Dr. Shigeru Morikawa deceased after the submission.
Genetic and epigenetic inactivation of tax gene in adult T-cell leukemia cells
Article first published online: 9 JAN 2004
Copyright © 2004 Wiley-Liss, Inc.
International Journal of Cancer
Volume 109, Issue 4, pages 559–567, 20 April 2004
How to Cite
Takeda, S., Maeda, M., Morikawa, S., Taniguchi, Y., Yasunaga, J.-i., Nosaka, K., Tanaka, Y. and Matsuoka, M. (2004), Genetic and epigenetic inactivation of tax gene in adult T-cell leukemia cells. Int. J. Cancer, 109: 559–567. doi: 10.1002/ijc.20007
- Issue published online: 24 FEB 2004
- Article first published online: 9 JAN 2004
- Manuscript Accepted: 24 OCT 2003
- Manuscript Revised: 17 OCT 2003
- Manuscript Received: 24 APR 2003
- Ministry of Education, Science, Sports and Culture of Japan
- Welfide Medicinal Research Foundation
- Public Trust Haraguchi Memorial Cancer Research Fund
- DNA methylation;
To clarify the status of tax gene, we analyzed human T-cell leukemia virus type-I (HTLV-I) associated cell lines and fresh adult T-cell leukemia (ATL) cells. We compared 2 types of HTLV-I associated cell lines: one was derived from leukemic cells (leukemic cell line) and the other from nonleukemic cells (nonleukemic cell line). Although all nonleukemic cell lines expressed Tax, it could not be detected in 3 of 5 leukemic cell lines, in which nonsense mutation or deletion (60 bp) of tax genes, and DNA methylation in 5′-LTR were identified as the responsible changes. We found such genetic changes of the tax gene in 5 of 47 fresh ATL cases (11%). The tax gene transcripts could be detected in 14 of 41 fresh ATL cases (34%) by RT-PCR. In ATL cases with genetic changes that could not produce Tax protein, the tax gene was frequently transcribed, suggesting that such cells do not need the transcriptional silencing. Although DNA methylation of 5′-LTR was detected in the fresh ATL cases (19 of 28 cases; 68%), the complete methylation associated with transcriptional silencing was observed only in 4 cases. Since partial methylation could not silence the transcription, and the tax gene transcription was not detected in 27 of 41 cases (66%), the epigenetic change(s) other than DNA methylation is considered to play an important role in the silencing. © 2004 Wiley-Liss, Inc.
Adult T-cell leukemia (ATL) is a neoplastic disease derived from helper T-lymphocytes and is considered as a distinct clinical entity based on clinical features and geographic distribution of patients.1, 2 Identification of the causative agent, HTLV-I, has allowed detailed analysis of the epidemiological, immunological and clinical characteristics of ATL.3, 4, 5, 6 There are 4 clinical subtypes of ATL: smoldering, chronic, acute and lymphoma-type ATL. The first 2 types exhibit insidious clinical course and after several years progress to acute or lymphoma-type ATL. Acute and lymphoma-type ATL are clinically aggressive forms with a mean survival time of about 1 year in spite of intensive chemotherapy. HTLV-I transmits mainly from mother to child via breast milk. The relative risk of development of ATL among carriers has been estimated in Japan to be about 5% after a long latent period of approximately 60 years.7 It is suggested that the process of leukemogenesis depends on multiple steps and is influenced by various factors such as viral protein.8
Tax is an important HTLV-I viral protein and is encoded by the pX region between env and 3′-long terminal repeat (LTR).5 Tax is thought to play a central role in ATL leukemogenesis by its pleiotropic actions such as transactivation of NFκB, CREB and SRF pathways,9, 10 and functional inactivation of p16, p53 and MAD1.11, 12, 13 However, the enigma of Tax-induced leukemogenesis is that the expression of Tax in leukemic cells remains unclear. In some ATL cells, deletion of 5′-LTR,15 which is a promoter of viral genes, and nonsense mutation of tax gene result in the loss of Tax protein.16 It is noteworthy that these changes are predominantly observed in the aggressive forms of ATL like acute and lymphoma-type ATL.
In the present study, we analyzed the tax gene in HTLV-I associated cell lines and fresh ATL cells, and identified the mechanisms that inactivated tax gene in Tax nonexpressing cell lines.
