or the B element (Fig. 2). Tax703 activated the B element more than TaxWT did, whereas it caused only a negligible activation of CRE. In contrast, TaxM22 activated CRE and produced only a slight inhibition of the B element. The activation (fold induction) by Tax of the B element in CTLL-2 cells was low relative to that reported previously with other cell lines such as the T-cell line Jurkat. IL-2 also activated NF-B activity in T-cell lines including CTLL-2 (see Fig. 4, lane 4). Thus, the activation by Tax of the B element may be partially masked by the activation of this element by IL-2. Indeed, Tax and Tax703 but not TaxM22 activated the B element much more efficiently in Jurkat cells than in CTLL-2 cells (data not shown). In this regard, it was difficult to measure the activation by Tax of the B element in IL-2-deprived CTLL-2 cells, since these cells died within 3 days.

View larger version (16K): [in this window] [in a new window]   FIG. 2.   trans-activation phenotype of TaxWT and its mutants. (A) Schematic structures of TaxWT and its mutants. The position of amino acid substitution in each mutant is indicated. (B) Luciferase assay. Reporter Luc plasmids with a CRE (LTR-luc) or B element were transfected into CTLL-2 cells together with expression plasmids encoding Tax or its indicated mutants. Luc activity in the harvested cells was measured as described in Materials and Methods. Fold activation shows the ratio of Luc activity in cells cotransfected with the expression plasmid encoding Tax or its mutant relative to that of cells cotransfected with the vector plasmid. The average Luc activity of three experiments is presented. Data are representative of three reproducible independent experiments.

View larger version (34K): [in this window] [in a new window]   FIG. 3.   Pathway of Tax responsible for IL-2-independent proliferation of CTLL-2 cells. (A) Expression of Tax proteins in CTLL-2 cells. Cell lysates were prepared from CTLL-2 cells transfected either with vector plasmid (pHPr.1-neo) or with expression vectors encoding TaxWT or the indicted Tax mutant proteins. The expression of Tax proteins in these lysates was measured by Western blot analysis with anti-Tax antibody (Taxy-8). (B) Cell growth analysis of CTLL-2 cells expressing each Tax mutant. CTLL-2 clones expressing the indicated Tax mutant were cultured in the absence of IL-2, and cell growth was measured by trypan blue staining.

Tax-mediated activation of NF-B in CTLL-2 cells. We next measured the binding activity of the B element (NF-B activity) in CTLL-2 cells expressing Tax by using EMSA. Consistent with the Luc assay, CTLL/WT-5 and CTLL/703-2 showed a higher NF-B activity (slow-migrating complex) than did CTLL/Vec in both the absence and presence of IL-2 (Fig. 4). The Tax-induced complex was specific to the B element, since it was inhibited by homologous B oligonucleotides but not by unrelated TRE (TPA-responsive element) from the collagenase gene (Fig. 4). The fast-migrating band was also specific to the B element, but the complex was not induced by Tax. Thus, the slow-migrating complex was associated with IL-2-independent proliferation (Fig. 3). Similar results were obtained with other Tax expressing CTLL-2 clones. In contrast, CTLL/M22-1 showed a lower NF-B activity than did CTLL/Vec in the presence of IL-2. Inhibition of NF-B activity by TaxM22 was consistent with that of the B reporter by TaxM22 (Fig. 2). In addition, absence of IL-2 reduced NF-B activity in CTLL-2 cells. Thus, reduced NF-B activity was associated with CTLL-2 cell death.

View larger version (71K): [in this window] [in a new window]   FIG. 4.   NF-B activity in CTLL-2 cells expressing Tax or its mutant. Nuclear extracts were prepared from CTLL-2 cells transfected with the indicated tax mutant plasmids. The extracts were then incubated with an NF-B binding site labeled with [-32P]ATP in the absence (lanes 1 to 8) or presence of homologous oligonucleotides (lane 10) or an unrelated one (TRE, lane 11). The formed complex was separated by polyacrylamide gel electrophoresis (5% polyacrylamide). Arrows indicate the specific complex.

