JAK–STAT and JAK–PI3K–mTORC1 Pathways Regulate Telomerase Transcriptionally and Posttranslationally in ATL Cells

Adult T-cell leukemia (ATL) is an aggressive lymphoproliferative disorder that occurs in individuals infected with human T-cell leukemia virus type 1 (HTLV-I; refs. 1–3). ATL is classified into 4 subtypes, including smoldering, chronic, lymphoma, and acute types, according to clinical manifestations with distinct molecular mechanisms (4). Treating aggressive ATL is very difficult. In addition, effective therapies have yet to be developed for indolent types of ATL such as smoldering or chronic ATL. More than 50% of such ATL subtypes have been reported to progress to acute ATL and prognosis is poor once this occurs (5). Thus, novel therapy targeting ATL tumor cells during smoldering/chronic phases is mandatory to improve the prognosis of this incurable and debilitating disease. This is reminiscent of chronic myeloid leukemia in which inhibitors of the BCR-ABL tyrosine kinase have been successfully used as a specific therapy for the chronic phase of the disease (6).

The mechanism of leukemogenesis or tumor progression is distinct in each phases of ATL. In the early phase of HTLV-1 infection in T cells, several lines of evidence indicate that the HTLV-1 Tax protein plays a central role in leukemogenesis. For example, Tax activates critical signaling pathways such as NF-kB, PI3K, and AP1. Tax also induces the expression of cytokines, such as interleukin (IL)-2, and their receptors, leading to cell proliferation and transformation. In addition, HTLV-1–infected cells contain elevated activity of human telomerase, an RNA-dependent DNA polymerase composed of a catalytic subunit termed human telomerase reverse transcriptase (hTERT), which elongates telomeres shortened by successive replication cycles. Tax has been shown to upregulate the transcriptional activity of hTERT via the NF-kB pathway, although contradicting results have also been reported (7, 8). Although Tax is required for the IL-2–dependent expansion of HTLV-1–immortalized cells in the early phases of viral infection, T-cells expressing Tax can be eliminated by host immune surveillance due to Tax immunogenicity. Indeed, most ATL cells do not express Tax and tend to become independent of IL-2 (9), in particular in acute ATL. Nevertheless, IL-2 is often required for ATL cell proliferation and survival, suggesting the importance of IL-2 in tumor growth in some patients, such as those suffering from chronic ATL, regardless of Tax protein expression (10, 11).

IL-2 signaling has recently been shown to be associated with the upregulation of the hTERT promoter in Tax-negative HTLV-1–transformed cells (12). These findings suggest that telomerase is an attractive target for anti-cancer drug discovery in ATL; however, the mechanism of IL-2–dependent telomerase activation in Tax-negative ATL tumor cells has not been completely elucidated. The Janus-activated kinase (JAK)–STAT pathway is involved in IL-2 signaling in normal T cells and is activated in IL-2–independent HTLV-1–transformed T cells and ATL cells (13, 14). We previously reported that STAT5 controls telomerase transcription in chronic myeloid leukemia cells and is implicated in resistance to imatinib (15, 16). Thus, the JAK–STAT pathway could be a critical component of telomerase activity regulation in ATL cells. The aim of the current study is to assess the role of the JAK–STAT and the related pathways in regulating IL-2–induced telomerase activity, which promotes cell proliferation in ATL cells derived from chronic ATL patients. Results may lead to the development of new molecularly targeted therapies in chronic/smoldering ATL.

This study used IL-2–dependent ILT-Hod cells derived from chronic ATL patient (17). HUT102 cell line, which has the capacity for IL-2–independent cell growth, was also used (1). ILT-Hod and HUT102 were kind gifts from Dr. Mari Kannagi (Tokyo Medical and Dental University, Tokyo, Japan) and Dr. Masahiro Fujii (Niigata University Graduate School of Medical and Dental Sciences, Niigata, Japan) in 2008, who published the manuscripts using these cell lines (18, 19). Cell identities were not authenticated by the authors other than confirming the cell lines are ATL cells by Southern blot analysis for viral gene integration and status of Tax expression in addition to characteristic T-cell phenotypic markers. For baseline growth, the ILT-Hod cells require the constant addition of IL-2 at a concentration of at least 5 U/mL. For telomerase induction experiments, cells were exposed to a concentration of 100 U/mL. To stimulate the T-cell antigen receptor, ILT-Hod cells were grown on either anti-CD3, anti-CD28, or anti-CD3/CD28 coated plates for 2 days, and the cells were subsequently harvested. Plates were coated at 37°C for 90 minutes before the cells were seeded.