MATERIAL AND METHODS
Eight HTLV-I-associated cell lines were used in the present study: 5 (ED, ATL-43T, ATL-48T, ATL-55T and ATL-2) were derived from a leukemic clone identified in vivo by the same integration sites of HTLV-I provirus or recombination of T-cell receptor genes.17, 18, 19, 20 The remaining 3 cell lines (MT-2, Sez627 and ATL-35T) were not derived from leukemic cells. The human embryonic kidney cell line, 293, was studied as a control. To study the effect of demethylation, ATL-43T was cultured in media supplemented with 10 μM 5-aza-2′-deoxycytidine (5-Aza-CdR) (Sigma Chemical Co., St. Louis, MO) for 3 days or 10 μM 5-Aza-CdR and 1 μM trichostatin A (TSA) (Sigma Chemical Co.) for the last 24 hr, and then followed by isolation of RNAs and proteins.
Peripheral blood mononuclear cells (PBMCs) were isolated from patients with ATL. Clinical subtypes of ATL were diagnosed as reported by Shimoyama et al.21 The percentage of CD4-positive cells among lymphocyte population was > 90% in each case examined. The subjects were 47 patients with ATL (30 cases with acute ATL, 4 with lymphoma-type ATL and 13 with chronic ATL). The diagnosis of ATL was confirmed by the monoclonal integration of HTLV-I provirus in the host genome by the Southern blot method.
cDNA synthesis and direct sequencing
Total RNAs were isolated from the cell lines and fresh ATL cells using Trizol reagent (Life Technologies, Inc., Paisley, UK) and cDNAs were prepared from 1 μg of total RNAs using the RNA LA PCR Kit (Takara, Shiga, Japan) as described by the manufacturer. Oligo dT primers were used to prime first-strand synthesis for the entire reaction. For PCR, 1 μl of the reverse transcriptase reaction mixture was diluted with 50 μl of PCR buffer containing 0.2 mM each of deoxynucleotide triphosphates, 1.5 mM MgCl2, 1.25 unit of Taq DNA polymerase (Takara) and 20 pmol of each primer. Primers used for RT-PCR were as follows: tax gene; 5′-CCGGCGCTGCTCTCATCCCGGT-3′ (sense) and 5′-GGCCGAACATAGTCCCCCAGAG-3′ (antisense). PCR was performed in the GeneAmp 2400 (Applied BioSystems, Foster City, CA) for 40 cycles under the following conditions: 2 min at 94°C and 40 cycles of 30 sec at 94°C, 30 sec at 61°C, 2 min at 72°C and finally 2 min at 72°C. Sequencing was performed using Big Dye Terminator (Applied BioSystems) with an ABI 377 autosequencer (Applied BioSystems).
Total proteins were isolated using RIPA buffer [1% Nonidet P-40, 0.5% sodium deoxycholate, and 0.1% sodium dodecyl sulfate (SDS)]. The 20 μg of protein with protease inhibitors (0.1 mg/ml PMSF, 3% Aprotinin, 1 mM sodium orthovanadate) was separated on SDS-10% polyacrylamide gel electrophoresis and transferred onto nitrocellulose membranes. The membranes were blocked overnight at 4°C with 5% bovine serum albumin, 5% skim milk and 0.01% NaN3 in PBS containing 0.1% Tween 20. After washing, anti-Tax (Lt-4)22 or anti-α-tubulin antibody was incubated with the membrane for 1 hr at room temperature, and a horseradish peroxidase conjugate anti-mouse IgG was used during the final 40 min of incubation. Chemiluminescent detection of blotted proteins was performed with an enhanced chemiluminescence kit (Amersham Biosciences Corp., Piscataway, NJ).