Inhibition of apoptosis in CTLL-2 cells expressing Tax. In the next step, we examined the extent of apoptosis in CTLL-2/Vec and CTLL-2 cells that expressed Tax proteins by using the annexin V staining method. Annexin V binding to cells is a marker of the early stages of apoptosis. Apoptosis of 10.3% of the CTLL-2/Vec cells was detected on day 0. Culture of cells in the absence of IL-2 resulted in a rapid increase in the population of apoptotic cells till day 3, and 96.6% of cells stained with annexin V on day 3 (Fig. 5). On the other hand, there was only a slight increase in apoptotic CTLL-2/WT (from 9.2 to 10.3%) and CTLL-2/703 (from 25.8 to 33.7%) cells during a 3-day culture in the absence of IL-2. Therefore, Tax and Tax703 proteins inhibited apoptosis of CTLL-2 cells in the absence of IL-2. Annexin staining experiments also showed that apoptosis was less prevalent in CTLL-2/WT-5 cells than in CTLL-2/703-2 cells. Thus, both these Tax proteins induced IL-2-independent proliferation of CTLL-2, but TaxWT produced a limited degree of apoptosis compared with Tax703. Unlike CTLL-2 cells expressing TaxWT and Tax703, apoptosis of cells that expressed TaxM22 occurred much faster than did apoptosis of CTLL-2/Vec cells. These results were consistent with those of the cell growth analysis (Fig. 3). In addition, more apoptotic cells were detected in CTLL-2/M22 than CTLL-2/Vec cells even in the presence of IL-2 (Fig. 5, compare M22 with Vec on day 0). These results indicated that TaxM22 enhanced apoptosis of CTLL-2 in the absence as well as the presence of IL-2.

View larger version (33K): [in this window] [in a new window]   FIG. 5.   Tax and Tax703 inhibit apoptosis induced by IL-2 deprivation in CTLL-2 cells. CTLL-2, CTLL-2/WT, CTLL-2/703, and CTLL-2/M22 cells were cultured in the absence of IL-2 for the indicated time intervals. They were then incubated with fluorescein isothiocyanate-conjugated annexin, and staining was measured with a FACScalibur. Numbers represent the percentages of unstained cells (left) and those stained with annexin V (right).

CFSA of CTLL-2 cells expressing Tax. CFSA is a transformation assay for various types of cells including lymphocytes (31). We examined CFSA activity of CTLL-2 cells that expressed Tax protein, and the results are summarized in Table 1. CTLL-2 and CTLL-2/M22 cells produced only a few colonies in the absence of IL-2, but all three CTLL-2/WT cells reproducibly formed colonies (0.43 to 16.4%). On the other hand, only two of three CTLL-2/703 clones in the first series of experiments and one in the second series of experiments showed CFSA activity, and the colony numbers (0.17 to 0.27%) were smaller than for CTLL/WT (0.8 to 7.4%). Thus, CFSA activity of CTLL-2 cells expressing Tax703 was lower than for those expressing TaxWT, although both TaxWT and Tax703 induced IL-2-independent proliferation of CTLL-2 cells.

Cell growth of HTLV-1-transformed T-cell lines in the absence of IL-2. Two types of T-cell lines are transformed by HTLV-1; one is IL-2 dependent, and the other is IL-2 independent. Since Tax induced IL-2-independent proliferation of CTLL-2 cells, we next examined the involvement of Tax in IL-2-independent transformation of T cells by HTLV-1. Six IL-2-dependent and three IL-2-independent cell lines were examined for their expression of Tax as well as for cell growth characteristics in the absence of IL-2. ILT-Koy cells expressed Tax most abundantly among the six IL-2-dependent cell lines and proliferated even in the absence of IL-2 for 6 days, but their number started to decrease on day 8 (Fig. 6) and most died after 3 to 6 weeks (data not shown). Tax could not be detected in ILT-Mat cells by Western blot analysis, and only low levels of expression were detected by Northern blot analysis (data not shown). After deprivation of IL-2, ILT-Mat was the quickest to die among the six IL-2-dependent cell lines. The other four cell lines (ILT-Hod, ILT-Kan2, ILT-Oot, and ILT-MC), which expressed detectable levels of Tax but lower levels than did ILT-Koy, did not grow in the absence of IL-2, although they survived longer than ILT-Mat. Thus, the duration of survival of these cell lines in the absence of IL-2 correlated with the level of expression of Tax in these cells. Three IL-2-independent cell lines expressed relatively high levels of Tax. However, in two IL-2-dependent cell lines (ILT-Koy and ILT-Oot), the level of expression of Tax was equivalent to that in an IL-2-independent cell line (TL-Su), suggesting that even a sufficient level of Tax could not induce IL-2-independent transformation of T-cells by HTLV-1. All three IL-2-independent cell lines showed persistent growth in the absence of IL-2 (Fig. 6B). TL-Su expressed Tax less abundantly and proliferated more slowly than the other two cell lines. Thus, the growth rate of these three IL-2-independent cell lines in the absence of IL-2 correlated with the amount of Tax being expressed by these cells.

