Inhibition of the Htert Promoter by the Proto-Oncogenic Protein TAL1

Inhibition of the hTERT promoter by the proto-oncogenic protein TAL1

J-M Terme1, V Mocquet1, A-S Kuhlmann2, L Zane3, F Mortreux3, E Wattel3, M Duc Dodon2 and P Jalinot1

1Laboratoire de Biologie Molećulaire de la Cellule, Unite´ Mixte de Recherche 5239, Centre National de la Recherche Scientifique, Ecole Normale Supe´rieure de Lyon, Institut Fe´de´ratif de Recherche 128 Biosciences Lyon Gerland, Lyon cedex 07, France; 2Virologie Humaine, Unite´ 758, Institut National de la Sante´ et de la Recherche Me´dicale, Ecole Normale Supe´rieure de Lyon, Institut Fe´de´ratif de Recherche 128 Biosciences Lyon Gerland, Lyon cedex 07, France and 3Unite´ d’Oncovirologie et de Biothe´rapies, Formation de Recherche en Evolution 3011, Centre National de la Recherche Scientifique, Centre León Be´rard, Lyon cedex 03, France

Telomerase activity, which has fundamental roles in develop- expression is regulated by many transcription factors (for a ment and carcinogenesis, strongly depends on the expression review, see Kyo et al.8). Its proximal transcription promoter of human telomerase reverse transcriptase (hTERT), its catalytic subunit. In this report, we show that the basic helix-loop- sequences include two E-boxes and five Sp1-binding sites. These helix factor, TAL1 (T-cell acute lymphoblastic leukemia 1), is a E-boxes have been shown to be regulated positively by the Myc– 9–12 negative regulator of the hTERT promoter. Indeed, TAL1 Max complex and negatively by Mad–Max. overexpression leads to a decrease in hTERT mRNA abundance An interesting model of hTERT regulation focused on the and hence to reduced telomerase activity. Conversely, suppres- opposite effects exerted by the Tax and HBZ (HTLV-1 basic sion of TAL1 by RNA interference in Jurkat cells increases leucine zipper) proteins expressed by the human T-cell hTERT expression. Analysis by chromatin immunoprecipitation assays showed that TAL1 binds to the hTERT proximal leukemia virus type 1 (HTLV-1), the etiological agent of adult promoter and recruits HDAC1. Considering the relationship T-cell leukemia (ATL). Indeed, it has been shown that the Tax recently established between TAL1 and the human T-cell transactivator decreases the activity of the hTERT promoter leukemia virus type 1 (HTLV-1) Tax protein, which was through the first E-box.13 Conversely, the HBZ protein, a factor confirmed in T lymphocyte clones derived from adult T-cell expressed by the antisense strand of the provirus, activates in leukemia patients, we analyzed the effect of TAL1 with respect cooperation with JunD, the hTERT promoter activity.14 This led to the earlier characterized effects of Tax and HBZ (HTLV-1 basic leucine zipper) on hTERT expression. TAL1 was observed to a model suggesting that early after infection Tax decreases to reinforce the negative effect of Tax, whereas hTERT telomerase activity, thereby favoring genomic instability. At a transactivation by the HBZ–JunD complex was repressed by later stage, Tax expression is repressed and the expression of TAL1 overexpression. Moreover, HBZ was found to induce HBZ reactivates telomerase expression allowing evolution proteasome-mediated degradation of TAL1. These observa- toward fully transformed leukemic cells.14 tions support a model in which Tax and TAL1 by repressing Recently, we have shown that TAL1 (T-cell acute lympho- hTERT would initially favor genomic instability, whereas expression of factors such as HBZ allows at a later stage an blastic leukemia 1), which is known to have a key function in increase in hTERT production and consequently in telomerase T-cell acute lymphoblastic leukemia (T-ALL) by impairing 15 16 activity. differentiation, is a target of Tax. Indeed, through activation Leukemia (2009) 23, 2081–2089; doi:10.1038/leu.2009.131; of its transcription promoter 1b, Tax increases TAL1 expression. published online 9 July 2009 By considering these data, we asked whether TAL1 is involved Keywords: T-ALL; hTERT; TAL1; HTLV-1; Tax; HBZ in hTERT regulation both in T-ALL and ATL cells. This factor is known to activate or repress many cellular genes.17 This class II basic helix-loop-helix factor, which has a key function in the differentiation of both hematopoietic and endothelial cells, has Introduction been shown to associate with E-boxes as a heterodimer with E protein as E47.18 TAL1 has also been shown to associate with Telomerase enzyme, which has the unique property of increas- many different transcriptional regulators, including the LIM-only ing the length of the telomeres, has a key function in cancer 1,2 proteins LMO1 and LMO2, GATA factors, Sp1, mSIN3A and cells. It has been shown that telomerase is active in many 19–23 3,4 p300. Depending on the combination of factors in which leukemic cells. Telomerase consists of an RNA molecule, TAL1 is engaged, its effect on a target transcriptional promoter is hTR, and of a catalytic protein subunit, human telomerase 5 either positive or negative. reverse transcriptase (hTERT). The high telomerase activity in We show in this report that TAL1, similar to Tax, is a repressor leukemic cells is often linked to an elevated expression of 6 of hTERT promoter activity. This effect relies on the Sp1 sites hTERT, and in various types of leukemia a decrease in hTERT present in the hTERT promoter. Interestingly, Tax and TAL1 can expression has been linked to reduced proliferation and to 7 cooperate for hTERT repression, but their effect is counteracted by higher sensitivity to drug treatment. The hTERT subunit JunD and HBZ. Our results suggest that TAL1 by inhibiting hTERT might favor genome instability both in T-ALL and ATL cells. Correspondence: Dr P Jalinot, Laboratoire de Biologie Molećulaire de la Cellule, Unite´ Mixte de Recherche 5239, Centre National de la Recherche Scientifique, Ecole Normale Supe´rieure de Lyon, Institut Materials and methods Fe´de´ratif de Recherche 128 Biosciences Lyon Gerland, 46 Alleé d’Italie, 69364 Lyon cedex 07, France. E-mail: [email protected] Constructs Received 29 October 2008; revised 27 May 2009; accepted 2 June pSGF-TAL1, pSUPER-TAL1 and pSG5-Tax have been described 2009; published online 9 July 2009 earlier.16 pSG5-MYC-TAL1 was generated by amplifying it with hTERT inhibition by TAL1 J-M Terme et al 2082 the appropriate primers, the TAL1-coding sequences from diluted 1:1000 or as indicated by the manufacturer, and detection a plasmid including the complete human TAL1 open-reading