The Phytoestrogen Quercetin Impairs Steroidogenesis and Angiogenesis in Swine Granulosa Cells In Vitro

Experimental evidence documents that nutritional phytoestrogens may interact with reproductive functions but the exact mechanism of action is still controversial. Since quercetin is one of the main flavonoids in livestock nutrition, we evaluated its possible effects on cultured swine granulosa cell proliferation, steroidogenesis, and redox status. Moreover, since angiogenesis is essential for follicle development, the effect of the flavonoid on Vascular Endothelial Growth Factor output by granulosa cells was also taken into account. Our data evidence that quercetin does not affect granulosa cell growth while it inhibits progesterone production and modifies estradiol 17β production in a dose-related manner. Additionally, the flavonoid interferes with the angiogenic process by inhibiting VEGF production as well as by altering redox status. Since steroidogenesis and angiogenesis are strictly involved in follicular development, these findings appear particularly relevant, pointing out a possible negative influence of quercetin on ovarian physiology. Therefore, the possible reproductive impact of the flavonoid should be carefully considered in animal nutrition.


Introduction
Quercetin (3,30,40,5,7-pentahydroxyflavone) is a flavonoid belonging to a group of plant-derived nonsteroidal compounds known as phytoestrogens [1]. In plants, these compounds are involved in energy production and exhibit strong antioxidant properties; in mammals, they have been shown to exert various biological and pharmacological effects [2,3]. Several studies [4][5][6] have demonstrated their significant health promoting activities, due to free radical scavenging and metal chelating properties. On the other hand, quercetin can also exhibit pro-oxidant effects [7]. It has been demonstrated [8][9][10][11] a relationship between quercetin free radical scavenging activity and its anticarcinogenic and anti-inflammatory properties. Most experimental data about the activity of flavonoids in different biological systems have been obtained by in vitro studies, and the knowledge of the pharmacokinetics and systemic availability of food-derived flavonoids in humans and animals is still incomplete: in particular, the bioavailability of quercetin appears low, but reported values range from 0% to 52% [12,13].
Recent data obtained in vitro or in animal studies indicate that flavonoids may also modify cell function independently of their antioxidant power [14][15][16], affecting the overall process of carcinogenesis by different mechanisms. Since it is well known that neovascularization represents a key process in tumor growth, invasion, and metastasis, a lot of scientific interest has developed in recent years about potential inhibitors of angiogenesis. Among natural health products, dietary flavonoids have shown to possess antiangiogenic effects, inhibiting several important steps of new vessel growth: an impairment of VEGF expression by different flavonoids has been documented [17,18], as well as inhibitory effects on proliferation, migration, tube formation of endothelial cells in vitro [19][20][21].
In particular, many experimental evidences suggest a close link between quercetin antineoplastic effect and its antiangiogenic potential [22][23][24][25]. While both the antioxidant and antiangiogenic properties of quercetin are generally related to its protective effects against oxidative stress and should also be relevant for its cancer preventive impact [11,26], the potential of the inhibition of vessel formation on female reproductive efficiency remains elusive; in fact, it is well known that the ovarian angiogenic process is strictly associated with follicular development [27]. Furthermore, ROSs play a crucial role in the ovarian follicle as signalling molecules in both the angiogenic cascade [28] and the ovulatory process. Additionally, phytoestrogens are able to bind to estrogen receptors thus activating several estrogenresponsive genes [3]. In particular, quercetin has been found to exhibit both estrogenic and antiestrogenic actions in vitro thus suggesting different potential effects on reproductive function.
Therefore, the aim of this study was to evaluate possible effects of quercetin on the function of granulosa cells, which play a complex and fundamental role in the development of the ovarian follicle. To this purpose, we studied the flavonoid action on granulosa cell proliferation, steroidogeneis, and redox status. Finally, since follicle growth requires new vessel formation, we also evaluated the effect of quercetin on the production of VEGF, the main proangiogenic factor.

Materials and Methods
All the reagents were obtained from Sigma (St. Louis, Mo, USA) unless otherwise specified.
Quercetin was dissolved in dimethyl sulfoxide (DMSO), and the final concentration of DMSO added to the medium was 0.2% (vol/vol). The same concentration of DMSO was added to specific control wells.

