|Home | About | Journals | Submit | Contact Us | Français|
We summarized the scientific literature published from 2007 to 2016 on the potential effects of bisphenol A (BPA) on female fertility. We focused on overall fertility outcomes (e.g., ability to become pregnant, number of offspring), organs that are important for female reproduction (i.e., oviduct, uterus, ovary, hypothalamus, and pituitary), and reproductive related processes (i.e., estrous cyclicity, implantation, and hormonal secretion). The reviewed literature indicates that BPA may be associated with infertility in women. Potential explanations for this association can be generated from experimental studies. Specifically, BPA may alter overall female reproductive capacity by affecting the morphology and function of the oviduct, uterus, ovary, and hypothalamus-pituitary-ovarian axis in animal models. Additionally, BPA may disrupt estrous cyclicity and implantation. Nevertheless, further studies are needed to better understand the exact mechanisms of action and to detect potential reproductive toxicity at earlier stages.
Female infertility is generally defined as the inability to get pregnant naturally and to deliver a live healthy newborn. According to the Center for Disease Control and Prevention (CDC; http://www.cdc.gov/nchs/nsfg/key_statistics/i.htm#infertility), 6.1% of married women were considered to be infertile between 2011 and 2013 in the USA alone. The percentage of infertile women can reach 30% world-wide (1). Infertility in women can be the result of various factors, including physical problems, endocrine problems, lifestyle habits, and environmental factors. Environmental factors such as exposure to chemicals with endocrine disrupting properties can mimic or block the endocrine activity of endogenous hormones and thus, adversely affect reproduction.
One of the most extensively studied endocrine disrupting chemicals is bisphenol A (BPA). BPA is incorporated in many daily used products as it is used by the manufacturers of polycarbonate plastics and epoxy resins. Despite the relatively short half-life of BPA (6–24 hours) (2), it was measured in various reproductive tissues (3), including ovarian follicular fluid, placenta, breast milk, and colostrum. Findings from previous publications suggest that BPA is a reproductive toxicant (4–6).
The current review focuses on the scientific evidence for BPA-induced fertility problems in females. We summarized the main findings of epidemiological and experimental studies that examined the potential effects of BPA on female fertility and that were published between 2007 and 2016. We included the morphological and mechanistic findings reported in the reviewed manuscripts. We focused on the reported outcomes of BPA exposure on overall: 1) fertility, 2) reproductive related processes including the ovarian cycle, and 3) reproductive tissues.
Pubmed (http://www.ncbi.nlm.nih.gov/pubmed) searches for the years 2007–2016 were conducted using the following key words: BPA, bisphenol A, fertility, female, reproduction, ovary, pregnancy, oviduct, ovulation, fertilization, uterus, implantation, hypothalamus, and pituitary. We focused on manuscripts published in 2007–2016 to expand upon previous review papers on the same topic (4, 5, 7–11). Additionally, references included in other review papers were examined for relevant information. We included manuscripts that dealt with fertility/infertility outcomes related to overall fertility, implantation, uterine morphology and function, estrous cyclicity, hypothalamus-pituitary, hormone levels (luteinizing hormone; LH, follicular stimulating hormone; FSH, and prolactin; PRL), oviduct, and ovary. We excluded manuscripts about topics that were out of the scope of this review paper or ones that will be reviewed by other authors in this special issue (e.g., sexual maturation/behavior, oocyte quality and maturation, ovarian steroidogenesis, pregnancy, miscarriage, endometriosis, polycystic ovarian syndrome, and uterine fibroids/leiomyoma).
BPA studies have used various study designs and included a wide range of doses. Based on the definitions in other studies, we considered a “low dose” of BPA as follows: a dose below the lowest observable adverse effect level (LOAEL) of 50 mg/kg/day in animal models (4, 5, 12, 13), 17.2 mg/l for aquatic animals (5, 14), 1 × 10−7 M for cell culture experiments (5, 15), and a dose in the range of typical (not occupational) human exposures for epidemiological studies (5, 16). The majority of the studies described in this review used doses that are within the category of “low dose”. Throughout the text of this review, we indicated if the doses were considered low or high based on the categories above. In the tables, the specific doses that were used in each study are described in detail. Lastly, similar to Peretz et al. (4), we defined exposure time during pregnancy as “in utero”; exposure after birth that ended before weaning as “neonatal”; and exposure any time after weaning as “postnatal” or adult exposure.