was performed by chemiluminescence using enhanced chemilu- frame, provided by D Mathieu. The PCR product was inserted minescence (ECL) or ECL þ kits (GE Healthcare). between the BglII and XhoI restriction sites in the pSG5-MYC Antibody against HTLV-1 Tax was obtained from the AIDS vector. pCDNA3-HBZ-myc and pCMV-JunD-Flag have been Research and Reference Reagent program catalog. The follow- described earlier.14 pCMV-Sp1 was kindly provided by Katie ing antibodies were purchased: FLAG (clone M2; Sigma- Flaig.24 pGL3-hTERT3300, pGL3-hTert-378 and pGL3-hTert- Aldrich), b-actin (Sigma-Aldrich) and MYC (clone 9E10; 181 wild-type (WT), mt1, mt2, mt and mutSp1 were kindly Sigma-Aldrich). provided by Satoru Kyo.10 pCMV-p300 was kindly provided by Louis Gazzolo.25 Real-time quantitative telomeric repeat amplification protocol Cell culture, transfection and T-cell limiting dilution Stably transfected HeLa cells were trypsinized, washed twice in cloning phosphate-buffered saline and resuspended in 3-((3-cholamido- Jurkat cells were cultured in RPMI 1640 medium with 10% fetal propyl)-dimethylammonio)-1-propanesulfonate (CHAPS) lysis calf serum (FCS) (Sigma-Aldrich, St Louis, MO, USA), 100 U/ml buffer (10 mM Tris-HCl (pH 7.5), 1 mM MgCl2,1mM EGTA, penicillin and 100 mg/ml streptomycin (Gibco Invitrogen, San 0.5% CHAPS (w/v), 10% (w/v) glycerol, 0.1 mM phenylmethyl- 7 Diego, CA, USA) at 37 1C in a 5% CO2-humidified atmosphere. sulfonyl fluoride, 5 mM b-mercaptoethanol) at 10 cells per ml. HeLa cells were cultured in Dulbecco’s modified Eagle’s After 30 min lysis at 4 1C, lysates were centrifuged at 14 000 g for medium with 10% FCS (Gibco), 100 U/ml penicillin and 30 min at þ 4 1C and the supernatants were stored at 80 1C. 100 mg/ml streptomycin at 37 1C in a 5% CO2-humidified Samples were diluted at a final concentration of 300 ng/ml. The atmosphere. For transfection, FCS concentration was reduced sample dilution (1 ml) was used to perform the real-time to 5%. Jurkat cells were transfected using the Nucleofector kit quantitative telomeric repeat amplification protocol (RQ-TRAP) (Amaxa, Cologne, Nordrhein-Westfalen, Germany). HeLa cells assay with 5 ng/ml of TS primer (50-AATCCGTCGAGCAGAGT were stably transfected with the pCEP-FLAG-TAL or the control T-30), 2.5 ng/ml of ACX primer (50-GCGCGGCTTACCCT 16 0 pCEP-FLAG-GFP constructs, as described earlier. Expression TACCCTTACCCTAACC-3 ), 0.5 mM MgCl2 and the reactions vectors were transfected using the calcium phosphate precipita- were carried out in 20 ml reactions containing the LC-FastStart tion method as described earlier.16 To achieve inhibition of the DNA Master SYBR Green I mix (Roche, Basel, Basel-Stadt, proteasome, we added MG132 (Sigma-Aldrich) at 10 mM for 6 h. Switzerland). Using the LightCycler (Roche), we incubated the Cellular extracts were normalized with respect to protein samples for 30 min at 30 1C, for 10 min at 95 1C and 40 cycles concentrations that were quantified with the DC protein assay were performed: 30 s at 95 1C and 90 s at 60 1C. The values kit (Bio-Rad, Hercules, CA, USA). For immunoprecipitations, represented correspond to the means of triplicate reactions from cellular extracts adjusted to equal protein concentrations were one representative experiment (out of three experiments) with incubated for 2 h with antibodies diluted 1:250. Protein A beads standard deviation. (40 ml) were then added to the mixes that were then incubated for 1 h. Beads were collected by centrifugation and washed three times in lysis buffer. Proteins were eluted in 2 SDS Luciferase assays sample buffer at 80 1C for 10 min. Peripheral blood mono- HeLa cells (1.4 105 cells) were transfected with 1.5 mgoffirefly nuclear cells (PBMCs) were cloned in limiting dilution (0.1 cell luciferase constructs (promoter hTERT 3300, 378 or 181, per well) in Terasaki plates after removal of adherent cells. The either WT or mutated) and 15 ng of thymidine kinase Renilla- medium used for T lymphocytes was RPMI 1640 containing luciferase (pRLTK) by calcium phosphate precipitation. Reporter penicillin and streptomycin, sodium pyruvate, nonessential gene analysis was performed 48 h after transfection for HeLa amino-acid solution, 2-mercapto ethanol, 10% filtered human cells or 24 h after nucleofection of Jurkat cells by using a Dual- AB serum and 100 U/ml recombinant IL-2 (Chiron Corporation, Luciferase Reporter Assay System (Promega, Madison, WI, Emeryville, CA, USA). The cloning medium was the medium USA). The luciferase activity associated with each construct used for T lymphocytes supplemented with 1 mg/ml PHA (Abbott was normalized on the basis of pRLTK activity. The values Murex HA 16, Abbott, Abbott Park, IL, USA) and 5 105 per ml correspond to the means of triplicate reactions from one irradiated (30 Gy) allogeneic PBMCs (feeder cells). The Terasaki representative experiment (out of three experiments) with plates were stored at 37 1C for 10 days in aluminum foil, then standard deviation. One and two stars indicate a Student’s screened for growing cells under the microscope. Positive t-test P-value less than 0.05 and 0.01, respectively. cultures were transferred to 96-well U-bottom plates in medium for T lymphocytes and re-stimulated after a few days. T lymphocytes were re-stimulated every 14 days with PHA Real-time quantitative RT-PCR (1 mg/ml) and fresh feeder cells (106 per ml). Lethally irradiated Total RNAs were extracted from frozen cells using the RNeasy PBMCs from three distinct allogeneic, HTLV-I-negative donors mini kit (Qiagen, Hilden, Nordrhein-Westfalen, Germany). were used as feeder cells to exclude the possibility of clones One-step reverse transcription (RT)–PCR reactions were becoming infected in vitro. To preserve the original growth performed using the QuantiTect SYBR Green characteristics of the cells, we maintained clones in this way for RT-PCR kit (Qiagen) and the LightCycler apparatus (Roche). no more than 4 months, after which a fresh aliquot was thawed. Sequences of sense and antisense primers were as followsF TAL1: 50-CAGCCTAGTGGCTTGTCCTC-30,50-GGAGCCTGA AATTGAATGGA-30; b-actin: 50-GGACTTCGAGCAAGAGAT Immunoblot GG-30,50-AGCACTGTGTTGGCGTACAG-30; h-TERT: 50-TGT After separation by SDS–polyacrylamide gel electrophoresis, TTCTGGATTTGCAGGTG-30,50-GTTCTTGGCTTTCAGGATG proteins were transferred to a polyvinylidene difluoride membrane G-30; Tax: 50-ATCCCGTGGAGACTCCTCAA-30,50- CCAAACAC (GE Healthcare). For immunoblot, primary antibodies were GTAGACTGGGTATCC-30.