Long PCR was performed by a hot start PCR amplification with AmpliWax (Applied BioSystems) as described previously.15 The lower mixture contained 1×LA PCR buffer II (Mg++-free; Takara), 1.5 mM MgCl2, 0.3 μM of each primer and 0.2 mM deoxynucleotide triphosphates. The upper mixture contained 1×LA PCR buffer II (Mg++-free), 0.5 units of ExTaq (Takara) and 0.2 μg of genomic DNA. PCR cycle conditions were as follows: 25 cycles of 15 sec at 94°C, 30 sec at 68°C and 10 min at 72°C. Sequences of primers used in these experiments were as follows: primer 1; 5′-GTTCCACCCCTTTCCCTTTCATTCACGACTGACTGC-3′; primer 2; 5′-GGCTCTAAGCCCCCGGGGGATATTTGGGGCTCATGG-3′; primer 3; 5′-GGGGTCCCAGGTGATCTGATGCTCTGGACAGGTGGC-3′; primer 4; 5′-GGCGACTGGTGCCCCATCTCTGGGGGACTATGTTCG-3′. To amplify the entire provirus, primers 1 and 2 were used. To determine the deletion of 5′ LTR (defective type 2), the sets of primers 1 and 3 or primers 2 and 4 were used for PCR.
Determination of genomic sequence adjacent to 5′-LTR
To amplify the genomic DNA adjacent to 5′-LTR, we used inverse long PCR as described previously.23 In brief, the genomic DNAs (1 μg) were digested with Pst I and then ligated by T4-DNA ligase. The circularized DNA was used as a substrate for nested PCR. We performed the first PCR (40 cycles) with LA Taq (Takara) as follows: denaturation at 94°C for 30 sec, annealing at 61°C for 30 sec and extension at 72°C for 5 min, followed by a final 10 min extension at 72°C, and the second PCR as follows: denaturation at 94°C for 30 sec, annealing at 57°C for 30 sec and extension at 72°C for 5 min, followed by a final 10 min extension at 72°C. Primers used in this experiment were as follows: the first PCR, 5′-AAGCAAGAAGTCTCCCAAGC-3′ (gag: 1261—1280) (sense) and 5′-AGTTAAGCCAGTGATGAGCG-3′ (gag: 861–880) (antisense); the second PCR, 5′-CCAGTTTATGCAGACCATCC-3′ (gag: 1296–1315) (sense) and 5′-TTCAGACTTCTGTTTCTCGG-3′ (U3: 64–83) (antisense). The numbering of nucleic acids was in reference to ATK-1 according to Seiki et al.5 The sequence of PCR product was determined as described above.
DNA methylation of U3 region determined by sequencing of sodium bisulfite-treated genomic DNAs
DNA methylation was detected by sequencing of PCR products, which were amplified with sodium bisulfite-treated genomic DNAs as described previously.20 In brief, 1.0 μg DNA samples were denatured by the addition of the same volume of 0.6 M NaOH and then incubated for 15 min at 37°C. Then, 208 μl of 3.6 M sodium bisulfite (pH 5.0) and 12 μl of 10 μM hydroquinone were added to the samples. The samples were covered by mineral oil and then incubated at 55°C for 16 hr. The treated DNAs were purified using Wizard DNA Clean-Up System (Promega, Madison, WI) and then incubated for 10 min in 0.3 N NaOH following the ethanol precipitation. The sodium bisulfite-treated DNAs were amplified by nested PCR with specific primers to the U3 region of HTLV-I LTR, and the genomic sequence of integration site as described above (5′-LTR) or pX region (3′-LTR) as follows: the first PCR, 5′-GGTGTGAGATGGTATTTTATTGTGG-3′ (the genomic sequence adjacent to 5′-LTR of ATL-43T) (sense) or 5′-(C/T)GATGGTA(C/T)GTTTATGATTTT(C/T)GGG-3′(pX) (sense) and 5′-AACTCCTATTATTTTATTAAACC(A/G)TATAC(A/G)-3′(U3) (antisense); the second PCR 5′-GGTGTGAGATGGTATTTTATTGTGG-3′ (the genomic sequence adjacent to 5′-LTR of ATL-43T) (sense) or 5′-(C/T)GATGGTA(C/T)GTTTATGATTTT(C/T)GGG-3′(pX) (sense) and 5′-CC(A/G)TATAC(A/G)TACCATAAAAA-3′ (U3) (antisense). The conditions for PCR were as follows; the first PCR was carried out for 35 cycles (30 sec at 94°C, 30 sec at 47°C (for 5′-LTR) or 49°C (for 3′-LTR), and 30 sec at 72°C) and followed by a final 2 min extension at 72°C, and the second PCR was performed for 40 cycles (30 sec at 94°C, 30 sec at 54°C (for 5′- and 3′-LTR), and 30 sec at 72°C) and followed by a final 2 min extension at 72°C. The final PCR products were subcloned into TOPO XL PCR Cloning kit (Invitrogen, Carlsbad, CA), and the sequences of each clone were determined. DNA methylation of 5′-LTR in the fresh ATL cells was determined using the specific primers to the integration sites of each case.
