Unique Epigenetic Code of the SHP-1 P2 Promoter in HTLV-1 Transformed T-Cells, a Possible Mechanism of Resistance to Chemotherapy for Adult T-Cell Leukemia.

Abstract 1269

Poster Board I-291

SHP1 is a protein tyrosine phosphatase with tumor suppressor function. SHP-1 downregulation has been observed in hematological malignancies. We showed HTLV-1 transformed cells have dense SHP-1 P2 promoter DNA methylation compared with HTLV-1 negative T-cell leukemia/lymphoma cells, and HTLV-1 Tax is responsible for the promoter silencing. To investigate the epigenetic mechanisms involved in SHP-1 P2 promoter regulation, we analyzed and compared expression of SHP-1 mRNA as well as histone code and transcription factors involved in the P2 promoter regulation in normal CD4+ T-cells, HTLV-1 negative T-cell lines and HTLV-1 transformed cell lines.

Expression of SHP-1 gene from its ubiquitous P1 promoter and hematopoeitic cell specific P2 promoter were first analyzed in normal CD4+ T-cells freshly isolated from healthy donors, HTLV-1 negative T-cell leukemia/lymphoma cell lines (Jurkat, CEM-T4, SupT1, MOLT4 and HUT78), HTLV-1 transformed cell lines (MT2, MT4, B1, C5, HUT102, C10/MJ and OS-P2), CML cell line (K562) and epithelial cell lines (Hela and 293T) through real time PCR designed to distinguish SHP-1 (I) from (II) transcripts. Upon normalizing the absolute copy number of SHP-1 transcript to that of GAPDH, the results indicate that HTLV-1 transformed cell lines express significantly lower level of SHP-1 (II) compared with that from HTLV-1 negative T-cell lines (mean 17.9 vs 666.6, p=0.0476) while their SHP-1 (I) expression were similar (mean 9.4 vs 7.2, p=0.5804). Note that CD4+ T-cells from healthy donors have much higher SHP-1 (I) (mean 606.8) and (II) (mean 7414.7) expression. This result suggests HTLV-1 transformation leads to specific repression of the SHP-1 P2 promoter.

DNA methyltransferase inhibitor decitabine (DAC) was tested for its effect on SHP-1 expression restoration in SHP-1 P2 promoter DNA methylation positive cell lines. DAC can effectively restore SHP-1 mRNA expression in HTLV-1 negative T-cells and CML cells but with little to no SHP-1 restoration both at mRNA and protein level in HTLV-1 transformed cells, indicating additional factors are involved in SHP-1 silencing in HTLV-1 transformed cells.

In order to further evaluate SHP-1 P2 promoter regulation, ChIP assay in combination with real time PCR analyses were performed with chromatin isolated from CD4+ T-cells, MOLT4, HUT78 and B1 cells. As expected, amount of Sp1 and RNA polymerase II binding to SHP-1 P2 promoter are proportional to the expression level of SHP-1 (II). The binding profile for DNA methyltransferase DNMT3a, DNMT3b, methyl DNA binding protein MBD1 and MBD3 were studied. DNMT3a and MBD1 associate with the P2 promoter only in B1 cells. At the histone level, although histone deacetylation at H3K9 and H3K27 was observed in both HUT78 and B1 cells, presence of HDAC1 on SHP-1 P2 promoter was only observed in B1 cells. In addition, MBD3, a subunit of the NuRD multisubunit complex containing nucleosome remodeling and histone deacetylase activities, also preferentially bind to the P2 promoter in B1 cells. Interestingly, H3K27 is specifically tri-methylated in B1 cells, which is accompanied with the presence of EZH2 and other factors involved in histone methylation. It is thus likely that both histone deacetylation and tri-methylation are responsible for SHP-1 silencing in HTLV-1 transformed cells.

Combining the results from these observations and our previous reports, we propose a “three steps” SHP-1 silencing model in T-cell neoplasia. First silencing step (represented by MOLT4) is mainly mediated by dissociation of transcription factors from SHP-1 P2 promoter, and low level of SHP-1 (II) expression is still observed. Second step (represented by HUT78) is DNA methylation mediated by DNMT3b and deacetylation at H3K9 and H3K27, in which SHP-1 (II) is very little, but the silencing is still reversible. Third step (represented by B1) is additional DNA methylation by DNMT3a, H3K9 and H3K27 deacetylation by HDAC1 and MBD3, and H3K27 tri-methylation and binding of EZH2, in which SHP-1 P2 promoter is tightly silenced irreversibly.

Our observation showed SHP-1 epigenetic silencing mechanisms is different between HTLV-1 transformed cells and HTLV-1 negative cells, which could contribute to establish a novel treatment strategy for ATL.

Presenter's Current address: Division of Hematology and Oncology, Ehime Prefectural Central Hospital, Matsuyama, Japan