Leukemia hTERT inhibition by TAL1 J-M Terme et al 2083 Chromatin immunoprecipitation assay 1.2 Chromatin immunoprecipitation (ChIP) assays were performed essentially by using the Upstate Biotechnology Inc. (Charlottes- 1 ville, VA, USA) recommendations with minor modifications. 6 Formaldehyde cross-linked chromatin from 5 10 cells per 0.8 antibody was used for each immunoprecipitation. Cross-linking reactions were quenched with 125 mM glycine, cells were lysed 0.6 * and chromatin was sonicated to obtain an average DNA length of 500 bp. After centrifugation, the chromatin was diluted 0.4 10-fold and pre-cleared with protein A-agarose containing salmon sperm DNA and bovine serum albumin (Upstate 0.2 Biotechnology). Pre-cleared chromatin (2 ml) was incubated overnight at þ 4 1C with 10 mg of antibody to Sp1, to HDAC1 or Relative hTERT mRNA amount 0 to p300 (sc-14027, sc-6928 or sc-585, respectively; Santa Cruz pSG5 pSG-TAL1 Biotechnology, Santa Cruz, CA, USA), or anti-FLAG antibody followed by protein A-agarose immunoprecipitation. Eluted 1.2 protein–DNA cross-links were reversed by heating at 65 1C overnight, and 25% of the recovered DNA was used in the PCR 1 reaction to amplify the 278 bp region of the hTERT proximal promoter with the Phusion high-fidelity DNA polymerase 0.8 enzyme (Ozyme, Saint Quentin Yvelines, Yvelines, France) along with the forward primer (190/171) 50-CACAGACGCC ** 0.6 CAGGACCGCG-30 and the reverse primer ( þ 69/ þ 88) 50-G CGCGCGGCATCGCGGGGGT-30. A control PCR (negative 0.4 control leading to a 150 bp fragment) was also performed from a region devoid of Sp1 sites within the distal region of hTERT 0.2 using forward primer (2916/2897) 50-GGCAGGCACGAGT GATTTTA-30 and reverse primer (2782/2763) 50-CTGAGG Relative hTERT mRNA amount CACGAGAATTGCTT-30 to show the specificity of Sp1 sites 0 located on the proximal hTERT in the ChIP assay. DNA samples pCEP-FGFP pCEP-FTAL1 recovered from chromatin samples before immunoprecipitation, 1.1 which corresponds to 1% of chromatin samples included in each immunoprecipitation reaction, were also PCR-amplified as loading controls. The PCR reactions for hTERT were processed 1 through 32 cycles of 98 1C for 10 s and 71 1C for 30 s, followed by one cycle for 7 min at 72 1C. PCR products were separated on 0.9 2% agarose gel and visualized with ethidium bromide staining.

0.8 ** Results 0.7

Inhibition of hTERT expression by TAL1 Relative telomerase activity To examine the effect of TAL1 on hTERT expression, we 0.6 transfected HeLa cells either with a control or with a TAL1 expression vector, and mRNA produced by the endogenous pCEP-FGFP pCEP-FTAL1 htert gene was quantified by quantitative RT-PCR (qRT-PCR). In Figure 1 T-cell acute lymphoblastic leukemia 1 (TAL1) regulates these cells, which lack endogenous TAL1, expression of this telomerase activity by modulating human telomerase reverse tran- protein leads to a 40% drop in the amount of hTERT mRNA scriptase (hTERT) mRNA amount. (a) Transfected TAL1 regulates (Figure 1a). A similar effect was observed in HeLa cells stably endogenous hTERT mRNA. HeLa cells were transfected with 1 mgof expressing TAL1 from an Epstein–Barr virus (EBV)-derived pSGF-TAL1. Total RNAs were extracted and analyzed by quantitative RT-PCR. (b) Regulation of hTERT mRNA in TAL1 stably transfected episomic vector (Figure 1b). In this experiment, cells trans- cells. HeLa cells were stably transfected with the pCEP-FLAG-TAL or formed by a similar vector expressing green fluorescent protein the control pCEP-FLAG-GFP constructs. After selection of stably (GFP) were taken as control. Considering this negative effect of transfected cells, total RNAs were prepared and analyzed by real-time TAL1 on hTERT expression, we further examined how it affected quantitative RT-PCR. Results are represented as in a.(c) Regulation of the telomerase activity. This was measured by an RQ-TRAP telomerase activity in TAL1 stably transfected cells. Stably transfected assay using HeLa cells transformed by the GFP and TAL1 HeLa cells were lysed in 3-((3-cholamidopropyl)-dimethylammonio)- 1-propanesulfonate (CHAPS) buffer and analyzed by real-time episomic vectors. In agreement with the earlier data, an quantitative telomeric repeat amplification protocol (RQ-TRAP) assay. approximate 30% drop in telomerase activity was observed The telomerase activity is expressed as a ratio FLAG-TAL/FLAG-GFP. when TAL1 was expressed (Figure 1c). The effect of TAL1 on One and two stars indicate a Student’s t-test P-value less than 0.05 and hTERT expression was also examined in lymphoid cells. In the 0.01, respectively. leukemic T-cell line Jurkat, overexpression of TAL1 markedly reduced in a dose-dependent manner the activity of a reporter construct driving luciferase production under the control of the expression (Figure 2c) led to an increase in the activity of the hTERT promoter (Figure 2a). Conversely, expression in these reporter construct (Figure 2b). This effect showed that the cells of an shRNA that was effective in reducing TAL1 endogenous level of TAL1 present in these cells exerts a negative

Leukemia hTERT inhibition by TAL1 J-M Terme et al 2084 2500 effect on transcription driven by the hTERT promoter. Collec- tively, these data suggest that TAL1 is a negative regulator of 2000 hTERT transcription in T-ALL cells.

