Resveratrol and endometrium: a closer look at an active ingredient of red wine using in vivo and in vitro models

Abstract

Resveratrol is a natural phytoestrogen with antiproliferative properties present in red wine, grapes, and berries. Published reports on the effects of resveratrol in human endometrial function are limited. The objective of this study was to investigate the expression of estrogen receptor α (ESR1), Ki-67 (a proliferative marker), aryl hydrocarbon receptor (AhR), and members of the cytochrome P450 superfamily of enzymes (CYP1A1 and CYP1B1) in an in vitro and vivo assay. Alkaline phosphatase assay of estrogenicity was used to compare estrogen activity of different concentrations of resveratrol to estradiol (E2) and diethylstilbestrol (DES), using Ishikawa cell culture. Immunohistochemical expression of ESR1 and Ki67, and reverse transcriptase polymerase chain reaction of AhR, CYP1A1, and CYP1B1 were analyzed from xenograft implants of human endometrial tissue in ovariectomized immunodeficient RAG-2-γ(c) mice, after 30 days of treatment with subcutaneous pellets of E2, E2 plus progesterone (P4), or E2 plus resveratrol (6, 30, or 60 mg) for 30 days. Compared to E2, resveratrol acted as an agonist and antagonist of estrogen in low and high concentrations, respectively, when combined with E2. Xenografts of human endometrial tissues in RAG-2 mice exhibited reduced expression of ESR1 and proliferative activity (Ki67) with 60 mg of resveratrol. This study suggests that resveratrol, at high doses, has the potential benefit to reduce proliferation of human endometrium through ESR1.

Keywords: Ishikawa; cell proliferation; endometriosis; estrogen receptor; human endometrium; resveratrol.

Figure 1.

Composite showing the RAG2-γ(c) knockout mouse model for endometriosis. Subcutaneous implants of human endometrium were stimulated by treatment with estrogen pellets for 4 weeks. The implants are visible under the skin (A), visible endometrial xenografts (B) are harvested (C) and the histologic appearance is shown in D (scale bar = 100 µm). In a typical experiment, 2 implants would be placed and the mice treated with E2 plus 1 of 3 doses of resveratrol prior to tissue collection after 30 days.

Figure 2.

A, Estrogenicity in Ishikawa cells after treatment of estradiol during different time periods. Alkaline phosphatase assay of Littlefield provided a means to compare the estrogenicity. ANOVA P < .0001, n = 4 in each group. B, Estrogenicity obtained from Ishikawa cells treated for 3 days with different concentrations of resveratrol in the presence or absence of estradiol or DES at 10⁻⁸ mol/L, or with DES + ICI (an antiestrogen). Bars represent mean (±SEM); post hoc test (pairing letters), all P < .05. ANOVA indicates analysis of variance; DES, diethylstilbestrol; ICI, Imperial Chemical Industries; SEM, standard error of the mean.

Figure 3.

Immunohistochemical appearance of estrogen receptor α (ERα) and the proliferation marker Ki67 in human eutopic endometrium used as a xenografts tissue recovered from the mouse model (mean ± SEM). A, ERα expression in epithelial endometrium is reduced in the presence of treatment with estradiol (1.6 μg/d) and progesterone (357 μg/d) and with estradiol (1.6 μg + resveratrol 60 mg/d. B, ERα expression is reduced in stroma cells, compared to those treated with estradiol only. Ki-67 expression is reduced in endometrial stromal cells treated with estradiol plus progesterone (C and D). All ANOVA—Dunnett post hoc test (pairing letters)—All P ≤ .01. Epi indicates epithelial endometrium; SEM, standard error of the mean.

Figure 4.

Photomicrograph of human eutopic endometrium used a xerograph in RAG mouse. Estrogen treated implants contained more estrogen receptors and Ki67, a marker of proliferation, than the implants from mice treated with E2 plus progesterone (E2 + P). In contrast to the E2-treated animals, mice treated with 60 mg of resveratrol (E2 + R) had reduced Ki67 and ERα immunostaining. Magnification ×400. E2 indicates estradiol; P, progesterone; R, resveratrol; ERα, estrogen receptor α.

Figure 5.

Comparisons are shown for AhR, Cyp1A1, and Cyp1B1 mRNA by RT-PCR (mean ± SEM) in human xenografts endometrial tissue from RAG-2 mice treated with estradiol (E2), E2 plus progesterone (P), or E2 plus 60 mg of resveratrol. No significant difference was found among groups. AhR indicates aryl hydrocarbon receptor; mRNA, messenger RNA; RT-PCR, reverse transcriptase polymerase chain reaction; SEM, standard error of the mean.