Methylation specific PCR (MS-PCR)
MS-PCR was performed as described previously by Herman et al.24 Sodium bisulfite treated DNA was first amplified with PCR primers in U3 region of 5′-LTR and gag region, and then amplified by primers in R region and gag region. Primers were made to anneal to methylated and unmethylated sequences after treatment by sodium bisulfite. Primers were as follows: the first PCR, 5′-TTAAGTCGTTTTTAGGCGTTGAC-3′ (U3; methylated) (sense) or 5′-TTAAGTTGTTTTTAGGTGTTGAT -3′(U3; unmethylated) (sense) and 5′-AAAAAAATTTAACCCATTACC-3′ (gag) (antisense); the second PCR 5′- GAGGTCGTTATTTACGTCGGTTGAGTC-3′ (R; methylated) (sense) or 5′- GAGGTTGTTATTTATGTTGGTTGAGTT-3′ (R; unmethylated) (sense) and 5′-AAAAAAATTTAACCCATTACC - 3′ (gag) (antisense). The conditions for PCR were as follows; the first PCR was carried out for 35 cycles [30 sec at 95°C, 30 sec at 53°C (for methylated and unmethylated), and 40 sec at 72°C] and followed by a final 2 min extension at 72°C, and the second PCR was performed for 35 cycles [30 sec at 95°C, 30 sec at 52°C (for methylated) or 57°C (for unmethylated) and 30 sec at 72°C] and followed by a final 2 min extension at 72°C.
Quantification of tax gene transcript
Transcriptions of tax and GAPDH (internal control) genes were quantified with real-time PCR as described previously.25 Probes and primers for tax and GAPDH genes were designed for quantification. The sequences of primers and probe for tax and GAPDH genes were as follows; tax primers; 5′-CCGCCGATCCCAAAGAA-3′ (sense) and 5′-CTCTGTCCAAACCCTGGGAA-3′ (antisense); tax probe; 5′-AAGACCACCAACACCA TGGCCCA-3′; GAPDH primers; 5′-ACCAACTGCTTAGCACCCCT-3′ (sense); 5′-GTCTTCTGGGTGGCAGTGAT-3′ (antisense); and GAPDH probe; 5′-CTTTGGTATCGTGGAAGGACTCATGACC-3′. The probes were labeled with fluorescent 6-carboxyfluorescein (FAM) (reporter) at the 5′ end and fluorescent 6-carboxy tetramethyl rhodamine (TAMRA) (quencher) at the 3′ end. A part (1/40) of RT-PCR products synthesized from 5 μg total RNA were used for real-time PCR in a 50 μl amplification reaction solution containing 1 × TaqMan buffer A, 3.5 μM MgCl2, 200 μM each of dATP, dCTP, dGTP, 400 μM dUTP, 1.25 unit of AmpliTaq Gold polymerase, 0.5 unit of AmpErase UNG, 300 nM of each primers and 200 nM of the probe. The reaction conditions were 95°C for 10 min (activation of the AmpliTaq Gold polymerase), 40 cycles of 15 sec at 95°C (denaturing) followed by 60 sec at 56°C (for tax) or 60°C (for GAPDH) (annealing and extension). All experiments were performed and analyzed by the ABI PRISM 7700 Sequence Detection System (Applied BioSystems). To compare the gene expression, we measured the expression of tax gene relative to that of GAPDH gene.