1500 Interference of TAL1 with stimulation of the hTERT 1000 promoter activity by Sp1 To better understand this negative effect of TAL1 on the hTERT 500 promoter activity, we set out to characterize the promoter

Relative luciferase activity 0 hTERT

TAL1: - + ++ Luc -181 WT E E 4000 3500 ** Luc -181 mE1 E E 3000 2500 2000 Luc -181 mE2 E E 1500 1000 500 Luc -181 mE Relative luciferase activity 0 E E shRNA ctlsh RNA TAL1

shRNA ctlsh RNA TAL1 Luc -181 mSP1 E E

I.B. α-FLAG

-181 WT -181 mE1 25 35 I.B. β-actin 30 20 25 12 15 20 Figure 2 Endogenous T-cell acute lymphoblastic leukemia 1 (TAL1) 10 15 ** regulates human telomerase reverse transcriptase (hTERT) promoter in ** 10 leukemic Jurkat cells. (a) Jurkat cells were transfected with 3 mgof 5 pGL3-hTERT3300 and either pCMV or increasing amounts of pCMV- 5 TAL1 (1 and 2 mg). At 24 h after transfection, the luciferase activity was 0 0 Relative luciferase activity measured and is represented. (b) The experiment was performed as 1 2 Relative luciferase activity 1 2 described in a, except that hTERT3300 was co-transfected with vectors expressing a control shRNA or an shRNA targeting TAL1. (c) HeLa cells were transfected with vectors expressing either control or -181 mE2 -181 mE anti-TAL1 shRNA. Protein extracts of these cells were analyzed by 30 120 immunoblotting using antibodies to FLAG (top) or b-actin (bottom). 25 100 Two stars indicate a Student’s t-test P-value less than 0.05 and 0.01, respectively. 20 80 15 60 ** 40 10 ** 5 20 0 0 Relative luciferase activity Figure 3 Requirement of Sp1-binding sites in human telomerase Relative luciferase activity 1 2 reverse transcriptase (hTERT) regulation by T-cell acute lymphoblastic 1 2 leukemia 1 (TAL1). (a) Schematic representation of hTERT 181 promoters wild type (WT) and bearing mutations in the distal E-box -181 mSP1 (mE1), in the proximal E-box (mE2), in both motifs (mE) and in the ﬁve 0.8 Sp1-binding sites located between E1 and E2 (mSP1). (b–e) E-boxes are 0.7 not required for hTERT regulation by TAL1. HeLa cells were 0.6 transfected with 300 ng of pSGF (lane 1) or pSGF-TAL1 (lane 2), 0.5 ** together with 1.5 mg of hTERT 181 WT (b), mE1 (c), mE2 (d) and mE 0.4 (e). Cell extracts were prepared 48 h after transfection and analyzed for luciferase activity. (f) Sp1-binding sites are required for hTERT 0.3 regulation by TAL1. HeLa cells were transfected with 300 ng of pSGF 0.2 or pSGF-TAL1, together with 1.5 mg of hTERT181 mSP1 and 0.1 luciferase activity was analyzed as described above for panels b–e. 0 Two stars indicate a Student’s t-test P-value less than 0.05 and 0.01, Relative luciferase activity 1 2 respectively.

Leukemia hTERT inhibition by TAL1 J-M Terme et al 2085 60

Relative luciferase activity 0 1234 TAL1: --+ + Tax: - - + +

α-Tax >

1 243

Figure 5 T-cell acute lymphoblastic leukemia 1 (TAL1) inﬂuences human telomerase reverse transcriptase (hTERT) deregulation by human T-cell leukemia virus type 1 (HTLV-1) proteins. (a) TAL1 cooperates with HTLV-1 Tax to inhibit the hTERT promoter. HeLa cells were transfected with 1.5 mg of hTERT378 wild type, along with 150 ng of pSG5 as control (lane 1) or pSG-Tax (lanes 3 and 4) and 300 ng of pSGF-TAL1 (lanes 2 and 4). Cell extracts were prepared 48 h after transfection and analyzed for luciferase activity. (b) An aliquot of the transfected cells was analyzed by immunoblot with an antibody directed against Tax.

Table 1 Correlations between TAL1 and hTERT expression in HTLV-1 infected or noninfected T lymphocytes

Pearson’s Signiﬁcance correlation level

HTLV-1-positive CD4+ clones (n ¼ 8) TAL1-Tax +0.799 0.009 TAL1-hTERT 0.224 0.297 Tax-hTERT 0.245 0.279 Figure 4 T-cell acute lymphoblastic leukemia 1 (TAL1) competes with human telomerase reverse transcriptase (hTERT) induction by Sp1 ATL cells (n ¼ 4) and p300 and recruits HDAC1. (a) TAL1 prevents Sp1 induction of TAL1-hTERT 0.521 0.145 hTERT promoter. HeLa cells were transfected with 100 ng of pCMV (lane 1) or pCMV-Sp1 (lanes 2–5), together with 1.5 mg of hTERT378 Noninfected T lymphocytes (n ¼ 14) wild type and increasing amount of pSGF-TAL1 (lanes 2–5; 25, 100 TAL1-hTERT 0.190 0.110 and 200 ng). Cell extracts were prepared 48 h after transfection and Expression of TAL1, hTERT and Tax was quantified by qRT-PCR in analyzed for luciferase activity. (b) TAL1 hinders p300 induction of three different types of T lymphocytes: HTLV-1-positive CD4+ clones hTERT promoter. HeLa cells were transfected with 4 mg of pCMV (lane obtained by limiting dilution (see Materials and methods), ATL cells and 1) or pCMV-p300, together with 1.5 mg of hTERT378 wild type and noninfected T lymphocytes prepared from patients. The HTLV-1- increasing amount of pSGF-TAL1 (lanes 2–6; 100, 200, 300 and positive CD4+ clones and noninfected T lymphocytes (CD4+ or CD8+) 600 ng). Cell extracts were prepared 48 h after transfection and were cultured both under nonactivation and activation conditions. The analyzed for luciferase activity. (c) Chromatin immunoprecipitation qRT-PCR results normalized with respect to a housekeeping gene (ChIP) assays were performed with extracts of HeLa cells transformed were analyzed for Pearson’s correlation using the SPSS, Chicago, with Epstein–Barr virus (EBV) episomic vectors expressing either Illinois, USA (HTLV-1-positive CD4+ clones) or Excel Microsoft, FLAG-GFP (lanes 1, 3, 5, 7, 9 and 11) or FLAG-TAL1 (lanes 2, 4, 6, 8, Redmond, Washington, USA (ATL cells and noninfected T lympho- 10 and 12). Immunoprecipitations were performed without (lanes 1 cytes) software. Together with the Pearson’s correlation coefficients the and 2) or with antibodies to FLAG (lanes 3 and 4), HDAC1 (lanes 5 significance levels are indicated. and 6), Sp1 (lanes 7 and 8) or p300 (lanes 9 and 10). DNA extracted from immunoprecipitates was PCR-amplified with primers specific for either the proximal (upper panel) or the distal (lower panel) hTERT promoter. observed similarly with the three constructs starting at positions 3300, 378 and 181 with respect to the transcription start site (Supplementary Figure 1). This effect of TAL1 was dose dependent (Supplementary Figure 2). This indicated that the sequence mediating the activity of TAL1. As a first step, reporter TAL1 effect is exerted on the proximal part of the promoter constructs with various portions of the 50 upstream promoter that includes two E-boxes and a series of five Sp1-binding sites sequences were tested. The negative effect of TAL1 was (Figure 3a). When E-boxes were mutated either individually