In recent years, several research groups have examined the effects of BPA on overall fertility. Epidemiological studies examined if BPA levels are higher in infertile women than in fertile women (Table 1). Findings from these studies indicate that infertile women have higher serum BPA levels compared to fertile women (17, 18). Further, studies conducted in women undergoing in vitro fertilization (IVF) treatments show that BPA levels (total or unconjugated BPA) were inversely associated with peak estradiol levels, number of oocytes retrieved, oocyte maturation, fertilization rates, and embryo quality (19–23). Thus, increased levels of BPA may decrease the success rate of IVF treatments. Nevertheless, these studies did not take into account potential modifying factors such as co-exposure to other chemicals and the location of sample collection as pointed out by Teeguarden et al (24, 25). Thus, additional studies are needed to fully understand the associations between BPA exposure and fertility outcomes in women.
Limited information is available on the potential molecular targets of BPA in infertile women, but Hanna et al. reported an association between higher serum levels of unconjugated BPA and decreased methylation within the TSP50 gene promoter in whole blood samples of women undergoing IVF treatments (26). However, the researchers did not provide any mechanistic explanations of these findings other than to indicate that TSP50 may be an oncogene based on previous research by other groups (27, 28). Interesting findings reported by Chavarro et al. suggest a potential modifying effect of soy food consumption on the inverse correlations between urinary total BPA concentrations and fertility treatment outcomes (29). Overall, these studies are suggestive for potential associations between BPA and infertility. However, additional studies are needed to determine possible cause and effect relationship and the mechanism of action of BPA-mediated effects on fertility in healthy women.
Not all epidemiological studies found an association between BPA exposure and fertility outcomes. Null associations were reported between urinary total BPA concentrations and impaired fecundity or time to pregnancy in generally healthy women (30–32). In another study, null associations between urinary total BPA concentrations and number of oocytes retrieved, embryo quality, and fertilization rates were reported in women undergoing IVF treatments (33). The differences in the results may be explained by differences in sample characteristics (i.e., generally healthy women without any reported infertility issues versus women undergoing IVF treatments) and by differences in sample size.
Studies utilizing animal models provide further insights on the effects of BPA exposure on female fertility (Table 2). In mice, Berger et al. reported that low dose BPA exposure of pregnant dams during the pre-implantation period significantly reduced the number of litters and litter size compared to controls (34). Further, in utero post-implantation low dose BPA exposure affected the fertility of the females in the subsequent generations (35, 36). Cabaton et al. performed a forced breeding study and found that low dose BPA-exposed females had fewer pregnancies and overall reduced cumulative number of pups compared to controls (37). Moore-Ambriz et al. examined the effects of BPA exposure in young adult mice on fertilization capacity later in adult life (38). The fertilization rate of BPA exposed females was reduced compared to controls (38). Further, impaired fertility was also reported in a study that examined the effects of in utero low dose BPA exposure in three subsequent generations of mice (35, 36). Specifically, F1 females that were gestationally exposed to BPA had reduced fertility, reduced litter size (36), and reduced ability to maintain pregnancy to term (i.e., reduced gestational index) compared to controls (35). Further, F2 females had a reduced gestational index compared to controls (35). In addition, F3 females exhibited reduced fertility and decreased ability to become pregnant compared to controls, indicating a potential transgenerational effect of BPA on female reproduction (35). In chickens, in ovo high dose BPA exposure reduced hatchability (39), whereas in fish, low dose BPA exposure increased the observed hatching rate (40).