Granulosa Cell Proliferation.
Cell proliferation was evaluated by 5-bromo-2 -deoxyuridine (BrdU) incorporation assay test (Roche, Mannheim, Germany). Briefly, 10 4 cell/200 μL CM were seeded in 96-well plates and treated as described above. After addition of 20 μL BrdU to each well during the last 4 hour of incubation, culture media were removed, and DNA denaturing solution was added in order to improve the accessibility of the incorporate BrdU for antibody detection. Thereafter, 100 μL anti-BrdU antibody were added to each well. After a 1.5-hour incubation at room temperature (21 • C), the immune complexes were detected by the subsequent substrate reaction. The reaction product was quantified by measuring the absorbance at 450 nm against 690 nm using Spectra Shell Microplate reader (SLT Spectra, Milan, Italy). To establish viable cell number, absorbance was related to a standard curve prepared by culturing in quintuplicate granulosa cells at different plating densities (from 10 3 to 10 5 /200 μL) for 48 hours. The curve was repeated in four different experiments. The relationship between cell number and absorbance was linear (r = 0.92). Cell number/well was estimated from the resulting linear regression equation. The assay detection limit was 10 3 cell/well, and the variation coefficient was less than 5%. The number of cells obtained by this calculation was used for correcting hormones, VEGF production, and redox status data.

Steroid Production.
Briefly, 10 4 cells/well were seeded in 96-well plates in 200 μL CM supplemented with androstenedione (28 ng/mL) and treated with quercetin at concentrations previously described. Culture media were collected after incubation, frozen and stored at −20 • C until progesterone (P4) and 17β estradiol (E2) determination by validated RIAs [29]. P4 assay sensitivity and ED50 were 0.24 and 1 nM/L, respectively; E2 assay sensitivity and ED50 were 0.05 and 0.2 nM/L. The intra-and interassay coefficients of variation were less than 12% for both assays.

Nitric Oxide Production.
Nitric oxide (NO) was assessed by measuring nitrite levels in culture media by microplate method based on the formation of a chromophore after reaction with Greiss reagent, which was prepared fresh daily by mixing equal volumes of stock A (1% sulfanilamide, 5% phosphoric acid) and stock B (0.1% N-[naphtyl] ethylenediamine dhydrochloride). Briefly, 10 5 cells/200 μL CM were seeded in 96-well plates and treated with quercetin as already mentioned. After incubation with Greiss reagent, the absorbance was determined with a Spectra Shell Microplate Reader using a 540 nm against 620 nm filter. A calibration curve ranging from 25 to 0.39 μM was prepared by diluting sodium nitrite in culture medium.

Superoxide Anion Production. Superoxide anion (O 2
− ) generation was measured by the cell proliferation WST-1 test (Roche, Mannheim, Germany). Briefly, 10 5 cells/200 μL CM were seeded in 96-well plates and treated with quercetin as described above. Since evidence exists [30,31] that tetrazolium salts can be used as a reliable measure of intracellular O 2 − production, 20 μL of WST-1 were added to the cell during the last 4 hours of incubation. The absorbance was then determined using a Spectra Shell Microplate reader at 450 nm against 620 nm.

Hydrogen Peroxide Production.
Hydrogen peroxide (H 2 O 2 ) production was measured by an Amplex Red Hydrogen Peroxide Assay Kit (Molecular Probes, PoortGebouw, The Netherlands); the Amplex Red reagent reacts with H 2 O 2 to produce resorufin, an oxidation product. Briefly, 2 × 10 5 cells/200 μL CM were seeded in 96-well plates. After incubation with treatments, plates were centrifuged for 10 minutes, at 400 × g, then the supernatants were discarded and cells were lysed by adding cold Triton 1% in Tris-HCl (100 μL per well), incubating on ice for 30 minutes. The test was performed on cell lysates and read against a standard curve of H 2 O 2 ranging from 0.195 to 12.5 μM. A Spectra Shell microplate reader set to read 540 nm emission was used to quantify the reaction product.