Leukemia hTERT inhibition by TAL1 J-M Terme et al 2086 mutations of both E-boxes than on the WT construct (Figure 3, 2000 compare b and e). By contrast, mutation of the ﬁve Sp1-binding 1800 1600 sites led to a reduction in the intensity of the TAL1 effect on the 1400 promoter (Figure 3, compare b and f). This result indicated that 1200 TAL1 might reduce the hTERT promoter activity through an 1000 effect on Sp1. To further test this point, we examined the effect 800 of TAL1 overexpression on stimulation of the hTERT promoter 600 activity by Sp1. Transfection of cells with an Sp1 expression 400 vector increased the activity of the hTERT promoter (Figure 4a, 200 compare lanes 1 and 2), but co-expression of increasing Relative luciferase activity 0 amounts of TAL1 was able to block this activity (Figure 4a, 1 2 3 4 5 compare lanes 1, 2 and 5). A similar effect was observed with HBZ / the p300 cofactor (Figure 4b). These results show that TAL1 JunD: - + + + + expression interferes with the activation of the hTERT promoter TAL1: - - by the DNA-binding Sp1 transcription factor and the p300 cofactor, both well-established positive regulators of hTERT. To further understand how TAL1 modiﬁes Sp1 and p300 associa- 1400 tion with the hTERT promoter, we performed ChIP experiments 1200 using cells transformed with episomic vectors expressing either FLAG-GFP or FLAG-TAL1. These assays were performed with 1000 two primer couples amplifying either the distal or the proximal 800 hTERT promoter. Interestingly, this approach revealed that 600 FLAG-TAL1 associates with the proximal hTERT promoter (Figure 4c , upper panel, compare lanes 3 and 4). This was 400 also observed by performing a DNA precipitation assay using 200 extracts of these cells (Supplementary Figure 3). The ChIP assays

Relative luciferase activity 0 also showed that TAL1 increased Sp1 binding to the proximal 1 23promoter, while slightly reducing p300 association (Figure 4c , HBZ / upper panel, compare lanes 7 and 8 and lanes 9 and 10, JunD: - + + respectively). As the binding of both Sp1 and p300 to the shRNA proximal hTERT promoter did not provide a clear explanation of α-TAL1: - - + the effect of TAL1, we also tested the histone deacetylase (HDAC1), which has been shown earlier to form a complex with TAL1 and mSIN3A.19 In the absence of TAL1, HDAC1 did not 0.75 associate with the hTERT proximal promoter, whereas in the 0.7 presence of TAL1 a clear signal was detected (Figure 4c , upper 0.65 panel, compare lanes 5 and 6). None of these factors were 0.6 observed to bind to the distal hTERT promoter. Collectively, 0.55 these results indicate that the repression of the hTERT proximal 0.5 promoter correlates with TAL1 and HDAC1 binding. 0.45 0.4 0.35 Cooperation between Tax and TAL1 to repress hTERT 0.3 and impairment of this negative effect by HBZ Relative hTERT mRNA amount Earlier studies have shown that hTERT expression was modu- ctl TAL1 HBZ/Jun DHBZ/JunD 13,14 + TAL1 lated by proteins expressed by the HTLV-1 provirus. It has been established that the Tax transactivator has a negative effect Figure 6 (a) T-cell acute lymphoblastic leukemia 1 (TAL1) hinders on hTERT expression, whereas the HBZ protein expressed from human telomerase reverse transcriptase (hTERT) promoter induction the reverse strand of the provirus26 was able to stimulate this by human T-cell leukemia virus type 1 (HTLV-1) HTLV-1 basic leucine expression in combination with JunD. We also observed earlier zipper (HBZ). Jurkat cells were transfected with 3 mg of hTERT3300 together with 2 mg of vectors expressing HBZ and JunD (lanes 2–5) a cross talk between Tax and TAL1, which mutually reinforced along with increasing amounts of pSGF-TAL1 (lanes 2–5; 0.5, 1 and their expression through stimulation of their respective promo- 1.5 mg). At 24 h after transfection, the luciferase activity was measured. ters.16 After considering this, it appeared interesting to examine (b) Endogenous TAL1 competes with hTERT induction by HBZ. Jurkat the effect of TAL1 on hTERT with respect to these viral proteins. cells were transfected with 3 mg of pGL3-hTERT3300 and 2 mgof As observed earlier, Tax was able to reduce hTERT expression, vectors expressing either control or anti-TAL1, as well as 3 mg of HBZ in the same way as the effect exerted by TAL1 (Figure 5a, and JunD expression vectors (lanes 2 and 3). The luciferase activity was measured as in a.(c) TAL1 prevents induction of endogenous compare lanes 2 and 3). Co-expression of both proteins led to a hTERT mRNA by HBZ. HeLa cells were transfected with 2 mgof reinforced negative effect (Figure 5a, compare lanes 2 and 4). vectors expressing TAL1, HBZ and JunD as indicated at the bottom of This effect was not related to an increase in the amount of Tax the graph, which represents the amount of hTERT mRNA quantiﬁed by protein in the presence of TAL1 (Figure 5b, compare lanes 3 and real-time RT-PCR. 4), showing that both factors can cause a marked reduction in hTERT production. We also tested whether this effect could be or both together, the effect of TAL1 was still apparent observed in HTLV-1-infected cells. A ﬁrst analysis was (Figures 3b–e). The negative effect of TAL1 was similar and performed in CD4 þ clones expressing Tax derived from even slightly stronger on the reporter construct bearing infected patients.27 In these cells, a positive correlation was