In contrast, some experimental studies reported that BPA exposure does not affect fertility outcomes. Specifically, a few studies indicate that gestational low dose BPA exposure did not alter number of litters (41–44) or litter size (41–48) in mice, rats, and fish. Xi et al., also indicate that gestational BPA exposure at a dose of 50 mg/kg/day (i.e. LOAEL) did not alter litter size in mice (46). Similarly, Moore-Ambriz et al. reported that low dose BPA exposure did not affect the size of preovulatory follicles, the number of shed oocytes, and zygotes in adult mice that were exposed to BPA at a younger age (38). One of the reasons for differences between the reported results may be the age of the animals. Studies that examined reproductive capacity in older animals were more likely to observe a difference between the BPA treated females and controls.
In summary, several studies indicate that BPA levels may be higher among infertile women than fertile women and that BPA exposure may reduce fertility in animal models. However, further studies are needed to link findings from epidemiological studies and experimental studies.
To provide information about the potential mechanisms by which BPA impairs fertility, below we focus on reproductive organs that are targeted by BPA in a manner that could reduce fertility. Specifically, the sections below focus on BPA-induced abnormalities in reproductive organs that stem from changes in morphology, function, gene expression, and levels of proteins or hormones related to reproduction. We review the recent studies on the effects of BPA on the oviduct, uterus and implantation, estrous cyclicity, ovary, and hypothalamic-pituitary axis.
Following ovulation, the oocyte travels from the ovary through the oviduct to allow fertilization. Upon successful fertilization, the conceptus will continue to travel through the oviduct until it is settled in the uterus. Thus, a normal functioning oviduct is required for fertility. The available recent evidence on the effects of BPA exposure on the oviduct is extremely limited and is based only on experimental studies in mice (Table 3). Specifically, in utero low dose BPA exposure resulted in the appearance of progressive proliferative lesions in the oviduct and remnants of the Wolffian duct during adult life (49, 50). Further, studies indicate that in utero high dose BPA exposure delayed development and transport of the conceptus compared to controls (51, 52). Taken together, the current data indicate that gestational BPA exposure may affect both oviduct morphology and function; however, further studies are needed to confirm whether this is the case in women and to examine potential molecular mechanisms of BPA action on the oviduct.
Implantation is required for the establishment of pregnancy. During this stage, the blastocyst attaches to the uterine wall. Thus, exposures that interfere with implantation have the potential to impact fertility. The scientific evidence for a possible link between BPA exposure and impairments in implantation is based on one epidemiological study (Table 4) and several experimental studies (Table 5). Specifically, higher urinary total BPA levels were associated with increased implantation failure defined by a serum β-human chorionic gonadotropin test (β-hCG < 6 IU/L) conducted 17 days after egg retrieval in women undergoing IVF treatments (53).
Experimental studies examining the effects of BPA exposure at early gestational stages report a reduced number (34, 51, 54–57) or complete ablation of implantation sites (52) in mice and rats when compared to controls. These studies include both low (34, 54–57) and high (51) BPA doses. Further, a recent study by Li et al. (54) demonstrated that low dose BPA exposure reduced uterine levels of leukemia inhibitory factor (Lif), progesterone receptor (Pgr), heart and neural crest derivatives expressed transcript 2 (Hand2), and homeobox A10 (Hoxa10) compared to controls. These observed BPA mediated effects can impair fertility because these factors are part of the progesterone-mediated signaling pathway and are important in uterine receptivity and implantation.
Although the majority of studies indicate that BPA alters implantation, two experimental studies report no effect of low dose (58) or a relatively high dose (122 mg/kg/day) (59) of BPA on the number of implantation sites. The reasons for these discordant results are unclear, but it may be that the effects of BPA on implantation are ablated at high doses.
For proper blastocyst invasion, implantation, and successful pregnancy, the uterine endometrium transforms and reorganizes under the influence of estrogen and progesterone. Thus, exposures that interfere with uterine function have the potential to adversely impact fertility. Based on our search criteria, we did not locate epidemiological studies that focus on BPA exposures and associations with uterine outcomes. However, several in vivo and in vitro experimental studies have examined the effects of BPA exposure on the uterus or uterine cells (Tables 6 and and77).
