Identification of the Core Methylated CpG Regions Responsible for Silencing of steroid-5-alpha-reductase 2 (SRD5A2) in Human Prostate
Zongwei Wang, PhD1; Hongbo Wang, PhD1, Rongbin Ge, MD; PhD2; Shulin Wu, MD1; Shahin Tabetabaei, MD1; Chin-Lee Wu, MD; PhD1; Aria F. Olumi, MD3
1Massachusetts General Hospital, Boston, MA; 2UMass Memorial Medical Center, Worcester, MA; 3Beth Israel Deaconess Medical Center, Boston, MA
BACKGROUND:
5-alpha reductase type 2 (SRD5A2), an enzyme that is critical for prostatic development and growth, is utilized as an inhibitory target by finasteride for patients with bladder outlet obstruction (BOO) secondary to benign prostatic hyperplaisa (BPH). However, we have found that many aging benign prostate tissues do not express the enzyme. We have demonstrated that hypermethylation affects expression of SRD5A2 gene and that DNA methylation of SRD5A2 gene is regulated by DNMT1 in human BPH tissues. Here we define the methylation pattern and identify specific core CpG dinucleotides that account for absence of SRD5A2 expression in human adult prostate tissues when methylated.
METHODS:
Sixteen prostate samples from patients who were treated by transurethral resection of prostate (TURP) for BOO secondary to BPH were used. In order to perform methylation profiling of this gene, the regulatory regions from the Ensembl Regulatory Built and the CpG islands based on the CpG islands identified by the UCSC Genome Browser are the focus of the in silico design. These designs cover the regulatory components that include 2 CpG islands, 1 CTCF binding site and one open chromatin region in the SRD5A2 promoter regulatory region. The initial assessment resulted in a total of 14 in silico designs that cover the CpG sites. Process of testing: DNA samples were processed for direct bisulfite modification using EZ DNA Methylation Direct Kit, then performed gradient PCR, followed by capillary electrophoresis of the PCR products using the QIAxcel Advanced System. Libraries were prepared using the KAPA Library Preparation Kit, followed by library molecules purification, quantification, and sequencing. Finally, FASTQ files were aligned to the local reference database using open-source Bismark Bisulfite Read Mapper with the Bowtie2 alignment algorithm. Methylation levels were calculated in Bismark by dividing the number of methylated reads by the total number of reads.
RESULTS:
Eight samples had high SRD5A2 expression and eight samples had low SRD5A2 expression confirmed by immunohistochemistry. With deep CpG dinucleotide methylation analysis we found the highest percentage of methylation in the regions -813 (CpG#-60) to -630 (CpG#-45) and -266 (CpG# -29) to -26 (CpG#-6) of SRD5A2 low-expressing BPH tissues compared with SRD5A2 high-expressing tissues. The CpG dinucleotides that most notably differentiated between expressing and non-expressing SRD5A2 prostate tissues were located at nucleotide -813(CpG# -60), -630 (CpG#-45), -266(CpG# -29), -165(CpG# -21), -160(CpG# -20), -158(CpG# -19), -47(CpG# -7), and -26(CpG# -6). These CpG hotspots can be divided into two broad regions: nucleotide -813 to -630 and segment -266 to -26.
CONCLUSIONS:
SRD5A2 low-expression samples showed significantly higher methylation patterns than SRD5A2 high-expression samples. Methylation of two specific regions on the promoter (-813 to -630 and -266 to -26) most notably determine expression of SRD5A2. The specific CpG dinucleotides that are methylated will enable us to evaluate mechanisms and patterns of SRD5A2 promoter methylation to study the functional significance of SRD5A2 methylation, and will enable the first effort to decipher molecular mechanisms accounting for unresponsiveness to the SRD5A2 inhibitors for management of BPH.
Funding Source: NIH/R01 DK091353
Back to 2018 Program