Leukemia hTERT inhibition by TAL1 J-M Terme et al 2087 TAL1: + + + + observed between Tax and TAL1 expression (Table 1). This is in accordance with our previous work that showed a mutual HBZ: - positive effect between Tax and TAL1 expression.16 In this analysis, a negative correlation was observed between either TAL1 and hTERT or Tax and hTERT, although with limited TAL1 > signiﬁcance levels probably as a consequence of the limited number of clones available for analysis (Table 1). A negative correlation between TAL1 and hTERT expression was also observed in ATL cells and in uninfected T lymphocytes HBZ > reﬂecting the negative function of TAL1 on hTERT expression (Table 1). In agreement with earlier observations, by contrast with Tax or TAL1, co-expression of HBZ and JunD had a clear stimulatory β-actin > effect on the activity of the reporter construct including the hTERT promoter (Figure 6a, compare lanes 1 and 2), but production of increasing amounts of TAL1 antagonized this 1 2 3 4 effect (Figure 6a). Conversely, co-expression of the shRNA targeting TAL1 was able to reinforce the stimulatory activity of TAL1: +++ HBZ/JunD (Figure 6b). This shows that the endogenous TAL1 HBZ: - + + present in leukemic Jurkat cells had a negative effect on HBZ/ MG132: - - + JunD transactivation. In accordance with these observations, it was also observed that HBZ/JunD was able to impair the TAL1 > negative effect of TAL1 on endogenous hTERT mRNA expression (Figure 6c). To further understand how HBZ interferes with TAL1, the effect of the viral protein on the amount of TAL1 was analyzed by HBZ > immunoblot. Unexpectedly, expression of increasing amounts of HBZ was found to cause a profound reduction in the TAL1 β-actin > concentration (Figure 7a, upper panel). As the proteasome inhibitor, MG132, was able to block this effect (Figure 7b, 1 23 compare lanes 2 and 3, upper panel), it appeared that HBZ stimulates TAL1 degradation by the proteasome. This effect is extracts IP α-TAL1 likely related to an interaction between both proteins. Indeed, in thepresenceofMG132,TAL1andHBZwerefoundto TAL1: ++ - + + - coprecipitate (Figure 7c, lane 5). Despite this interaction between HBZ: - ++ - + + TAL1 and HBZ, TAL1 expression did not weaken the association of HBZ with JunD (Supplementary Figure 4). Collectively, these MG132: + + + + + + data show that HBZ by interacting with TAL1 triggers its degradation, thereby impairing its negative effect on the activity TAL1 > of the hTERT promoter. Tax and HBZ therefore appeared to have opposite effects on TAL1 and hTERT expression.

HBZ > 123 4 5 6 Discussion Figure 7 (a) HBZ (HTLV-1 basic leucine zipper) induces a decrease in the T-cell acute lymphoblastic leukemia 1 (TAL1) intracellular TAL1 has an important function in T-ALL by being inappropri- amount. HeLa cells were transfected with 1 mg of pSGF-TAL1 and ately expressed in the T-cell lineage as a consequence of increasing amounts of MYC-HBZ expression vector (lanes 2–4; 1, 2 translocation or deletion placing the gene under the control of and 3 mg). Cell lysates were used for immunoprecipitation with an the SIL promoter.28,29 It is also estimated that TAL1 is often antibody to FLAG. Immunoprecipitated proteins were analyzed by abnormally expressed in T-ALL as a consequence of epigenetic immunoblot using antibodies to FLAG (top), to MYC (middle) and to b- activation.30 As telomerase is often highly active in transformed actin (bottom). The positions of the TAL1, HBZ and b-actin proteins are indicated on the left. (b) HBZ-mediated degradation of TAL1 relies cells, a negative effect of TAL1 on the hTERT expression was on proteasome activity. HeLa cells were transfected with 1 mg of pSGF- unexpected. Telomerase activity is determined by the expres- TAL1 without (lane 1) or with 1 mg of MYC-HBZ expression vector sion level of both its hTERT and hTR components, which are (lanes 2 and 3). In lane 3, cells were treated with the proteasome limiting.31 In agreement with this notion, we observed that inhibitor MG132. Cell lysates were analyzed by immunoblot as overexpressing TAL1 in cells leads to a decrease in the described in a.(c) HBZ interacts with TAL1. HeLa cells were telomerase activity. We also observed that suppression of transfected with 1 mg of pSGF-TAL1 (lanes 1 and 4), MYC-HBZ vector (lanes 3 and 6) or both constructs together (lanes 2 and 5) and treated TAL1 by RNA interference in Jurkat cells raised the amount of with the MG132 proteasome inhibitor. Cell lysates were analyzed by hTERT indicating that the negative effect of TAL1 intervenes at immunoblot using antibodies to FLAG (top) and to MYC (bottom). physiological concentrations. In agreement with these experi- Positions of the signals corresponding to TAL1 and HBZ are indicated ments performed in cell lines, we also observed a negative on the left. correlation between TAL1 and hTERT expression in T lymphocytes expressing either Tax or not, as well as in ATL cells. However, these correlations were obtained with uncertain signiﬁcance levels and have to be conﬁrmed with a larger set

