MicroRNA 135 regulates HOXA10 expression in endometriosis

Abstract

Context: Homeo box A10 (HOXA10) regulates endometrial receptivity and its expression is decreased in women with endometriosis. Although sex steroids regulate HOXA10, these hormones are unaltered in endometriosis. We hypothesized a role for microRNA in the regulation of HOXA10.

Objective: MicroRNA 135a and -b are small noncoding RNA with predicted targets that include HOXA10. We evaluated miR135a/b expression and HOXA10 regulation in endometrium from subjects with and without endometriosis.

Design: The design of the study was the measurement of miR135a/b expression by quantitative PCR and in vitro analysis of HOXA10 regulation.

Setting: The study was conducted at a university medical center.

Patients: Patients included 50 controls and 32 women with endometriosis.

Interventions: Study interventions included endometrial biopsies and in vitro transfection.

Main outcome measures: miR135a/b and HOXA10 expression were measured in the study.

Results: All endometrial samples expressed miR135a and -b. miR135a expression in controls was increased during the proliferative phase, decreased at the time of ovulation, and increased during the luteal phase. Subjects with endometriosis had 3-fold higher expression of miR135a in the proliferative phase than controls. miR135b showed less variation across the menstrual cycle; however, it was significantly increased in women with endometriosis in the proliferative and secretory phases. HOXA10 expression was simultaneously repressed in the endometrium of women with endometriosis. Transfection of endometrial stromal cells with mir135a/b or miR135a/b inhibitors resulted in the altered expression of HOXA10 mRNA and protein. miR135a or -b decreased luciferase expression driven by the HOXA10 3' untranslated region containing the miR135 binding site. miR135a regulation of HOXA10 was absent in MCF-7 cells, demonstrating cell specificity.

Conclusions: HOXA10 was aberrantly regulated in the endometrium of women with endometriosis by both miR135a and miR135b. Increased microRNA expression likely suppresses genes required for implantation.

Fig. 1.

miR135a, miR135b, and HOXA10 expression during the menstrual cycle. A, miR135a and miR135b expression in endometrium of controls and women with endometriosis measured by qPCR and normalized to U6 expression. Top panel, In controls (solid bars), expression of miR135a was increased during the proliferative phase of the menstrual cycle, dropped after ovulation, but increased again during the secretory phase. Endometrial expression of miR135a was increased in the proliferative phase of women with endometriosis (open bars) compared with that of controls. *, P < 0.05, expression in endometriosis compared with controls. Bottom panel, miR135b expression is distinct from miR135a expression. Peak expression occurs in the early secretory phase when miR135a is at a nadir. The expression of miR135b is higher in endometrium from women with endometriosis than controls in the proliferative, mid-, and late secretory phases of the menstrual cycle. *, P < 0.05, expression in endometriosis compared with controls. Shown are miRNA expression normalized to U6 expression mean ± sem. P, Proliferative; ES, early secretory; MS, midsecretory; LS, late secretory. B, HOXA10 expression in the endometrium of women without and without endometriosis evaluated by qPCR and normalized to β-actin expression. Solid bars show mean HOXA10/β-actin expression in controls, and open bars are mean expression in endometrium from women with endometriosis. HOXA10 expression was diminished in the proliferative, midsecretory, and late secretory phases of endometrium from women with endometriosis relative to controls. *, P < 0.02, expression in endometriosis compared with controls. Error bars are sem. P, Proliferative; ES, early secretory; MS, midsecretory; LS, late secretory.

Fig. 2.

