The Polycomb Group Protein EZH2 Impairs DNA Damage Repair Gene Expression in Human Uterine Fibroids

Abstract

Uterine fibroids are benign, smooth muscle tumors that occur in approximately 70%-80% of women by age 50 yr. The cellular and molecular mechanism(s) by which uterine fibroids (UFs) develop are not fully understood. Accumulating evidence demonstrates that several genetic abnormalities, including deletions, rearrangements, translocations, as well as mutations, have been found in UFs. These genetic anomalies suggest that low DNA damage repair capacity may be involved in UF formation. The objective of this study was to determine whether expression levels of DNA damage repair-related genes were altered, and how they were regulated in the pathogenesis of UFs. Expression levels of DNA repair-related genes RAD51 and BRCA1 were deregulated in fibroid tissues as compared to adjacent myometrial tissues. Expression levels of chromatin protein enhancer of zeste homolog 2 (EZH2) were higher in a subset of fibroids as compared to adjacent myometrial tissues by both immunohistochemistry and Western blot analysis. Treatment with an inhibitor of EZH2 markedly increased expression levels of RAD51 and BRCA1 in fibroid cells and inhibited cell proliferation paired with cell cycle arrest. Restoring the expression of RAD51 and BRCA1 by treatment with EZH2 inhibitor was dependent on reducing the enrichment of trimethylation of histone 3 lysine 27 epigenetic mark in their promoter regions. This study reveals the important role of EZH2-regulated DNA damage-repair genes via histone methylation in fibroid biology, and may provide novel therapeutic targets for the medical treatment of women with symptomatic UFs.

Keywords: BRCA1; DNA damage repair; EZH2; H3K27me3; RAD51; fibroid.

FIG. 1

Expression levels of DNA damage repair genes RAD51 and BRCA1 in fibroids (F) and adjacent myometrial tissues (MyoF) as well as primary cells. Each experiment was conducted in replicates. A) Expression of RAD51 was determined by real-time PCR in fibroid and adjacent myometrial tissues. *P < 0.05 compared with the respective control (n = 5). B) Expression of BRCA1 was measured by real-time PCR in fibroid and adjacent myometrial tissues. *P < 0.05 compared with the control (n = 5). Expression of RAD51 (C) and BRCA1 (D) was determined by real-time PCR in primary cells from a fibroid and adjacent myometrial tissue. The cells from a fibroid and from myometrium tissue were plates in 6 × 60-mm dishes, including three dishes for primary fibroid cells and three dishes for primary myometrium cells. Three biological samples from both fibroid and myometrium samples were subjected to RNA isolation, cDNA synthesis, and q-PCR analysis. *P < 0.05 compared with the respective control.

FIG. 2

Restoring of RAD51 and BRCA1 in primary cells from fibroids treated with EZH2 inhibitor. A) Expression of RAD51 in primary cells from fibroid tissues was determined by real-time PCR in the presence or absence of 5-Aza-dC, TSA, and DZNep, respectively. *P < 0.05 compared with the respective control. B) Expression of BRCA1 in primary cells from fibroid tissues was determined by real-time PCR in the presence or absence of 5-Aza-dC, TSA, and DZNep, respectively. *P < 0.05 compared with the control.

FIG. 3

Expression of EZH2 expression in fibroid tissues and adjacent myometrium. A) Immunohistochemistry of EZH2 expression in fibroid and adjacent tissues. Arrows indicate EZH2-positive cells (magnification ×400). B) Semiquantitative analysis of EZH2-positive cells in myometrium and fibroids from fibroid and matched myometrium samples (n = 2). C) Western blot analysis of EZH2 in fibroid and adjacent tissues from African American women (n = 8). Total lysates from fibroids and adjacent myometrial tissues were extracted and subjected to Western blot analysis using antibody against EZH2. βactin was used as an endogenous control. Asterisks indicate the upregulation of EZH2 in fibroid tissues as compared to adjacent myometrium tissues. EZH2 protein bands were quantified and normalized to βactin, and relative values were used to generate data graphs (C, bottom panel). F, uterine fibroid; M, adjacent myometrial tissue.