Chromatin immunoprecipitation assay
Chromatin immunoprecipitation (ChIP) assays were performed as previously described.26 In brief, ATL cell lines (ATL-43T and ATL-55T) and the fresh ATL cells from an acute ATL patient (5×105 cells/antibody) were fixed with formaldehyde and then sonicated to obtain soluble chromatin. The chromatin solutions were immunoprecipitated with 4 μl of anti-acetylated histone H3 antibody, anti-acetylated histone H4 (Upstate Biotechnology, Lake Placid, NY), or normal rabbit IgG overnight at 4°C, and then the immunoprecipitates were collected with 50% protein A and G-Sepharose slurry preabsorbed with 0.1 mg/ml sonicated salmon sperm DNA. Consequent purified DNAs were subjected to PCR reactions using primer sets specific for 5′-LTR. Primers used are as follows: ATL-43T; 5′-GCTACTCAGAGATAACCACTGC-3′ (sense), ATL-55T; 5′-GTCACCACGTACAGCAAGAA-3′ (sense), case 1; 5′-GGGAGCTGAACTGCATTATC-3′ (sense), and U3 region of LTR; 5′-TAAACTTACCTAGACGGCGG-3′ (antisense primer for all PCR reactions). The condition for PCR was as follows; 2 min at 94°C and 35 cycles of 30 sec at 94°C, 30 sec at 62.5°C (ATL-43T and 55T) or 61°C (ATL sample), 45 sec at 72°C, and finally 2 min at 72°C. PCR products were separated on agarose gel and the results were quantified using ATTO densitometry software. Values were calculated as signal intensity of samples normalized by input DNA.
Tax in HTLV-I associated cell lines derived from leukemic and nonleukemic cells
HTLV-I can transform T-lymphocytes in vitro similar to the transformation of B-lymphocytes by Epstein-Barr Virus (EBV).18 When PBMCs from patients with ATL are cultured in the presence of IL-2, most of established cell lines are derived from nonleukemic cells [nonleukemic (NL) cell line].19 However, a few cell lines were derived from leukemic cells [leukemic (L) cell line]. To clarify the role of Tax in leukemogenesis, we first characterized the expression of Tax in HTLV-I associated cell lines. Tax was expressed in all NL cell lines and 2 L cell lines (ATL-2 and ATL-48T), whereas 3 of the L cell lines (ED, ATL-43T and ATL-55T) did not produce Tax protein (Fig. 1a). To clarify the underlying mechanism of the lack of Tax expression in these cell lines, we analyzed the sequences of cDNAs and genomic DNAs of HTLV-I proviruses. HTLV-I proviruses identified in ED contained nonsense mutation (W56*) as shown in Figure 1b. On the other hand, RT-PCR detected somewhat larger tax gene transcript from ATL-55T than wild-type tax transcript as shown in Figure 1a. Sequencing of this product revealed that a 60 bp deletion containing the third exon of tax gene resulted in aberrant splicing and premature termination in ATL-55T (Fig. 1c). However, ATL-43T contained a complete provirus, and the coding sequences for tax gene were intact (data not shown), suggesting that mechanism(s) other than mutation or deletion suppressed the transcription of tax gene.
DNA methylation in 5′-LTR of ATL-43T
The intact HTLV-I provirus sequence encoding tax gene in ATL-43T suggested the role of transcriptional silencing for the lack of Tax protein in this cell line. Therefore, we analyzed DNA methylation of 21 bp repeats [Tax-responsive element (TRE)-1] and TRE-2 within U3 region of LTR.27, 28, 29 To distinguish 5′- and 3′-LTR, we studied each U3 region, which has been shown to be critical for viral transcription, of 5′- and 3′-LTR using sodium bisulfite-treated DNA from ATL-43T. The U3 region of 5′-LTR was amplified with the primer derived from genomic sequence of integration site, which was determined by inverse PCR, and that in U3 region. Genomic sequence adjacent to 5′-LTR of ATL-43T was identified in 7p15.3 (Genbank: AC006377). The U3 region of 3′-LTR was amplified by the primer in pX region and that in the U3 region as described in Material and Methods. Then, the PCR product was subcloned into plasmid DNA, and the sequences of 10 clones from each PCR products were determined (Fig. 2). Only CpG sites in the U3 region of 5′-LTR was heavily methylated whereas there were no methylated CpG sites in the U3 region of 3′-LTR. Indeed, CpG sites were present in CRE site of TRE-1 sites, which has been shown to interact with binding proteins in vivo,30 and most of CpG sites in TRE-1 sites were methylated in ATL-43T. It was noteworthy that the promoter-proximal TRE-1 that was critical for Tax-induced viral transcription was completely methylated in ATL-43T.30 Since 5′-LTR was the promoter for viral transcription, this result suggested that viral transcription was silenced by the selective DNA methylation of 5′-LTR.