A, Expression of miR135a and miR135b in endometrial stromal cells treated with has-mir-135a or has-mir-135b inhibitor and has-mir-negative control. Transfection with either the miR135a or miR135b inhibitor construct significantly decreased the expression of the miRNA when compared with transfection with the control (CT). Error bars are sem. *, P < 0.01. B, HOXA10 mRNA expression is regulated by miR135a and miR135b. HOXA10 mRNA levels were decreased in endometrial stromal cells treated with either miR135a or miR135b compared with the corresponding microRNA control. *, P < 0.04 (top panel). Transfection with the miR135a inhibitor or with the miR135b inhibitor resulted in increased HOXA10 mRNA expression when compared with control treatment. *P < 0.01 (bottom panel). Shown are mean HOXA10 mRNA expression levels normalized to β-actin ± sem measured by quantitative real-time PCR using the 2^−ΔΔCT method.

Fig. 3.

HOXA10 protein expression after treatment with each miRNA or miR inhibitor. A, Endometrial cells were transfected with miR135a, miR135b, the miR135a or miR135b inhibitors, or the respective controls. Shown are representative Western blots showing HOXA10 protein expression (molecular mass 43 kDa) and corresponding glyceraldehyde-3-phosphate dehydrogenase (GAPDH) loading control (molecular mass 37 kDa). Top panel, HOXA10 protein expression was decreased by treatment with miR135a and increased by treatment with the miR135a inhibitor relative to the respective miR or inhibitor controls. Bottom panel, Similarly, HOXA10 protein expression was decreased by treatment with miR135b and increased by treatment with the miR135b inhibitor relative to the respective miR or inhibitor controls. Each experiment was repeated three times in triplicate. B, Quantification of the Western blot demonstrating decreased HOXA10 expression after treatment with miR135a or miR135b compared with the miRNA control (Ctl). Reciprocally, treatment with the miR135a or miR135b inhibitor s each led to increased HOXA10 expression compared with controls. HOXA10 densitometry was normalized to the glyceraldehyde-3-phosphate dehydrogenase (GAPDH) density from the same lane. miR135a decreased the HOXA10 protein expression by approximately 60%, whereas miR135b decreased HOXA10 expression by nearly 80%. *, P < 0.02. miR135a inhibitor treatment led to a nearly doubling of HOXA10 protein expression, whereas the miR135b inhibitor treatment more than doubled the HOXA10 expression. *, P < 0.05.

Fig. 4.

miR135a inhibitor increases HOXA10 expression in normal primary endometrial stromal cells and those from women with endometriosis. Primary endometrial stromal cells were obtained from women with and without endometriosis. To determine whether miR135 affects these cells equally, we chose a representative treatment and administered it to both normal cells and those from women with the disease. HOXA10 protein expression was similarly affected in each cell type. Shown are representative Western blot analyses. This experiment was repeated three times using cells from separate subjects, and each experiment was performed in triplicate. GAPDH, Glyceraldehyde-3-phosphate dehydrogenase.

Fig. 5.

miR135a and miR135b directly regulate HOXA10 through the HOXA10 3′ UTR. A luciferase reporter construct incorporating the HOXA10 3′ UTR that contained the predicted miR135a and miR135b binding site was transfect into endometrial stromal cells along with either miR135a or miR135b. Treatment with miR135a or miR135b resulted in decreased lucifierase activity compared with transfection with the miRNA control (Ctl). *, P < 0.03 and P < 0.02, respectively. Shown are mean and sem. Each experiment was repeated four times using cells from separate subjects, and each experiment was performed in triplicate.

Fig. 6.

miR135 regulation of HOXA10 expression in MCF-7 breast cancer cells. Like endometrial cells, MCF-7 cells express HOXA10. MCF-7 cells were transfected with miR135a, miR135b, miR135 control (Ctl), hsa-mir-135a (anti-a), hsa-mir-135b (anti-b), or hsa-mir-negative control (Ctl) for 72 h. HOXA10 mRNA expression was not significantly different between cells that were treated with miR135a or mir135b compared with control treated cells. HOXA10 expression was also not altered by treatment with the miR135a inhibitor. Treatment with the miR135b inhibitor led to increased HOXA10 mRNA expression. HOXA10 mRNA levels were measured by quantitative real-time PCR using the 2^−ΔΔCT method. *, P < 0.05 by Student t test.