View larger version (24K): [in this window] [in a new window]   FIG. 6.   Tax expression is higher in IL-2-independent T-cell lines than in IL-2-dependent cells. (a) Expression of Tax proteins in HTLV-1-transformed T-cell lines. Cell lysate was prepared from six IL-2-dependent and three IL-2-independent virus-transformed cell lines. The expression of Tax in these cell lines was measured by Western blot analysis as described in Materials and Methods. (b) Cell growth analysis of HTLV-1-transformed T-cell lines in the absence of IL-2. The indicated cell lines were washed twice with PBS and then cultured in the absence of IL-2 for 8 days. Viable-cell numbers were counted by trypan blue staining. Data are representative of three independent reproducible experiments.

	DISCUSSION

Top Abstract Introduction Materials and methods Results Discussion References

The major finding of the present study was that Tax, in addition to transforming normal T cells in an IL-2-dependent manner, stimulated the cell growth machinery of T-cell lines in the absence of IL-2. Tax converted the cell growth characteristics of the mouse T-cell line from IL-2 dependent to IL-2 independent (Fig. 1). The duration of cell survival and cell growth of HTLV-1-infected T-cell lines in the absence of IL-2 correlated with the level of expression of Tax in these cells (Fig. 6). Our results suggest that Tax plays a critical role in IL-2-independent transformation of T cells by HTLV-1.

Tax enhanced continuous IL-2-independent proliferation of CTLL-2 cells, but it failed to do so in HTLV-1-transformed T-cell lines (Fig. 6). In this regard, Miyazaki et al. (25) have shown that Tax does not modify the cell growth property of a pro-B-cell line from being IL-3 dependent to IL-3 independent and that persistent IL-3-independent growth requires interaction between Tax and either Myc or Lck oncoproteins. Hence, it is possible that CTLL-2 and IL-2-independent HTLV-1-transformed T-cell lines also activate a cellular gene(s) that interacts with Tax to induce IL-2-independent proliferation of T cells.

Several investigators have attempted to establish T-cell lines expressing a large amount of Tax and failed to do so or selected ones expressing a small amount of Tax. The difficulty may be due to the ability of Tax to induce apoptosis of T cells (7). On the other hand, the CTLL-2/Tax cells used in the present study expressed a large amount of Tax, the extent of which was equivalent to that of HTLV-1-transformed T-cell lines (data not shown). Thus, CTLL-2 cells may be resistant to apoptosis induced by Tax.

At least two distinct activities are necessary for IL-2-independent proliferation of CTLL-2 cells; one is the inhibition of apoptosis induced by IL-2-deprivation (Fig. 5), and the other is promotion of the cell cycle (5, 24). The results of the present study suggested that inhibition of apoptosis of CTLL-2 cells by Tax is mediated by activation of NF-B in the following manner (Fig. 3). Tax and Tax703, which are activators of NF-B, inhibit apoptosis of CTLL-2 induced by lack of IL-2 and allow the cells to proliferate in an IL-2-independent manner (Fig. 3). Apoptosis of CTLL-2 induced by deprivation of IL-2 is associated with a reduction in NF-B activity (Fig. 4). On the other hand, Tax-induced enhancement of the cell cycle in CTLL-2 cells may not be mediated via activation of NF-B but, rather, may be due to activation of CDKs. This argument is based on previous studies showing that CDKs are key regulators of the pathways that promote cell cycle and that Tax stimulates their activity by interacting with their inhibitor (INK4) in fibroblasts (19, 38) and in Tax-transformed human T-cell lines (30). It is noteworthy, however, that Tax has a variety of functions other than those discussed here. Thus, further analysis is required to determine the exact function involved in Tax-induced IL-2-independent proliferation of CTLL-2.

