Gene-body hypermethylation controlled cryptic promoter and miR26A1-dependent EZH2 regulation of TET1 gene activity in chronic lymphocytic leukemia

The Ten-eleven-translocation 1 (TET1) protein is a member of dioxygenase protein family that catalyzes the oxidation of 5-methylcytosine to 5-hydroxymethylcytosine. TET1 is differentially expressed in many cancers, including leukemia. However, very little is known about mechanism behind TET1 deregulation. Previously, by characterizing global methylation patterns in CLL patients using MBD-seq, we found TET1 as one of the differentially methylated regions with gene-body hypermethylation. Herein, we characterize mechanisms that control TET1 gene activity at the transcriptional level. We show that treatment of CLL cell lines with 5-aza 2´-deoxycytidine (DAC) results in the activation of miR26A1, which causes decrease in both mRNA and protein levels of EZH2, which in turn results in the decreased occupancy of EZH2 over the TET1 promoter and consequently the loss of TET1 expression. In addition, DAC treatment also leads to the activation of antisense transcription overlapping the TET1 gene from a cryptic promoter, located in the hypermethylated intronic region. Increased expression of intronic transcripts correlates with decreased TET1 promoter activity through the loss of RNA Pol II occupancy. Thus, our data demonstrate that TET1 gene activation in CLL depends on miR26A1 regulated EZH2 binding at the TET1 promoter and silencing of novel cryptic promoter by gene-body hypermethylation.

The siRNA transfections and over expression assays, were performed using custom designed EZH2 siRNA and negative control siRNA (Invitrogen, Carlsbad, USA), miR26A1 mimic miRNA and miR26A1 inhibitor miRNA (Thermoscientific, Waltham, USA). Transient transfection was carried out on an Amaxa Nucleofection Device II (Nucleofector 2b device, Lonza group AG, Basel, Switzerland) according to the manufacturer instructions. The Luciferase assays were performed in MCF-7 cells, using Lipofectamine LTX plus reagent kit (Invitrogen, Carlsbad, USA) according to manufacture instructions.

Pyrosequencing assay
Pyrosequencing was performed according to the manufacturer's instructions using the PyroMark™ Q24 Advanced pyrosequencer (Qiagen, Hilden, Germany). The Genomic DNA was bisulfite converted using the EZ DNA Methylation Gold kit (Zymo Research, Irvine, USA) according to manufacturer's protocol. The PyroMark™ software (Qiagen, Hilden, Germany) was applied to design pyrosequencing primer sets. The percentage of DNA methylation (average value for all the CpG sites in the target region) used for statistical analysis was calculated using PyroMark Q24 advanced software. The statistical analysis and the calculation of p value for comparison between CLL samples and normal controls were done using Statistica12.0 (Stat Soft, Tulsa, USA).

Luciferase reporter assays
First, we amplified HMR region (632bp) and fine mapped cryptic promoter TSS region containing region (519bp) using primer sequences anchoring restriction sites for cloning these fragments directly into PGL3 multiple cloning sites in both orientations. The 5'-ends of the forward primers were added using the Kpn1 restriction site and the 5'-ends of reverse primers were added using the XhoI restriction site and for reverse orientation it is vice versa (primer sequences shown in Supplementary  Table 1). Two control vectors were used for transfections along with the cloned fragment vectors; positive control vector containing strong SV40 promoter (Promega, Madison, WI, USA) and the negative control vector without any promoter (Promega, Madison, WI, USA). The entire cloned sequence of the amplified fragments of PGL3 Luciferase reporter vectors were verified by sanger sequencing before transiently transfected using Lipofectamine LTX plus reagent kit (Invitrogen, Carlsbad, USA) in MCF-7 cells according to manufacture instructions in the presence of the β-galactosidase reporter gene (Promega, Madison, USA).
The relative light units were measured with a luminometer (Glomax 20/20 luminometer, Promega, Madison, USA). Each experiment was repeated for at least three independent sets, and the final luciferase values (relative light units) were calculated by dividing the luciferase activity by the β-galactosidase activity. Data represent average ± standard deviation (SD) of three independent experiments performed in triplicates.
The PCR analysis was performed with the 7900HT fast real-time PCR system instrument and software (Applied Biosystems, Warrington, UK). The average percentage of input and fold enrichment was calculated which represents the enrichment of histone modification or binding of transcription factor on specific region of TET1 gene using the ChIP reactions performed in triplicates.

Western blot analysis
Western blot analysis was performed using total cell lysates lysed from transfected CLL cell line samples using RIPA buffer (Sigma-aldrich, St.Louis, USA) with PI inhibitors (Roche, Basel, Switzerland). Equal amounts of lysates were loaded on NUPAGE 10% Bis-Tris gels (Invitrogen, Carlsbad, USA) and transferred to membranes (Amersham Hybond ECL; GE Health Care Life Sciences, Sweden). After blocking in 5% BSA with TBS with addition of 0.1% Triton X-100, the membranes were incubated with the appropriate primary and secondary antibodies, followed by washes with TBS containing 0.05% Triton X-100. Blots were visualized with SuperSignal West Dura Extended Duration Substrate (Thermo Scientific, Rockford, USA) using the ChemiDoc XRSC (Bio-Rad) instrument. The primary antibodies used for western blotting were: TET1 ( ab191698, Abcam, Cambridge, UK), EZH2 (3147, Cell Signaling Technology, Danvers, USA) and GAPDH (SC-25778; Santa Cruz Biotechnology, Dallas, USA), while the secondary antibodies were anti-mouse IgG (7076S, Cell Signaling Technology, Danvers, USA) and anti-rabbit IgG (7074S, Cell Signaling Technology, Danvers, USA).

TET1 expression using RNA sequencing data obtained from published CLL data set
We obtained the raw data of RNA-seq samples for 96 patients along with 9 normal B cell samples as described in our earlier paper [1]. In short, the obtained gene expression profile was normalized to reads per kilobase of transcript per million mapped reads (RPKM). The log-fold changes between B cell normal and two CLL groups (IGHV-mutated and IGHV-unmutated) were calculated based on obtained RPKM values. Wilcoxon rank sum test in R package is used for calculating statistical p value for differential expression between normal and CLL prognostic groups.    The percentage of DNA methylation at HMR of TET1 gene using DAC treated and untreated MEC1 and HG3 CLL cell lines. (E) Detection of anti-sense intronic transcripts using strand specific Reverse Transcription PCR. 5'GSP and 3'GSP are designed for specifically synthesizing cDNA from lower anti-sense strand and upper sense strand respectively. Lanes 1 to 7 shows the PCR amplified products from 5'GSP and 3'GSP cDNA synthesis using control siRNA and EZH2 siRNA MEC1 cell line samples.