Leukemia hTERT inhibition by TAL1 J-M Terme et al 2088 of patients. Despite this limitation, these data obtained in various reverts the negative effect of TAL1 on hTERT. Indeed, it has been cells by different experimental approaches collectively support shown earlier that Notch was able to induce polyubiquitination the notion that TAL1 is a negative regulator of telomerase. and proteasome-mediated degradation of TAL139 and that Myc This effect of TAL1 on hTERT depends on the proximal can transactivate hTERT.39–41 promoter. As E-boxes are present in this part of the promoter, In conclusion, our data established TAL1 as an important there is a possibility of interference of TAL1 with these motifs, regulator of hTERT. Our results add to the complexity of the which have been shown to be regulated by the Myc–Max regulation of this key protein, which has an essential function in complex. However analysis of mutated constructs clearly cellular transformation. This strengthens the interest of potential showed that the effect of TAL1 did not depend on these manipulation of the activity of telomerase in a therapeutic goal. sequence motifs, but on the Sp1-binding sites. In agreement Alternatively, it could be important to inhibit TAL1 at early with this observation, overexpression of TAL1 was able to stages to avoid genome instability, thus paving the way for full impair stimulation of hTERT promoter activity by Sp1, as cellular transformation. well as by p300. A negative effect of TAL1 on different promoters is well established,17 and it has been shown to allow recruitment of transcriptional repressors such as mSIN3A Acknowledgements and HDAC1.19 The ChIP assays performed in this study clearly show that TAL1 associates with the hTERT promoter, likely We thank Katie Flaig, Saturo Kyo, Louis Gazzolo, Daniele through an interaction with Sp1. A direct explanation to the Mathieu and Jean-Michel Mesnard for generously providing us consequent transcriptional repression was provided by the with expression vectors and antibodies. We also thank Armelle TAL1-induced recruitment of HDAC1 on the promoter. Indeed, Roisin for help with cell culture, Eric Gilson’s team for assistance the presence of p300 was weakly affected. These observations with telomerase activity measurement, Sevanna Shahbazian and show that, similar to other promoters, the repressive effect of Thomas Barber for a critical reading of the paper. This work was TAL1 is because of the induced association of a histone supported by Association pour la Recherche sur le Cancer (grant deacetylase to the promoter. and VM fellowship), by the comite´ du Rhoˆne de la Ligue A consequence of the negative effect of TAL1 on hTERT Nationale contre le Cancer and by the Fondation pour la expression is to favor genome instability, which has been Recherche Me´dicale (J-M T fellowships). observed in many types of leukemia as well as telomere shortening.7,32 This effect probably drives a mutagenic effect in References favor of the development of a fully transformed phenotype. At late stages of transformation, it cannot be excluded that this 1 Gilson E, Geli V. How telomeres are replicated. Nat Rev Mol Cell negative effect of TAL1 is countered by other events stimulating Biol 2007; 8: 825–838. hTERT expression. Our observations of the Tax and HBZ protein 2 Harley CB. Telomerase and cancer therapeutics. Nat Rev Cancer 2008; 8: 167–179. effects on the TAL1 inhibition of hTERT are in favor of such a 3 Cogulu O, Kosova B, Karaca E, Gunduz C, Ozkinay F, Aksoylar S model. Indeed, it has been reported earlier that Tax inhibits et al. Evaluation of telomerase mRNA (hTERT) in childhood acute hTERT transcription and that HBZ in combination with JunD leukemia. Leuk Lymphoma 2004; 45: 2477–2480. activates it.13,14 It is thought that Tax is highly expressed at early 4 Kubuki Y, Suzuki M, Sasaki H, Toyama T, Yamashita K, Maeda K stages of infection before an active immune response blocks et al. Telomerase activity and telomere length as prognostic factors expression of the protein. During later stages of infection, of adult T-cell leukemia. Leuk Lymphoma 2005; 46: 393–399. 5 Weinrich SL, Pruzan R, Ma L, Ouellette M, Tesmer VM, Holt SE several arguments support a very low expression of Tax in ATL et al. Reconstitution of human telomerase with the template RNA 0 cells because of deletion or hypermethylation of the 5 LTR and component hTR and the catalytic protein subunit hTRT. Nat Genet even in some cases to the appearance of stop codons in the Tax 1997; 17: 498–502. coding sequence.33–35 By contrast, the HBZ RNA is well 6 Koyanagi Y, Kobayashi D, Yajima T, Asanuma K, Kimura T, Sato T expressed in ATL cells.36,37 From these observations, it has et al. Telomerase activity is down regulated via decreases in hTERT been proposed that Tax by inhibiting hTERT favors an unstable mRNA but not TEP1 mRNA or hTERC during the differentiation of leukemic cells. Anticancer Res 2000; 20 (2A): 773–778. genomic state at early steps of transformation, whereas at late 7 Ohyashiki JH, Ohyashiki K, Fujimura T, Kawakubo K, Shimamoto stages HBZ would drive with JunD a high expression of hTERT T, Iwabuchi A et al. Telomere shortening associated with disease and consequently a high telomerase activity.14 We show here evolution patterns in myelodysplastic syndromes. Cancer Res that TAL1 can reinforce hTERT repression. As we have shown 1994; 54: 3557–3560. earlier that there was a positive cross talk between the 8 Kyo S, Takakura M, Fujiwara T, Inoue M. Understanding and expression of both factors,16 our data support the notion that exploiting hTERT promoter regulation for diagnosis and treatment 13 of human cancers. Cancer Sci 2008; 99: 1528–1538. TAL1 participates in the negative effect of Tax on hTERT. 9 Hao H, Nancai Y, Lei F, Xiong W, Wen S, Guofu H et al. siRNA However, HBZ can block this inhibition. As the activity of the directed against c-Myc inhibits proliferation and downregulates JunD–HBZ complex depends on the Sp1-binding sites of the human telomerase reverse transcriptase in human colon cancer hTERT promoter, it is possible that some competition exists Colo 320 cells. J Exp Clin Cancer Res 2008; 27: 27. between JunD–HBZ and TAL1, the former impeding the 10 Kyo S, Takakura M, Taira T, Kanaya T, Itoh H, Yutsudo M et al. Sp1 association of the latter with the promoter. Alternatively, cooperates with c-Myc to activate transcription of the human telomerase reverse transcriptase gene (hTERT). Nucleic Acids Res JunD–HBZ by allowing an active recruitment of p300 could 2000; 28: 669–677. block the negative effect of TAL1. However, we also observed 11 Lebel R, McDuff FO, Lavigne P, Grandbois M. Direct visualization that HBZ degrades TAL1. Such an effect of HBZ has already of the binding of c-Myc/Max heterodimeric b-HLH-LZ to been observed with c-Jun.38 So HBZ clearly blocks the negative E-box sequences on the hTERT promoter. Biochemistry 2007; 46: effect of TAL1 on hTERT and this is likely to participate in the 10279–10286. reactivation of the telomerase activity in fully transformed cells. 12 Xu D, Popov N, Hou M, Wang Q, Bjorkholm M, Gruber A et al. Switch from Myc/Max to Mad1/Max binding and decrease in Such an effect might be mediated by other factors in non-viral histone acetylation at the telomerase reverse transcriptase promo- T-ALL because of mutation events. In particular, it is possible ter during differentiation of HL60 cells. Proc Natl Acad Sci USA that activation of the Notch–Myc pathway in leukemic cells 2001; 98: 3826–3831.