Primary leukemic cells were obtained from 6 patients with ATL. All patients were newly diagnosed (untreated), chronic ATL according to the diagnostic criteria by Shimoyama (4). Leukemic cells were isolated from blood samples using a Ficoll–Conray density gradient, washed twice, and their viability was determined by trypan blue dye exclusion. Cells showing more than 80% viability were used in experiments to exclude unreliable data derived from cell death and protein degradation. Four of 6 samples were found to be suitable for use. All samples were collected after obtaining informed consent, and the study protocol was approved by the Human Investigation Committee of our institution.

LY294002, PD98059, geldanamycin, radicicol, and rapamycin were all purchased from Calbiochem. IL-2 was kindly provided by Takeda Pharmaceutical and Shionogi Pharmaceutical. Polyclonal rabbit antibodies against PI3K, AKT, phospho-AKT(Ser473), p70S6K, mTOR, YB-1, hTERT, HSP90, JAK1, JAK2, JAK3, STAT3, STAT5, phospho-STAT5 (Tyr694), ß-actin, and the anti-CD3 mouse monoclonal antibody (OKT3), anti-CD28 (9.3), STAT5, phospho-tyrosine (4G10), and α-tubulin were used as already reported (16). A monoclonal antibody against Tax1 protein was generated previously (20).

Total RNA was isolated using Isogen (Nippongene), and cDNA was synthesized using the Advantage RT-for-PCR Kit (Clontech) as reported previously (21). The resulting cDNA (25 ng) was subjected to PCR amplification. The relative concentrations of the PCR products were determined by comparing the ratio of the product in each lane to β-actin. The PCR primers used were already reported (15).

Telomerase activity was measured using the telomere repeat amplification protocol (TRAP) as previously described (22). To quantify telomerase activity in each sample, enzyme activity was expressed in arbitrary units as reported previously (15). Internal telomerase assay standard (ITAS) was used as the internal control (23).

Separate aliquots of 2 × 10⁶ cells were transfected with a double-stranded short interfering RNA (siRNA) targeting STAT5A mRNA or a control nonsilencing siRNA (purchased from Dharmacon) using the Amaxa Nucleofector electroporation technique (Amaxa) according to the manufacturer's guidelines. The final concentration of each siRNA was 0.5 μg/mL, and the siRNA sequence targeting STAT5A was designed using siRNA-design software (Dharmacon). One hour after transfection with siRNA, 100 U/mL of IL-2 or solvent alone was added to the cell cultures. After 2 days of culture, the cells were harvested for immunoblot analysis and telomerase activity assays.

Cells (2 × 10⁷) were lysed in radioimmunoprecipitation assay (RIPA) buffer, and precleared samples were incubated with the appropriate antibodies for 2 hours or overnight. Immune complexes were resolved by 5% linear or 5% to 20% gradient SDS-PAGE and transferred to polyvinylidene difluoride membranes (BioRad). Blots were incubated with primary antibodies and processed as already reported (15).

Chromatin immunoprecipitation (ChIP) assays were conducted as described previously (24). Cells were fixed in formaldehyde for 10 minutes at 25°C and then were resuspended in lysis buffer, followed by preclearing with a salmon sperm DNA/protein G-Agarose slurry at 4°C. Each precleared solution was then incubated with STAT5 antibody or normal rabbit serum overnight at 4°C. Immune complexes were separated by incubation with salmon sperm DNA/protein G-Agarose slurry at 4°C. After washing the pellets, DNA elution buffer was added and the samples were heated to 65°C, followed by treatment with 4 mg of RNase at 37°C and 10 mg of proteinase K at 55°C. The DNA was then purified and subjected to PCR as reported before (24).