Scavenging Enzyme Activity
2.5.1. Superoxide Dismutase Activity . Superoxide dismutase (SOD) activity was determined by an SOD Assay Kit (Dojindo Molecular Technologies, Japan). Briefly, 2 × 10 5 cells/200 μL CM were seeded in 96-well plates and, after treatments, the plates were centrifuged for 10 minutes at 400×g. The supernatants were discarded and cells were lysed by adding cold Triton 1% in Tris-HCl (100 μL/10 5 cells) and incubating on ice for 30 minutes. Cell lysates were tested without dilution, and a standard curve of SOD ranging from 0.156 to 20 U/mL was prepared. The colorimetric assay was performed measuring formazan produced by the reaction between a tetrazolium salt (WST-1) and a superoxide anion (O 2 − ), produced by the reaction of an exogenous xantine oxidase. The remaining O 2 − is an indirect hint of the endogenous SOD activity. The absorbance was determined with a Spectra Shell Microplate Reader reading at 450 nm against 620 nm.

Peroxidase
Activity. Peroxidase activity was measured by an Amplex Red Peroxidase Assay Kit (Molecular Probes, PoortGebouw, The Netherlands) based on the formation of an oxidation product (resorufin) derived from the reaction between H 2 O 2 given in excess and the Amplex Red reagent. Briefly, 10 5 cells/200 μL CM were seeded in 96-well plates and incubated in the conditions described above. After centrifugation for 10 minutes. at 400 × g, the supernatants were discarded, and cells were lysed by adding cold Triton 1% in TRIS HCl (100 μL/10 5 cells) and incubating on ice for 30 minutes. Cell lysates were used undiluted to perform the test. The absorbance was determined with a Spectra Shell Microplate Reader using a 540 nm filter and read against a standard curve of peroxidase ranging from 0.039 to 5 mU/mL.
Briefly, 2 × 10 5 cells/200 μL CM were seeded in 96-well plates and treated as already described. At the end, plates were centrifuged for 10 minutes at 400 × g, supernatants were discarded and cells were lysed by adding cold Triton 0.5% + PMSF in PBS (200 μL/well), incubating on ice for 30 minutes. The test was performed on 40 μL of cell lysates added to Fe 3+ TPTZ reagent and then incubated at 37 • C for 30 minutes. The absorbance of Fe 2+ TPTZ was determined by Spectra Shell Microplate Reader at 595 nm. The ferric reducing ability of cell lysates was calculated by plotting a standard curve of absorbance against FeSO 4− 7H 2 O standard solution.

VEGF Production.
The VEGF content in culture media was quantified by an ELISA (Quantikine, R & D System, Minneapolis, Mich, USA); this assay, which was developed for human VEGF detection, has been validated for pig VEGF [32]. Briefly, 10 6 cell/1 mL CM were seeded in 24-well plates and treated as described above. The assay sensitivity was 8.74 pg/mL, and the intra and interassay CVs were always less than 7%. The absorbance was determined with a Spectra Shell Microplate using a 450 nm filter.

Statistical Analysis.
Each experiment was repeated at least 4 times (6 replicates/treatment). Experimental data are presented as mean ± SEM; statistical differences were calculated with ANOVA using Statgraphics package (STSC Inc., Rockville, Md, USA). When significant differences were found, means were compared by Scheffè's F-test.

Granulosa Cell Proliferation.
BrdU incorporation assay test showed that basal swine granulosa cell proliferation was unaffected by both concentrations of quercetin.

Hydrogen Peroxide Production.
Colorimetric assay did not show any significant effect of quercetin on H 2 O 2 production by swine granulosa cell.

Scavenging Enzyme Activity
Superoxide Dismutase and Peroxidase Activity. Both SOD and peroxidase activities were unaffected by quercetin.