Leukemia hTERT inhibition by TAL1 J-M Terme et al 2089 13 Gabet AS, Mortreux F, Charneau P, Riou P, Duc-Dodon M, Wu Y factor that down-regulates viral transcription. J Virol 2002; 76: et al. Inactivation of hTERT transcription by Tax. Oncogene 2003; 12813–12822. 22: 3734–3741. 27 Sibon D, Gabet AS, Zandecki M, Pinatel C, Thete J, Delfau-Larue 14 Kuhlmann AS, Villaudy J, Gazzolo L, Castellazzi M, Mesnard JM, MH et al. HTLV-1 propels untransformed CD4 lymphocytes into Duc Dodon M. HTLV-1 HBZ cooperates with JunD to enhance the cell cycle while protecting CD8 cells from death. J Clin Invest transcription of the human telomerase reverse transcriptase gene 2006; 116: 974–983. (hTERT). Retrovirology 2007; 4: 92. 28 Aplan PD, Lombardi DP, Reaman GH, Sather HN, Hammond GD, 15 O’Neil J, Shank J, Cusson N, Murre C, Kelliher M. TAL1/SCL Kirsch IR. Involvement of the putative hematopoietic transcription induces leukemia by inhibiting the transcriptional activity of E47/ factor SCL in T-cell acute lymphoblastic leukemia. Blood 1992; HEB. Cancer Cell 2004; 5: 587–596. 79: 1327–1333. 16 Terme JM, Wencker M, Favre-Bonvin A, Bex F, Gazzolo L, 29 Macintyre EA, Smit L, Ritz J, Kirsch IR, Strominger JL. Disruption of Duc Dodon M et al. Cross talk between expression of the the SCL locus in T-lymphoid malignancies correlates with human T-cell leukemia virus type 1 Tax transactivator and the commitment to the T-cell receptor alpha beta lineage. Blood oncogenic bHLH transcription factor TAL1. J Virol 2008; 82: 1992; 80: 1511–1520. 7913–7922. 30 Ferrando AA, Neuberg DS, Staunton J, Loh ML, Huard C, 17 Palomero T, Odom DT, O’Neil J, Ferrando AA, Margolin A, Raimondi SC et al. Gene expression signatures define novel Neuberg DS et al. Transcriptional regulatory networks downstream oncogenic pathways in T cell acute lymphoblastic leukemia. of TAL1/SCL in T-cell acute lymphoblastic leukemia. Blood 2006; Cancer Cell 2002; 1: 75–87. 108: 986–992. 31 Cairney CJ, Keith WN. Telomerase redefined: integrated regulation 18 Lecuyer E, Hoang T. SCL: from the origin of hematopoiesis to stem of hTR and hTERT for telomere maintenance and telomerase cells and leukemia. Exp Hematol 2004; 32: 11–24. activity. Biochimie 2008; 90: 13–23. 19 Huang S, Brandt SJ. mSin3A regulates murine erythroleukemia cell 32 Kleideiter E, Bangerter U, Schwab M, Boukamp P, Koscielniak E, Klotz differentiation through association with the TAL1 (or SCL) U et al. Telomeres and telomerase in paediatric patients with T-cell transcription factor. Mol Cell Biol 2000; 20: 2248–2259. acute lymphoblastic leukaemia (T-ALL). Leukemia 2005; 19: 296–298. 20 Huang S, Qiu Y, Stein RW, Brandt SJ. p300 functions as a 33 Furukawa Y, Kubota R, Tara M, Izumo S, Osame M. Existence of transcriptional coactivator for the TAL1/SCL oncoprotein. escape mutant in HTLV-I tax during the development of adult Oncogene 1999; 18: 4958–4967. T-cell leukemia. Blood 2001; 97: 987–993. 21 Lecuyer E, Herblot S, Saint-Denis M, Martin R, Begley CG, Porcher 34 Miyazaki M, Yasunaga J, Taniguchi Y, Tamiya S, Nakahata T, C et al. The SCL complex regulates c-kit expression in hemato- Matsuoka M. Preferential selection of human T-cell leukemia virus poietic cells through functional interaction with Sp1. Blood 2002; type 1 provirus lacking the 50 long terminal repeat during 100: 2430–2440. oncogenesis. J Virol 2007; 81: 5714–5723. 22 Osada H, Grutz G, Axelson H, Forster A, Rabbitts TH. Association 35 Takeda S, Maeda M, Morikawa S, Taniguchi Y, Yasunaga J, of erythroid transcription factors: complexes involving the LIM Nosaka K et al. Genetic and epigenetic inactivation of tax gene in protein RBTN2 and the zinc-finger protein GATA1. Proc Natl Acad adult T-cell leukemia cells. Int J Cancer 2004; 109: 559–567. Sci USA 1995; 92: 9585–9589. 36 Satou Y, Yasunaga J, Yoshida M, Matsuoka M. HTLV-I basic 23 Valge-Archer VE, Osada H, Warren AJ, Forster A, Li J, Baer R et al. leucine zipper factor gene mRNA supports proliferation of adult T The LIM protein RBTN2 and the basic helix-loop-helix protein cell leukemia cells. Proc Natl Acad Sci USA 2006; 103: 720–725. TAL1 are present in a complex in erythroid cells. Proc Natl Acad 37 Usui T, Yanagihara K, Tsukasaki K, Murata K, Hasegawa H, Yamada Y Sci USA 1994; 91: 8617–8621. et al. Characteristic expression of HTLV-1 basic zipper factor (HBZ) 24 Yao J, Grant C, Harhaj E, Nonnemacher M, Alefantis T, Martin J transcripts in HTLV-1 provirus-positive cells. Retrovirology 2008; 5:34. et al. Regulation of human T-cell leukemia virus type 1 gene 38 Matsumoto J, Ohshima T, Isono O, Shimotohno K. HTLV-1 HBZ expression by Sp1 and Sp3 interaction with TRE-1 repeat III. DNA suppresses AP-1 activity by impairing both the DNA-binding ability Cell Biol 2006; 25: 262–276. and the stability of c-Jun protein. Oncogene 2005; 24: 1001–1010. 25 Riou P, Bex F, Gazzolo L. The human T cell leukemia/ 39 Nie L, Wu H, Sun XH. Ubiquitination and degradation of Tal1/SCL lymphotropic virus type 1 Tax protein represses MyoD-dependent are induced by notch signaling and depend on Skp2 and CHIP. transcription by inhibiting MyoD-binding to the KIX domain of J Biol Chem 2008; 283: 684–692. p300. A potential mechanism for Tax-mediated repression of the 40 Sharma VM, Draheim KM, Kelliher MA. The Notch1/c-Myc transcriptional activity of basic helix-loop-helix factors. J Biol pathway in T cell leukemia. Cell Cycle 2007; 6: 927–930. Chem 2000; 275: 10551–10560. 41 Weng AP, Millholland JM, Yashiro-Ohtani Y, Arcangeli ML, Lau A, 26 Gaudray G, Gachon F, Basbous J, Biard-Piechaczyk M, Devaux C, Wai C et al. c-Myc is an important direct target of Notch1 in T-cell Mesnard JM. The complementary strand of the human T-cell acute lymphoblastic leukemia/lymphoma. Genes Dev 2006; 20: leukemia virus type 1 RNA genome encodes a bZIP transcription 2096–2109.

Supplementary Information accompanies the paper on the Leukemia website (http://www.nature.com/leu)

Leukemia