Equal amounts of cell lysates were precleared with purified preimmune mouse or rabbit IgG and protein G-agarose for 1 hour at 4°C. Cell lysates were incubated with rabbit antisera against hTERT, AKT, HSP90, mTOR, or S6K for 1 hour or overnight at 4°C and then incubated with protein G-agarose for 1 hour at 4°C. Agarose beads were washed 5 times in wash buffer and 2 μL of each immunoprecipitate was subjected to TRAP assay analysis as described before.

Data are expressed as the mean ± SD of 3 or more independent experiments. Statistical analysis was carried out using the 2-tailed Student t test for paired data. P < 0.05 was considered statistically significant.

ILT-Hod cells, an IL-2–dependent T-cell line derived from chronic ATL patients, were IL-2 starved for 6 days and then stimulated with either anti-CD3/CD28 or IL-2 for 48 hours. Stimulation with IL-2 alone was sufficient to induce telomerase activity, whereas nonmitogenic stimulation with anti-CD28 and/or anti-CD3 alone (T-cell receptor stimulation) did not induce telomerase expression (Fig. 1A). For cell starvation and restimulation, ILT-Hod cells cultured with the constant addition of IL-2 were starved of IL-2 and then stimulated with 100 U/mL IL-2 for various time periods. As shown in Fig. 1B, telomerase activity decreased to less than 20% after 6 days and readdition of IL-2 induced telomerase activity by 24 hours and higher levels were observed at 48 hours. Cells that had more than 80% viability were recovered during the experiments. hTERT mRNA was detectable by 24 hours, and higher levels were detected 48 hours after IL-2 stimulation (Fig. 1C). Consistent with the hTERT mRNA results, hTERT protein was detectable by immunoblotting 24 hours after IL-2 stimulation (Fig. 1D). In addition, during IL-2 stimulation, the number of cells in S phase increased together with the induction of telomerase activity (Fig. 1E).

The pleiotropic functions of the HTLV-I Tax protein are thought to cooperate in promoting the proliferation of infected T cells (25). We thus examined whether IL-2 induces Tax expression in ILT-Hod cells. IL-2 did not upregulate Tax expression in the IL-2–dependent ATL cell lines. In contrast, Tax is constitutively expressed in IL-2–independent HUT102 cells (Fig. 2A). Next, IL-2–induced signal transduction pathways downstream of IL-2R activation were examined. Upon binding to receptors, cytokines induce phosphorylation and activation of members of the JAK family, which leads to the recruitment and phosphorylation of STATs. To identify which JAKs are phosphorylated and associated with STAT proteins in cells stimulated with IL-2, cell extracts were immunoprecipitated with antibodies against JAK1, JAK2, JAK3, STAT3, or STAT5 and then immunoblotted with the 4G10 antiphosphotyrosine antibody. Immunoblotting with 4G10 revealed increased phosphorylation of JAK1, JAK2, JAK3, and STAT5 in IL-2–treated ILT-Hod cells and constitutive activation of all of these signaling proteins in HUT102 cells (Fig. 2B). Coimmunoprecipitation assays revealed that JAK1 and/or JAK2 but not JAK3 are involved in the IL-2 signaling events that activate STAT5 in ILT-Hod cells, whereas JAK2 and/or JAK3 are involved in both STAT3 and STAT5 activation in the IL-2–independent growth of HUT102 cells (Fig. 2C).

Because our data indicated that IL-2 could induce tyrosine phosphorylation of STAT5, we then studied whether the activation of STAT5 correlated with its potential to transcriptionally activate telomerase using DNA-binding experiments (Fig. 3A and B). As shown in Fig. 3B, STAT5 became associated with the promoter following activation by IL-2, suggesting that STAT5 is one of the transcription factors that regulates telomerase during IL-2 activation of ILT-Hod cells. To confirm the role of STAT5 in telomerase expression, we analyzed the relationship between STAT5 activation and telomerase activity. ILT-Hod cells were incubated with siSTAT5 or nonsilencing siRNA for 2 days, and lysates were then examined by immunoblot analysis with anti-STAT5. Telomerase activity, together with STAT5, was clearly downregulated within 48 hours of transfection with siSTAT5 (Fig. 3C).