Discussion
Phytoestrogen quercetin is abundant in many plants used in animal nutrition [33,34]. In recent years, this flavonoid has attracted a lot of interest, due to its putative health promoting activities likely resulting from its antioxidant effects. Nevertheless, its overall biological impact remains controversial, mostly because of limited information about its bioavailability, endogenous dynamics, and the relative contribution of different types of conjugates in humans and animals. Quercetin appears to be effective in in vitro assays at concentrations ranging from 0.3 and 30 μg/mL, but plasma levels resulting from high oral doses are generally below those values [13]. This issue needs to be properly assessed in future studies, in order to get a better insight on the optimal consumption levels which could provide pharmacological significant concentrations in body fluids and tissues. In our experimental model, high quercetin concentration (30 μg/ml) has also been tested, since a process of accumulation of this flavonoid in some body tissues cannot be excluded [13].
Present study confirms that quercetin can modulate ovarian function, interfering with porcine granulosa cell steroidogenic activity [35]. The negative effect on P4 production could result from an inhibition of steroidogenic enzymes [36,37], since no detectable changes of cell proliferation have been observed. In a previous work [37], Journal of Biomedicine and Biotechnology it has documented that a suppressive action of quercetin on cytochrome P45, catalyzing the conversion of cholesterol to pregnenolone, represents a rate-limiting step in the steroidogenic pathway. Several recent studies [35,38,39] show that the phytoestrogen-induced decrease of P4 production in granulosa cells could result from an inhibition of 3βhydroxysteroid enzyme. On the contrary, Chen et al. [40] evidenced that quercetin increases StAR mRNA expression in MA-10 mouse tumor Leydig cells thus resulting in a stimulatory effect on steroidogenesis. This difference could be due to a cell type-dependent action of the flavonoid. As for E2 production, quercetin displayed a biphasic action: in fact, our data show that low dosages appear to increase E2 levels, while higher concentrations strongly decrease E2 production. An inhibitory effect of quercetin on aromatase activity in human granulosa-luteal cells has been documented [39], but the mechanisms through which this phytoestrogen could modulate the enzyme expression and activity remain elusive: these could involve either a binding with the estrogen receptors and/or a modulation of cell signalling pathways. On the basis of our data, it is possible to speculate that quercetin effects on E2 production by granulosa cells could be mediated by NO. In fact, this free radical seems to represent an autocrine regulator of granulosa cells E2 production [41], strongly inhibiting P450 aromatase activity [42]; the increase in estradiol secretion by cultured human granulosa cell [42] treated with NO synthase (NOS) inhibitors [43] would further strengthen this hypothesis. Interestingly, in agreement with our previous works [44,45], current data document that quercetin at low doses strongly inhibits NO levels, while at higher doses it stimulates NO release. These overall data would indicate that quercetin can regulate aromatase activity , and thus E2 release, by an interference with the NO/NOS system in swine granulosa cells. Our results also document that the increase of NO levels after treatment with 50 μg/mL of quercetin is paralled by a decrease of O 2 − levels: this observation would suggest a possible relationship between these two ROSs. Quercetin has been shown [46] to protect nitric oxide (NO) from the scavenging actions of O 2 − ; Vera et al. [47] reported that NO production negatively affects superoxide anion (O 2 − ) accumulation with a resulting formation of peroxynitrite (ONOO). Even if the antioxidant potential of quercetin has been demonstrated by different studies [48,49], pro-oxidative dose-dependent effects were also shown [7]. Present data indicate that quercetin does not affect neither H 2 O 2 levels nor SOD activity. Since antioxidant scavenger gene expression depends mainly on ROS levels [50], both the inhibitory effect on O 2 − generation as well as the ineffectiveness in modulating H 2 O 2 production could, respectively, explain the lack of effect on SOD and peroxidase enzyme activity. On the other hand, the higher quercetin concentration increased total nonenzymatic antioxidant ability thus suggesting that the antioxidant protection could result from an increased ability of several molecules to chelate transition metals [51].
Quercetin, at both concentrations tested, displayed a strong inhibitory effect on VEGF output thus suggesting its involvement in the modulation of angiogenesis. Since follicular development is strictly dependent on the angiogenic process [25], driven mainly by VEGF, this finding acquires particular relevance, since it is possible to speculate a possible negative influence of quercetin on ovarian physiology. In agreement with our findings, Zhong et al. [52] demonstrated that quercetin treatment significantly decreases VEGF secretion by myeloblastic leukemia cells NB4 in vitro. In addition, several studies document that flavonoids inhibit VEGFinduced endothelial cell functions and signalling pathways acting at molecular level [53][54][55][56]. Moreover, quercetin was found to inhibit several steps of angiogenesis, including proliferation, migration, and differentiation of endothelial cells [20,57]. In conclusion, the present study demonstrates that quercetin affects porcine granulosa cell function by interfering with steroidogenic activity and redox status as well as by inhibiting VEGF output; these data would suggest that this phytoestrogen represents a potential modulator of ovarian functions. Further studies are needed to better define the in vivo effects of quercetin on reproductive physiology.