Genistein effects on stromal cells determines epithelial proliferation in endometrial co-cultures

Abstract

Background: Estrogen is the leading etiologic factor for endometrial cancer. Estrogen-induced proliferation of endometrial epithelial cells normally requires paracrine growth factors produced by stromal cells. Epidemiologic evidence indicates that dietary soy prevents endometrial cancer, and implicates the phytoestrogen genistein in this effect. However, results from previous studies are conflicting regarding the effects of genistein on hormone responsive cancers.

Methods: The effects of estrogen and genistein on proliferation of Ishikawa (IK) endometrial adenocarcinoma cells were examined in co-cultures of IK cells with endometrial stromal cells, recapitulating the heterotypic cell-to-cell interactions observed in vivo. The roles of estrogen receptor (ER)α and ERβ were evaluated using ERα and ERβ specific agonists. ER activation and cell proliferation in the IK epithelial cells were determined by alkaline phosphatase assay and Coulter counter enumeration, respectively.

Results: Both estrogen and genistein increased estrogen receptor-induced gene activity in IK cells over a range of concentrations. Estrogen alone but not genistein increased IK proliferation in co-cultures. When primed by estrogen treatment, increasing concentrations of genistein produced a biphasic effect on IK proliferation: nM concentrations inhibited estrogen-induced proliferation while μM concentrations increased proliferation. Studies with an ERβ-specific agonist produced similar results. Genistein did not influence the effects of estrogen on IK proliferation in monoculture.

Conclusions: Our study indicates that nutritionally relevant concentrations (nM) of genistein inhibit the proliferative effects of estrogen on endometrial adenocarcinoma cells presumably through activation of stromal cell ERβ. We believe that sub-micromolar concentrations of genistein may represent a novel adjuvant for endometrial cancer treatment and prevention.

Figure 1. Estrogen and genistein induction of ER-regulated alkaline phosphatase activity in IK cells in endometrial co-cultures

A. Estrogen activates the estrogen receptor in endometrial epithelial cells co-cultured with endometrial stromal cells as determined by measuring alkaline phosphatase activity, a known gene product of estrogen receptor activation. Column 1 = vehicle control, columns 2 through 7 are estrogen treatments at 0.01, 0.1, 1.0, 10, 100, 1000 nM. Results measured by the ordinate are the ratio of the estrogen treated co-cultures to the untreated vehicle controls, expressed as fold-increases. B. Genistein activates the estrogen receptor over a range of physiologic concentrations; column 1 = vehicle control, columns 2 through 6 are genistein treatments at 10, 50, 100, 500 and 1000 nM genistein. The height of the columns represents the average of samples studied (N ≥ 8). **P < 0.01.

Figure 2. Effects of estrogen and genistein on epithelial cell proliferation in endometrial co-cultures

A. Estrogen increases endometrial epithelial cell proliferation in co-culture in a dose dependent manner; column 1 = vehicle control, columns 2 – 6 = 0.1, 1.0, 10, 100 and 1000 nM estrogen. Results measured by the ordinate are the ratio of the estrogen treated co-cultures to the untreated vehicle controls, expressed as fold-increases. B. Genistein fails to induce endometrial epithelial proliferation in co-culture at concentrations shown to activate the estrogen receptor (Figure 1B); column 1 = vehicle control, columns 2 – 4 = 1.0, 10 and 100 nM genistein. The height of the columns represents the average of samples studied (N ≥ 4). **P < 0.01.

Figure 3. Effects of estrogen receptor α- or β-specific agonists (PPT and DPN) on ER-regulated alkaline phosphatase activity in endometrial co-cultures

A) The ERα-specific agonist PPT stimulates ER-regulated alkaline phosphatase activity over a concentration range of 1.0 to 1000 nM (columns 3 to 6) when compared to vehicle control (column 1), but does not achieve an activity equivalent to that of 10 nM estrogen (column 2). Results measured by the ordinate are the ratios of the estrogen or PPT-treated co-cultures to the untreated vehicle controls, expressed as fold-increases. B) The ERβ-specific agonist DPN stimulates ER-regulated alkaline phosphatase activity at concentrations ranging from 10 to 1000 nM (columns 4 to 6) when compared to vehicle control (column 1), but not at 1.0 nM (column 3). DPN also does not achieve an activity equivalent to that stimulated by 10 nM estrogen (10 nM, column 2) at the highest concentration tested (column 6). Results measured by the ordinate are the ratios of the estrogen or PPT-treated co-cultures to the untreated vehicle controls, expressed as fold-increases. The height of the columns represents the average of the samples studied (N ≥ 11). **P < 0.01 versus vehicle control.

Figure 4. Effects of estrogen receptor α- or β-specific agonists (PPT and DPN) on epithelial cell proliferation in endometrial co-cultures

Estrogen (10 nM, column 2) and the ERα-agonist PPT (10 nM, column 3) both increased endometrial epithelial cell proliferation in co-cultures significantly when compared to the vehicle control (column 1). The ERβ-agonist DPN (10 nM, column 4) had no effect on endometrial epithelial cell proliferation. Results measured by the ordinate are the ratios of the estrogen or receptor-specific agonists to the untreated vehicle controls, expressed as fold-increases. Data shown represents the average of the samples studied (N ≥ 3). **P < 0.01.

Figure 5. Effects of estrogen and genistein co-treatment on epithelial cell proliferation in monocultured versus co-cultured endometrial epithelial cells

A. In monocultures, estrogen induces a modest but significant increase in IK cell proliferation (column 2 = 10 nM estrogen). Concurrent treatment with 10 or 1000 nM genistein (column 3 = 10 nM estrogen + 10 nM genistein; column 4 = 10 nM estrogen + 1000 nM genistein) did not change IK cell proliferation. Results are presented as the ratios of the estrogen or estrogen plus genistein to the untreated vehicle controls (column 1) shown as fold-increases (ordinate). B. In co-cultures, genistein inhibits estrogen induced IK proliferation (10 nM estrogen, column 2) in a biphasic manner. Physiologic concentrations of genistein (1 to 100 nM) decreased estrogen-induced proliferation by 15%, 30% and 45% at 1.0 nM (column 3), 10 nM (column 4) and 100 nM (column 5), respectively. At a higher concentration (1000 nM), genistein significantly increases the proliferative effects of 10 nM estrogen (column 6) above that of estrogen alone (column 2). Results are presented as the ratios of the estrogen or estrogen/genistein combinations to the untreated vehicle controls (column 1) and are shown as fold-increases (ordinate). Data shown represent the average of the samples studied (N ≥ 3). *P < 0.05 versus vehicle control, # P < 0.05 versus estrogen-treated sample; **P < 0.01 versus vehicle control.

Figure 6. Effect of the ERβ-agonist DPN on epithelial cell proliferation of IK cells in endometrial co-culture

The estrogen receptor (ER)-β subtype specific agonist DPN inhibited estrogen-induced IK proliferation in co-culture over a range of concentrations similar to that produced by increasing concentrations of genistein. IK co-cultures were treated with 10 nM estrogen (column 1) or increasing concentrations of DPN (0.1, 1.0, 10 and 100 nM; columns 2 – 5, respectively). Inhibition of estrogen induced proliferation was maximal at a concentration of 10 nM DPN, above which DPN became less inhibitory. Results are presented as the ratios of the estrogen/DPN combinations to the 10 nM estrogen-alone control (column 1) and are shown as fold-increases (ordinate). **P < 0.01.