To further clarify the mechanism by which IL-2 increases telomerase activity, we treated cells with various reagents that block critical signaling pathways involved in cell proliferation and cell survival. The viability of ILT-Hod cells cultured in the presence of the inhibitors was not reduced significantly after 48 hours of treatment (Fig. 4A, lower panel). IL-2–induced telomerase activity was abolished in a dose–dependent manner by treatment with LY294002, a specific inhibitor of PI3K; radicicol and geldanamycin, inhibitors of HSP90; and rapamycin, an inhibitor of mTOR, which is a downstream effector of AKT (Fig. 4A, top and middle panels). In contrast, PD98059, a specific inhibitor of MEK1/2 that blocks extracellular signal-regulated kinase (ERK)1/2, did not affect IL-2–induced telomerase activity. We next examined whether IL-2–induced expression of hTERT mRNA or protein was affected by these inhibitors. When cells were treated with 20 μmol/L LY294002, IL-2–induced expression of hTERT mRNA and protein was significantly blocked (Fig. 4B and C). Treatment with 0.1 μmol/L radicicol, 0.1 μmol/L geldanamycin, and 5 μmol/L rapamycin also blocked increased hTERT mRNA and protein levels, but 50 μmol/L PD98059 had no effect on either IL-2–induced hTERT mRNA expression or protein levels, suggesting that the PI3K/AKT/HSP90/mTOR pathway is involved in transcriptional regulation of telomerase activity (Fig. 4B and C).

As shown in Fig. 5A, the low level of constitutive AKT phosphorylation was increased by exposure to IL-2 following IL-2 deprivation. Similarly, 2 other downstream components of the AKT signaling pathway, mTOR and p70S6K, also showed increased phosphorylation (Fig. 5A).

We showed that JAK1, JAK2, and JAK3 were tyrosine phosphorylated in response to IL-2 (Fig. 2B). Although these molecules have been reported to bind PI3K in other cell systems (26, 27), its interaction has not been investigated in ATL cells. To determine whether PI3K associates with JAKs in response to IL-2, coimmunoprecipitation assays were conducted. As shown in Fig. 5B, IL-2 enhanced the association of the PI3K p85 subunit with JAK2, indicating that IL-2 induces a physical interaction between JAK2 and PI3K, as well as between JAK2 and STAT5. To confirm the association of these PI3K signaling pathway molecules with telomerase, we carried out TRAP assays using immunoprecipitates (IP-TRAP) generated with specific antibodies against hTERT, mTOR, S6K, AKT, and HSP90. These immunoprecipitates contained telomerase activity that was reduced by treatment with the corresponding inhibitors of these proteins (Fig. 5C). In contrast, immunoprecipitates using antibodies against either IgG or ERK did not show telomerase activity, being consistent with the results that IL-2–induced telomerase activity was not sensitive to PD98059, a MEK/ERK inhibitor (Fig. 5C). These data indicate that IL-2–induced telomerase activity is also regulated posttranslationally through the JAK2/PI3K/AKT/HSP90/mTORC1 signaling pathway.

We next examined whether the components we identified were also involved in the regulation of telomerase activity in primary leukemic cells. Using 4 primary chronic phase leukemic samples, we conducted TRAP assays and immunoblotting before and after stimulation with 100 U/mL of IL-2 for 2 days. Increased telomerase activity was detected in 3 of 4 samples after IL-2 stimulation (Fig. 6A). Interestingly, an increased level of telomerase protein was observed concomitant with activated STAT5 (Fig. 6B), suggesting that the same components involved in upregulation of telomerase activity in ILT-Hod cells are also involved in the regulation of telomerase activity in primary ATL cells.