Reactivated expression of Tax protein in ATL-43T
To clarify the role of DNA methylation in the silencing of viral transcription, ATL-43T was treated with a demethylating agent, 5-aza-2′-deoxycytidine (5-Aza-CdR), or 5-Aza-CdR and a histone deacetylase inhibitor, trichostatin A (TSA), which has a synergistic effect with 5-Aza-CdR.31 To study the transcription of tax gene, the transcripts were quantified using real-time PCR as shown in Figure 3a. The transcription of tax gene was reactivated by 5-Aza-CdR and augmented by addition of TSA in ATL-43T. Furthermore, Tax protein was also produced in ATL-43T cells after treatment with 5-Aza-CdR and increased by addition of TSA (Fig. 3b). Treatment only with TSA could not reactivate the tax gene transcription (data not shown). These data showed that for reactivation of the tax gene transcription, the removal of methylated CpG was indispensable, and deacetylation of histones showed augmented effect.32, 33
Status of tax gene in primary ATL cells
To clarify the significance of tax gene in fresh ATL cells, we analyzed the types of provirus, sequences and expression of tax gene in ATL cells from 47 ATL cases. Among 47 cases, genetic changes in the tax gene, which included nonsense mutation, deletion and insertion, were identified in 5 cases (11%) (Table I). The transcripts of tax gene were analyzed by RT-PCR (Fig. 4). In MT-1 cell line, Western blot analysis could not detect Tax protein (data not shown). However, RT-PCR could detect the tax gene transcripts. On the other hand, no transcripts were observed in ATL-43T, and HTLV-I carriers by RT-PCR in our study. Since HTLV-I infected, nonleukemic cells might exist in the mononuclear cells from the patients, it is quite difficult to distinguish whether the tax gene transcripts were derived from leukemic cells or not. However, the finding that RT-PCR could not detect the tax gene transcripts in HTLV-I carrier suggested that it is less likely that detected tax gene transcripts were derived from HTLV-I infected, nonleukemic cells. Among ATL cases in which intact mRNAs were available, we found the tax gene transcripts in 14 out of 41 cases (34%). Interestingly, the tax gene transcripts could be detected in 3 of 4 cases with abortive genetic changes in the tax gene (cases 15, 20, 39 and 43), showing that the tax gene was transcribed when Tax protein could not be produced. This suggests that such ATL cells, which could not produce intact Tax protein, do not need to silence the viral transcription.
|Case number||ATL||Types of provirus||Methylation of 5′LTR||tax||tax expression|
|16||A||C (multiple)||ND||G7464G/A (W56W/*)||−|
|20||A||DT1||U||253 bp deletion||++|
|21||A||DT1||P||1 bp insertion||ND|
|40||Ch||C (multiple)||ND||G7464G/A (W56W/*)||+|
|43||Ch||DT1||U||G7464A (W56*), 8bp deletion||−|
We previously reported that 5′-LTR was preferentially deleted in ATL, which was designated as type 2 defective provirus. Since 5′-LTR is the promoter of viral genes, such provirus is thought to impair the transcription of viral gene. This type of provirus was observed in 14 out of 47 cases (30%). Among 14 ATL cases with type 2 defective provirus, 5 lacked the second exon of tax that existed in env region, and internal promoter in pol region. Among ATL cases with type 2 defective provirus, the tax gene transcription was detected in 3 cases, indicating that lack of 5′-LTR is not sufficient for silencing of viral transcription, and in such cases, the internal promoter or trapped cellular promoter drives the transcription. In ATL cells without 5′-LTR and internal promoter, the tax gene transcript was not detected.