FIG. 4

EZH2 directly targets epigenetic mark H3K27me3 in fibroid primary cells. A) Introduction of EZH2 into HuLM cells by adenoviral transduction showed a dose-dependent increase in the expression levels of EZH2. B) Expression levels of H3K27me3 were markedly upregulated in Ad-EZH2 transduced cells in comparison to control in a dose-dependent manner. *P < 0.05 compared with the control. C) Protein lysates from fibroid primary cells treated with different concentrations of EZH2 inhibitor (DZNep) for 3 days were subjected to Western blot analysis using antibody against H3K27me3 and HeK4me3, respectively. βactin was used as an endogenous control. D) Densitometry quantification of H3K27me3 levels for C. E) Densitometry quantification of H3K4me3 levels for C.

FIG. 5

Overexpression of EZH2 decreases expression levels of DNA damage repair and cell cycle arrest-related genes. Overexpression of EZH2 was achieved by adenoviral transduction approach in HuLM cells. The expression of EZH2 (A), RAD51 (B), BRCA1 (C), and p21 (D) was determined by real-time PCR after infection of HuLM cells with Ad-EZH2 (20 pfu) and Ad-GFP (20 pfu) for 3 days. *P < 0.05 compared with the control.

FIG. 6

DZNep treatment efficiently inhibited fibroid primary cell proliferation by increased expression of P21 and decreased protein expression of PCNA. A) Fibroid (F) and MyoF primary cells were grown in media containing either DZNep (0.25 and 1 μM) or vehicle (DMSO) for 5 days. Cell proliferation was assessed by cell counting. Trypan blue staining was used to distinguish nonviable from viable cells. P < 0.05 compared with the respective control. B) Real-time PCR was performed to determine expression levels of p21 in fibroid primary cells treated with DZNep or vehicle. C) Real-time PCR was performed to determine expression levels of p21 in MyoF primary cells treated with DZNep or vehicle. D) Western blot analysis was performed to examine expression levels of PCNA in fibroid primary cells treated with varying concentrations of DZNep. E) Western blot analysis was performed to examine expression levels of PCNA in MyoF primary cells treated with varying concentrations of DZNep. *P < 0.05 compared with the control.

FIG. 7

DZNep treatment changes cell phase distribution. Fibroid cells (2.5 × 10⁵) were grown in a 100-mm dish. The procedure for processing the cells for cell cycle analysis appears in the Materials and Methods section. Representative cell cycle analysis in fibroid cells treated with DMSO (A), 1 μM DZNep (B), and 5 μM DZNep (C) for 4 days. D) Comparison of cell phase distribution in fibroid cells treated with DMSO, 1 μM DZNep, and 5 μM DZNep, respectively. Each column represents the mean of three independent experiments. *P < 0.05; ^#P < 0.05.

FIG. 8

Measurement of cell apoptosis in response to DZNep treatment. Primary fibroid cells were grown in a 100-mm dish in the presence or absence of DZNep (5 μM) for 3 days. DMSO treatment served as a control. Cells were stained with both annexin-V and EthD-III. Apoptosis was measured by flow cytometry. Double-negative staining representing living cells is shown in quadrant Q4; positive staining for annexin-V and negative staining for PI represent the early apoptotic stage shown in quadrant Q3; double-positive staining representing the late apoptotic stage is shown in quadrant Q2; and dead cells are presented in quadrant Q1. Each column represents the mean of three independent experiments.

FIG. 9

DZNep treatment restores expression levels of RAD51 and BRCA1 through epigenetic mark H3K27me3. A) Location of regions analyzed by ChIP/PCR along human RAD51 and BRCA1 promoters. The position of the transcriptional start site was designated as +1. Short horizontal lines indicate the regions analyzed by ChIP/PCR. B) ChIP/quantitative PCR was performed with anti-H3K27me3 antibody in the RAD51 promoter region in fibroid primary cells in the presence or absence of DZNep. C) ChIP/quantitative PCR was performed with anti-H3K4me3 antibody in the RAD51 promoter region in fibroid primary cells in the presence or absence of DZNep. D) ChIP/quantitative PCR was performed to determine the enrichment of H3K27me3 in the distal and proximal promoter regions of BRCA1 in the presence or absence of DZNep. E) ChIP/quantitative PCR was performed to determine the enrichment of H3K4me3 in the distal and proximal promoter regions of BRCA1 in the presence or absence of DZNep. *P < 0.05 compared with the control.