TaxM22 was expressed less strongly in CTLL-2 cells than were TaxWT and Tax703 (Fig. 3A). This may suggest that TaxM22, if expressed in sufficient amounts, may induce IL-2-independent proliferation of CTLL-2. However, TaxM22 induced a more rapid apoptosis of CTLL-2 cells in the absence of IL-2, indicating that TaxM22 cannot induce IL-2-independent proliferation of CTLL-2 even when expressed in sufficient amounts.

Our results suggested that the IL-2-independent growth-stimulatory activity of Tax in T cells is dependent on Tax more than is the IL-2-dependent transforming activity (Fig. 6). Among IL-2-dependent HTLV-1-infected T-cell lines, ILT-Mat, with a low level of Tax, rapidly died in the absence of IL-2 whereas other cell lines were resistant to apoptosis to an extent dependent on the level of expression of Tax (Fig. 6). The results of annexin staining experiments also showed that ILT-Mat cells were sensitive to apoptosis induced by lack of IL-2 whereas other cell lines expressing detectable amounts of Tax (as determined by Western blot analysis) were resistant to apoptosis (data not shown). Thus, inhibition of apoptosis by activation of Tax may represent a mechanism for cell survival of not only CTLL-2 cells but also human T-cell lines.

TaxM22 stimulated apoptosis of CTLL-2 (Fig. 5). There are at least two mechanisms to explain this phenomenon. TaxM22 inhibited the activity of NF-B in CTLL-2 cells in the presence of IL-2 (Fig. 2 and 4), and reduced NF-B activity was associated with apoptosis of CTLL-2 cells after deprivation of IL-2 (Fig. 4). Thus, TaxM22-induced inhibition of NF-B may be responsible for the stimulation of apoptosis. Alternatively, the CRE pathway activated by TaxM22 may stimulate apoptosis of CTLL-2 cells (Fig. 2).

IL-2 induces the expression of antiapoptotic genes bcl-2 and bcl-X, and this induction is thought to play a role in the inhibition of apoptosis of IL-2-dependent T cells (24). Tax does not, however, induce the expression of bcl-2 and bcl-X genes in a pro-B-cell line (25), and HTLV-1-transformed IL-2-independent T-cell lines do not express the bcl-2 gene more strongly than IL-2-dependent ones do (3). Thus, Tax may inhibit apoptosis of CTLL-2 and HTLV-1-transformed T-cell lines in a manner different from IL-2.

Another major finding of the present study was the identification of a novel transformation-related activity of Tax (CFSA) in CTLL-2 cells and the finding that such activity was different from that of inducing IL-2-independent proliferation of CTLL-2 cells. Tax703, which exerts an IL-2-independent transformation activity equivalent to TaxWT (Fig. 3), showed 10-fold less CFSA activity than TaxWT did. Inhibition of apoptosis (also called anoikis) is postulated as a mechanism of CFSA of transformed cell lines that originated from fibroblasts (9). Inhibition of apoptosis may also play a role in CFSA of CTLL-2 cells, since apoptosis of CTLL-2/WT cells was less extensive than that of CTLL-2/703 cells (Fig. 5). It should, however, be noted that the high rate of apoptosis of CTLL-2/703 cells was not inhibited by IL-2 (data not shown). Thus, this apoptosis is likely to be distinct from that of CTLL-2 cells induced by deprivation of IL-2.

Our results obtained with Tax mutants suggested that the CFSA activity of Tax is mediated by activities that are diminished in Tax703 (Fig. 2 and Table 1). Thus, it seems that the CRE pathway is required for CFSA. In addition, Tax but not Tax703 activates the transcription via the CArG box enhancer, and the pathway is associated with the transformation of rat embryo fibroblasts by Tax together with the v-Ras oncogene (22). Thus, the CArG box pathway is also another possible candidate for mediating CFSA of CTLL-2/Tax cells.

The present studies identified two distinct transformation-related activities of Tax in CTLL-2 cells, and these two activities were different from each other and from IL-2-dependent transformation activity. Thus, analysis of the respective roles of these activities in T-cells will advance our understanding of the transformation of T cells by HTLV-1.