DNA methylation of 5′-LTR in primary ATL cells
To study the third mechanism of inactivating Tax production, DNA methylation, we first determined methylated CpGs of 5′-LTR in 4 cases by sodium bisulfite sequencing (cases 1, 2, 3 and 35). As shown in Figure 5, 5′-LTRs were heavily methylated in 2 cases (cases 3 and 35), whereas little methylation was found in the remaining 2 cases (cases 1 and 2). For further studies, we used methylation-specific PCR (MS-PCR) to detect DNA methylation in R region of 5′-LTR (Table I). DNA methylation in R region correlated with that in U3 region of 5′-LTR (data not shown). When 5′-LTR was completely methylated, only methylation band could be detected as shown in Figure 6 (lane 1), whereas only unmethylation band could be amplified if 5′-LTR was not methylated (lane 3). When 5′-LTR was partially methylated, both methylated and unmethylated bands were observed in ATL-55T (lane 2). In ATL-55T, the tax gene was transcribed, indicating that partial methylation of 5′-LTR could not silence the viral transcription. We analyzed DNA methylation of 5′-LTR in ATL cells except for them with type 2 defective provirus since they did not retain 5′-LTR. Among the ATL patients (28 cases), only methylation band could be detected in 5 cases (18%), and 5′-LTR was partially methylated in 14 cases (50%). However, it was not methylated in 9 cases (32%). Among 5 ATL cases in which only methylated bands were observed by MS-PCR, the tax gene transcript was observed only in a case (case 35). In this case, sodium bisulfite sequencing experiment revealed that the promoter-proximal TRE-1 was partially methylated, whereas CpG sites in the other 2 TRE-1s were completely methylated (Fig. 5). Since the promoter-proximal TRE-1 is critical for the viral transcription, partial methylation of TRE-1 is thought to allow the tax gene transcription. These data suggested that DNA methylation in 5′-LTR was implicated in silencing of tax gene expression; however, the complete DNA methylation of 5′-LTR associated with transcriptional suppression was not so frequent.
Acetylation of histones H3 and H4 tails in the 5′-LTR of HTLV-I
The formation of the Tax/CREB complex on the HTLV-I promoter is critical for the recruitment of the co-activators CBP and p300. CBP/p300 have been shown to directly acetylate lysine residues within the amino-terminal tails of all 4 core histones.34 Since the enhanced histone acetylation has been shown to correlate with activated transcription, we studied the acetylation of histones H3 and H4 in 5′-LTR of ATL cells with or without DNA methylation in this region by ChIP assay. Both histones were acetylated in ATL-55T cell (with partially methylated 5′-LTR) but not in ATL-43T cell (5′-LTR was hypermethylated as shown in Figs. 2 and 7). Since the transcription of tax gene was detected in ATL-55T, but not in ATL-43T, the levels of acetylation were coincident with the transcription of viral genes, which was consistent to the previous report.35 We also studied the histone acetylation in 5′-LTR of the fresh ATL cell (case 1) without DNA methylation of 5′-LTR and detected hyper-acetylation of H3 and H4 histones, suggesting that the transcription of tax gene was active in ATL cells without DNA methylation of 5′-LTR in vivo. Indeed, the transcript of tax gene was detected in this case by RT-PCR (Table I). We could not study the acetylation of histones in ATL cells without DNA methylation and the tax gene transcription since such ATL cells were not available.
In certain virus-induced malignancies, cell lines established in vitro are quite different from the real neoplastic cells in vivo. For example, EBV can transform mature B-lymphocytes in vitro. In such cell lines, several viral genes are expressed, such as LMP, EBNA and EBER, which play important roles in such immortalization. However, viral genes are selectively expressed in Burkitt's lymphoma cells,36 indicating that in vivo malignant cells are quite different from transformed cells in vitro. A similar phenomenon has been observed in HTLV-I associated cell lines. HTLV-I can transform CD4-positive T-lymphocytes in vitro in the presence of IL-2.18 Such cell lines expressed Tax that was essential for immortalization as shown previously.37 However, in the present study, we found Tax nonproducing cell lines, which were derived from leukemic cells, showing that Tax was not always essential for leukemogenesis. Furthermore, we identified the 3 different mechanisms in the inactivation of Tax production: nonsense mutation, deletion of tax gene and DNA methylation in of 5′-LTR.