In this study, we showed that stimulation of a T-cell line established from a patient with chronic ATL through the IL-2 receptor but not the T-cell receptor induces cell-cycle entry and telomerase activity within 24 hours. Tax was detected in the IL-2–independent HUT102 cells but not in the IL-2–dependent ILT-Hod cells. Leukemic cells generally do not express Tax in vivo due to several regulatory mechanisms (28). Thus, it is unlikely that Tax is involved in the induction or activation of telomerase in this system. Alternative mechanisms independent of the Tax protein may be involved in hTERT expression and telomerase activity in IL-2–dependent chronic ATL cells.

Binding of IL-2 to the IL-2R β and γ chains results in the activation and recruitment of JAK1, JAK2, and JAK3 and the phosphorylation and nuclear translocation of the STATs (13). JAK and STAT activation is associated with the proliferation and survival of IL-2–independent ATL cell lines and primary ATL cells (14). However, the mechanism by which IL-2 signaling is involved in the activation of the JAK–STAT pathway in IL-2–dependent ATL cells is not clear. To ensure that any basal phosphorylation of JAK/STAT proteins was not due to prior exposure to IL-2, the ILT-Hod ATL cells were cultured for 6 days without IL-2. Subsequent stimulation of ILT-Hod cells with IL-2 caused tyrosine phosphorylation of JAK1, JAK2, JAK3, and STAT5. In contrast, the constitutive activation of each of these signaling proteins was observed in IL-2–independent HUT102 cells. Coimmunoprecipitation assays revealed that JAK1 and/or JAK2 but not JAK3 are involved in the IL-2 signaling events that activate STAT5 in ILT-Hod ATL cells, whereas JAK2 and/or JAK3 are involved in both STAT3 and STAT5 activation in the IL-2–independent growth of HUT102 cells. These findings indicate that JAK1/JAK2/STAT5 phosphorylation is triggered by IL-2 stimulation in ILT-Hod ATL cells and that STAT5 is functional and presumably required for IL-2–mediated proliferation and telomerase induction. Intriguingly, IL-2 activated both STAT5 and telomerase in primary samples, suggesting that this signaling pathway is active in ATL tumors in vivo. To further investigate the transcriptional regulation of telomerase by STAT5, ChIP assays were carried out to examine whether STAT5 could directly bind to the promoter region of hTERT. STAT5 was associated with the telomerase promoter after exposure to IL-2, indicating that STAT5 is one of the transcription factors that regulate hTERT expression in IL-2–stimulated ILT-Hod cells. siRNA-mediated STAT5 knockdown resulted in the functional silencing of telomerase activity, confirming the role of STAT5 in this pathway. We have previously observed the direct binding of STAT5 to the promoter region of hTERT in chronic myeloid leukemia (15, 16). To the best of our knowledge, this is the first report of STAT5 as a direct regulator of hTERT transcription in lymphoid cells.

Because IL-2 promotes the entry of ILT-Hod cells into the cell cycle, the activation of telomerase by IL-2 should be associated with cell proliferation. IL-2 activates the ERK and AKT signaling pathways, which are associated with normal T-cell proliferation (29). Therefore, we hypothesized that these pathways might participate in the regulation of IL-2–induced telomerase activity in ATL cells. PD98059, a specific inhibitor of MEK1/2, did not block the IL-2–induced increase in telomerase activity or the expression of hTERT mRNA. In contrast, LY294002, a specific inhibitor of PI3K, blocked IL-2–induced telomerase activity and the expression of hTERT mRNA. Rapamycin, an inhibitor of mTOR, also blocked both telomerase activity and the increase in hTERT mRNA and protein observed in IL-2–stimulated cells. These results suggest that the PI3K/AKT/mTOR pathway transcriptionally regulates telomerase activity in response to IL-2. Moreover, antibodies against AKT, mTOR, and S6K immunoprecipitated the telomerase activity, indicating that AKT, mTOR, and S6K interact directly with active hTERT protein in an IL-2–dependent manner. Thus, the PI3K/AKT/mTORC1 signaling pathway may be implicated in the posttranslational regulation of hTERT.

