Persistent inhibition of telomerase reprograms adult T-cell leukemia to p53-dependent senescence

Abstract

The antiviral thymidine analog azidothymidine (AZT) is used to treat several virus-associated human cancers. However, to date the mechanism of AZT action remains unclear and thus, reasons for treatment failure are unknown. Adult T-cell leukemia/lymphoma (ATL) is an aggressive malignancy of poor prognosis. Here, we report that enduring AZT treatment of T-cell leukemia virus I-infected cells, in vitro and in vivo in ATL patients, results in inhibition of telomerase activity, progressive telomere shortening, and increased p14(ARF) expression. In turn, this elicits stabilization and reactivation of the tumor suppressor p53-dependent transcription, increased expression of the cyclin-dependent kinase inhibitor p21(Waf1), and accumulation of p27(kip1), thereby inducing cellular senescence and tumor cell death. While ATL patients carrying a wild-type p53 enter remission following treatment with AZT, those with a mutated p53 did not respond, and patients' disease relapse was associated with the selection of a tumor clone carrying mutated inactive p53.

Figure 1.

Telomerase inhibition induces telomere attrition in HTLV-I–infected cells. (A) Telomerase activity (expressed as telomeric product generated, TPG), measured by TRAP assay in HTLV-I cell lines in absence of or presence of telomerase inhibitors AZT (50 μM) and ddG (10 μM) for 72 hours. (B) Correlation between telomerase activity (TPG) and survival of AZT-treated HTLV-I cell lines (in weeks). (C) Inhibition of telomerase activity by AZT in the HTLV-I–transformed MT-2 cell line untreated or treated with AZT for 18 weeks. (D) Southern blot analysis of telomere shortening following AZT treatment.

Figure 2.

Persistent telomerase inhibition is not associated with apoptosis but induces senescence in HTLV-I–infected cells. (A) Western blot analysis for expression of apoptosis regulators in untreated MT-2 or treated with AZT (18 weeks). Equal amounts (50 μg) of each extract were used and confirmed by β-tubulin. (B) Absence of cleavage for procaspase 3 or PARP in AZT-treated MT-2 cells. (C) Senescence β-gal (SA-β gal) assay in MT-2 cells untreated and treated with AZT (18 weeks). A significant amount of senescence was detected only in end of cultures. (D) Western blot analysis of MT-2 cells treated with AZT. Samples were collected at different times after treatment from 0 to 15 weeks. Expression of caspase 3, Bcl-2, and Bax was tested. Actin was used as loading control. (E) Western blot analysis of MT-2 cells treated with AZT. Samples were collected at different times after treatment from 0 to 15 weeks. Expression of Tax was analyzed as described in “Patients, materials, and methods.” Beta-tubulin was used to confirm equal loading.

Figure 3.

Continuing inhibition of telomerase by AZT in HTLV-I–infected cells induces posttranscriptional stabilization of p53. (A) Western blot analysis for expression of p53 pathway regulators in untreated MT-2 and AZT-treated (18 weeks) MT-2 cells. Equal amounts (50 μg) of each extract were used and confirmed by β-tubulin. (B) Analysis of p53 mRNA expression by RT-PCR in MT-2 and after culture with AZT for 18 weeks. GAPDH was used as internal control for amplification. (C) Western blot analysis of p53 after 24 hours of AZT treatment. Equal amounts (50 μg) of each extract were used and confirmed by β-tubulin. (D) Increased expression of p53 detected by Western blot in several HTLV-I–infected cell lines (1185, C10MJ, LAF) treated for 4 weeks with AZT.

Figure 4.

AZT-mediated reactivation of p53 functions and stabilization of CDKI p21^WAF and p27^KIP. (A) Expression of p21waf and Bax mRNA before and after ionizing radiation in AZT-treated MT-2 cells (18 weeks). GAPDH was used as internal control for amplification. (B) Western blot analysis for expression of p21^WAF and p27^KIP in untreated MT-2 or after culture with AZT for 18 weeks. Equal amounts (50 μg) of each extract were used and confirmed by β-tubulin. (C) Analysis of p21^WAF and p27^KIP mRNA expression by RT-PCR in MT-2 and after culture with AZT for 18 weeks. GAPDH was used as internal control for amplification. (D) Western blot analysis for expression of p27^KIP in untreated MT-2 or after culture with AZT for 18 weeks, in absence or presence of proteosome inhibitor lactacystin. Equal amounts of each extract (50 μg) were used and confirmed by β-tubulin. (E) MT-2–derived p53 cDNA nucleotide sequence. (F) p53 amino acid sequence from MT-2 cells compared with wild-type p53. (G) RT-PCR for p53 responsive genes before and after gamma irradiation of MT2 cells.

Figure 5.

Analysis of CD25⁺ HTLV-I–infected cells. (A) FACS analysis of the presence of CD25 marker expression from CD25⁺ and CD25– sorted cells. (B) Telomerase activity detected by TRAP assay in CD25⁺ and CD25– fractions isolated from an HTLV-I donor and ATL PBMCs. (C) Integrated HTLV-I proviral DNA was detected by PCR using primers in the tax coding region. GAPDH was used as amplification control to ensure proper quality and quantity of extracted DNAs. (D) Staining control of PBMCs in presence or absence of FITC-telomere probe. (E) In situ hybridization of FITC-conjugated telomere probe (top) and DAPI (4′6-diamidino-2-phenylindole 2HCl) (bottom).

Figure 6.

AZT-induced telomerase inhibition and telomere shortening in ATL patients. (A) Inhibition of telomerase activity measured by TRAP in vivo in ATL patients before and during treatment with AZT. (B) FACS analysis of telomere size by Flow-FISH in CD25⁺ and CD25– cells isolated from uncultured ATL patient samples. (C) Detection of exon 1 and 2 of the p14^ARF gene by PCR in CD25⁺ cells from 2 ATL patients before AZT treatment or after AZT treatment; GAPDH amplification was used as control. D indicates diagnosis; R, relapse/death; and CR, complete remission. (D) Detection of provirus integration site by inverse PCR before and after AZT treatment. Arrows indicate different integration sites by inverse PCR.

Figure 7.

Response of ATL patients to AZT-based therapy correlates with their p53 transcriptional status. (A) Immunohistochemistry detection of p53 expression in ATL samples collected from AZT responder (top) and nonresponders (bottom) performed as previously described. (B) p53 from ATL patients was cloned into pCDNA3.1 and tested in a functional assay. Jurkat T cells were transfected with a p53-responsive vector and p53 expression vectors and luciferase activity detected 24 hours later. Results are representative of 2 independent experiments.