Based on findings observed in ATL cell lines, we analyzed the tax gene and its expression in the fresh ATL cells. Expression of the tax gene in fresh ATL cells remains obscure in spite of extensive immunohistochemical and molecular studies.14, 38 In our study, we detected the tax gene transcripts in 14 of 41 cases (34%). Genetic changes, such as nonsense mutation, insertion and deletion of the tax gene, were also identified in the fresh ATL cells, which is consistent to the finding that ATL cell lines maintain leukemic state without Tax protein. Among ATL cases without the tax gene expression, we also showed selective DNA methylation of 5′-LTR as a mechanism involved in the suppression of the tax gene transcription. Recently, Koiwa et al.39 reported that DNA methylation in 5′-LTR suppressed the expression of tax gene, suggesting its role in viral latency. It suggested the high frequency of DNA methylation of 5′-LTR. However, our study showed that complete methylation of 5′-LTR associated with loss of the tax gene transcription was not so frequent (4/28; 14%), and partial methylation of 5′-LTR was predominant in ATL cells (14/28; 50%). Since partial methylation of 5′-LTR did not silenced the tax gene transcription in both cell lines and fresh ATL cells, the transcriptional silencing by DNA methylation is not so common phenomenon. On the other hand, the tax gene transcript was detected in most fresh ATL cells with the genetically changed tax gene, which could not produce Tax protein. This finding also indicates that silencing of tax gene transcription is not necessary when Tax protein could not be produced, supporting the hypothesis that transcriptional silencing observed in ATL cells enables them to escape from the host immune system against Tax protein. Since the primary ATL cells did not express the tax gene transcripts (66%; 27 of 41 cases) even when 5′-LTR was unmethylated at all, it indicated that other mechanism than DNA methylation silenced the transcription of viral gene. Recent study showed that methylation of histone H3 lysine-9 is associated with silencing in the absence of DNA methylation, indicating that histone modification plays a critical role in the silencing of viral genes in HTLV-I infected cells.40, 41 The mechanism of such silencing should be clarified in the future study.
Such loss or suppression of Tax expression provided a significant clue to the leukemogenesis of ATL. The pleiotropic actions of Tax protein promoted proliferation and inhibited apoptosis of HTLV-I infected cells; however, at the same time, this protein was the major target of cytotoxic T-cells in vivo.42, 43, 44 Thus, the presence of Tax in HTLV-I-infected cells provided advantages and disadvantages to the survival of HTLV-I-infected cells. It is speculated that Tax plays an important role in persistent proliferation of HTLV-I-infected cells during the carrier state, and then genetic and epigenetic changes accumulate in the host genome by mutator phenotype of Tax,45 which finally lead to downregulation of Tax function and escape from the host immune system. Indeed, DNA methylation of 5′-LTR is less frequent in HTLV-I carriers compared to that of ATL cells (our unpublished data). DNA methylation and epigenetic changes of viral promoters down-regulate Tax expression and function, leading to escape from the host immune system.
We previously identified the type 2 defective provirus that lacked 5′-LTR and internal sequences such as gag and pol, in ATL cells.15 The U3 region of 5′-LTR contains the critical motifs for transcription of viral genes and DNA methylation associated with histone deacetylation of this region are thought to disturb the interaction with various transcriptional factors including CREB, ATF-1, ATF-2 and other members of CREB-ATF superfamily transcriptional factors,30, 32, 35 resulting in silencing of viral transcription. In our study, deletion of 5′-LTR was identified in 30% of fresh ATL cases. The presence of internal promoter in the pol region has been reported in HTLV-I,46 which might cause the transcription of tax gene without 5′-LTR. However, 5 of our 14 ATL cases with type 2 defective provirus also lacked the internal promoter and the second exon of tax gene. In such cases without 5′-LTR and internal promoter, the tax gene transcription could not be detected. However, the tax gene was transcribed in 3 cases with type 2 defective provirus, indicating that internal promoter or trapped cellular promoter was responsible for such tax gene transcription. Taken together, loss of 5′-LTR could not abolish the tax gene transcription, but diminish that.
In conclusion, we show that genetic and epigenetic changes of viral promoters attenuate Tax expression and function, which is considered to enable ATL cells to escape from host immune system. Further studies are necessary to clarify the epigenetic mechanism to silence the tax gene transcription.
We thank M. Fujii for valuable suggestions. KN is a Research Fellow of the Japan Society for the Promotion of Science.
- 1Adult T cell leukemia in Japan. In: SenoS, TakakuF, IrinoS, eds. Topic in hematology, the 16th International Congress of Hematology. Amsterdam, 1977. 73–77., , , .
- 6ATL and HTLV-I-related diseases. In: TakatsukiK, ed. Adult T-cell leukemia. New York: Oxford University Press, 1994. 1–27., , .
- 8Adult T-cell leukemia/Lymphoma. In: GoedertJJ, ed. Infectious Causes of Cancer. Totowa: Humana Press, 2000. 221–29..
- 36Epstein-Barr virus and Burkitt's lymphoma, Goedert, J.J. Totowa: Humana Press, 2000. 77–92..