Our studies also revealed a possible role for HSP90 in regulating telomerase activity. HSP90 is an abundant cytosolic protein that acts in concert with cochaperones, such as HSP70 and p23, to prevent aberrant protein folding, which can lead to protein inactivation and aggregation (30). HSP90 has been shown to be required for efficient telomerase assembly (31). In ATL cells, IL-2–induced telomerase activity was blocked by geldanamycin and radicicol, which are inhibitors of HSP90. These compounds also inhibited hTERT mRNA expression and the corresponding increase in hTERT protein, indicating that HSP90 is involved in the transcriptional modulation of hTERT. Furthermore, HSP90 immunoprecipitates of IL-2–stimulated ATL cells contained active telomerase. Thus, HSP90 may act as an hTERT chaperone. These findings indicate that PI3K, AKT, mTORC1, and HSP90 upregulate hTERT in both a transcriptional and posttranslational manner in IL-2–dependent ATL cells. This mechanism is different from what was observed in NK cell tumors, in which mTOR and HSP90 are only implicated in the posttranslational regulation of hTERT (32). Recently, Bellon and colleagues reported that IL-2 signaling is associated with PI3K-dependent transcriptional upregulation of hTERT through the suppression of WT-1 in Tax-negative HTLV-1–transformed cells (12). This mechanism might also contribute to the IL-2–induced telomerase activity observed in this ATL system, even in the absence of Tax protein.

Upon IL-2 stimulation, PI3K has been shown to localize to the IL-2R in a murine T-cell line (33, 34). We observed that IL-2 stimulation of ATL cells leads to the phosphorylation of downstream effectors of PI3K, such as mTOR, p70S6K, and AKT, and found that JAK2 coimmunoprecipitated with the p85 subunit of PI3K in an IL-2–dependent manner, consistent with previous evidence that JAK2 can bind to the N-terminal SH2 domain of the PI3K p85 subunit in other hematopoietic systems (26, 27). The current data indicate that IL-2–induced activation of JAK2 occurs upstream not only of STAT5 but also of the PI3K pathway. Therefore, JAK2 may be one of the key molecules in the regulation of IL-2–induced telomerase activity in ATL cells.

In conclusion, this study shows that IL-2 stimulation induces hTERT activation in IL-2–dependent T-cell line established from a patient with chronic ATL and in primary chronic ATL cells. IL-2 is required for tumor growth in smoldering/chronic ATL and even in some acute ATL patients (11, 35). Therefore, the dissection of the mechanism by which IL-2 promotes tumor formation in such ATL cells is important to develop effective therapies for ATL, particularly before progression to aggressive ATL. We showed that IL-2–induced telomerase activation involves transcriptional and posttranslational regulation through dual signaling pathways, JAK/STAT and JAK/PI3K/AKT/HSP90/mTORC1 (Fig. 6C). These results shed light on the mechanism of ATL cell proliferation in response to IL-2 and uncovered novel therapeutic targets in ATL (36).

No potential conflicts of interest were disclosed.

Conception and design: O. Yamada, K. Ozaki, M. Akiyama, K. Kawauchi

Development of methodology: O. Yamada, K. Ozaki, K. Kawauchi

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): O. Yamada, K. Kawauchi

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): O. Yamada, K. Ozaki, K. Kawauchi

Writing, review, and/or revision of the manuscript: O. Yamada, K. Kawauchi

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): O. Yamada, K. Ozaki, K. Kawauchi

Study supervision: O. Yamada, K. Kawauchi

The authors thank Mari Kannagi and Masahiro Fujii for their kind gift of cell lines, Naoki Mori and Yuetsu Tanaka for providing primary cells and a Tax antibody, and Atsushi Hatano and Tomohiro Watanabe for technical support.

This work was partially supported by Japan Society for the Promotion of Science KAKENHI (22501053), and supported by the Program for Promoting the Establishment of Strategic Research Centers and by Special Coordination Funds for Promoting Science and Technology granted by the Ministry of Education, Culture, Sports, Science and Technology, Japan.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.