Direct activation of the HDAC8 promoter by SOX4. (A) A schematic depiction of the HDAC8 promoter region from –712 to +126 bp with potential cis regulatory elements. (B) Mutation analysis. Luciferase reporter assays were performed with the HDAC8 promoter fragment from –712 to +126 bp with the SOX4 site at –243 to –236 bp being wild-type or mutated (ΔSOX4, from AACAAGGA to ACCATGGA). Data are presented as mean ± SEM of 3 separate experiments. *P < .05. (C) Reconstitution experiment. Raji cells were cotransfected with pGL3-HDAC8 (–712/+126) or pGL3-HDAC8 (–712/+126) ΔSOX4 and the SOX4 expression vector or control expression vector as indicated. After 24 hours, luciferase assays were performed in triplicate. Promoter activation was expressed as fold induction of luciferase activity in cells transfected with the SOX4 expression vector vs cells transfected with the control expression vector. Transfection efficiency was normalized by β-galactosidase activity. Each bar represents mean ± SEM of 3 separate experiments. *P < .05. (D) Effect of broad-spectrum HDAC inhibitors on cell growth. ATL cell lines (HUT102 and ST1) were treated with TSA, SAHA, or vehicle (DMSO) and cultured in a 96-well plate at 0.5 × 10⁴ cells per well. At indicated time points, viable cells were counted on FACSCalibur by gating out cells stained with propidium iodide. Data are shown as mean ± SEM of 3 separate experiments. *P < .05; **P < .01. (E) Effect of HDAC8 siRNA on cell growth. Two ATL cell lines (HUT102 and ST1) and Raji were transfected with control siRNA or HDAC8 siRNA. After 48 hours, quantitative PCR was performed for HDAC8 and GAPDH. Levels of HDAC8 expression were normalized by GAPDH (left). Cells were also cultured in a 96-well plate at 0.5 × 10⁴ cells per well. After 72 hours, viable cells were counted on FACSCalibur by gating out cells stained with propidium iodide (right). Data are presented as mean ± SEM of 3 separate experiments. *P < .05.