In mice, in utero low dose BPA exposure increased uterine anomalies in the luminal epithelium and glands (60) and caused uterine hyperplasia, stromal polyps, and retention of remnants of the Wolffian duct in the adult offspring compared to controls (49, 50). Further, in utero high dose BPA exposure reduced uterine weight in the second generation of pups compared to controls (61). Neonatal high dose BPA exposure (single dose, 100 mg/kg) reduced uterine weight in young adult mice compared to controls (62), and neonatal low dose BPA exposure decreased endometrial proliferation in adult ovariectomized rats compared to controls (63). In young adult rats, low dose BPA exposure from gestation day 6 until weaning increased the thickness of the uterine epithelia and stroma compared to controls (64). In adult mice, dietary supplementation with low dose BPA resulted in clinical signs that are typical of pyometra (44). Lastly, in hens, in ovo high dose BPA exposure resulted in abnormal uterine morphology compared to controls (39). Overall, the results from in vivo studies are suggestive for impaired morphology of the uterus following early life stage BPA exposures at both low and high doses.
Some of the molecular factors in the uterus that were altered following in vivo BPA exposure include members of the Hoxa family, vascular endothelial growth factor (Vegf), estrogen receptor alpha and beta (Esr1 and Esr2), and Pgr (57, 63, 65–70). These factors are important for endometrial proliferation and receptivity. In contrast, in utero high dose BPA exposure did not affect expression levels of coding complement component 3 (C3), Pgr, calbindin D9K (S100g), and Vegfa in the adult offspring of rats (71); hence, it is unclear whether they are primary targets for BPA-induced uterine toxicity.
Findings from in vitro studies utilizing various human cell lines indicate that low dose BPA exposure decreased endothelial cell proliferation (9, 72) and increased decidualized stromal cell proliferation (73) compared to controls. Similarly, high dose BPA exposure decreased endothelial cell proliferation compared to controls (74). Some of the potential mechanisms through which BPA may affect cell proliferation in the uterus include alterations in insulin-like growth factor binding protein 1 (IGFBP1), macrophage migration inhibitory factor (MIF), HOXA10, and left right determination factor 1 (LEFTY), steroidogenic receptors (e.g., ESR1, ESR2, PGR), and enzymes (e.g., cytochrome P450, family 11, subfamily a, polypeptide 1; CYP11A1, hydroxysteroid (17-beta) dehydrogenase 1; HSD17B1, hydroxysteroid (17-beta) dehydrogenase 2; HSD17B2), or other hormones (e.g., PRL, LH) (9, 61, 62, 64, 66, 69, 73, 75–77). Further, Nacif et al. (77) performed a microarray on Ishikawa cells that were cultured with a range of low and high BPA doses (1nM–100μM) and found that multiple molecular pathways (e.g., cell organization and biogenesis, proliferation, and intracellular transport) were altered in response to BPA compared to controls.
In contrast, two studies on human cell lines cultured with low and high BPA doses found no effect on proliferation of primary stromal endometrial cells (76, 78). Differences in cell viability or proliferation may be due to the study design and experimental model. For example, differences in the source of the cells (carcinogenic tissue versus normal) and differences in the experimental cell lines (primary versus established/immortal cell line lines such as Ishikawa). Nevertheless, the majority of the in vitro studies support the observations reported in the above-mentioned in vivo studies.
Lastly, at the end of pregnancy, the uterus needs to contract to induce labor. Uterine contractions are under the control of endogenous hormones such as estradiol, progesterone, oxytocin, and prostaglandins (79). The effects of BPA on uterine contractility were investigated in one study in rats (75). Findings from this study suggest that BPA exposure decreased uterine contractility and altered transcript and protein levels of contraction-associated factors (75). Specifically, high dose BPA exposure increased oxytocin and oxytocin receptor, and decreased prostaglandin (PG)-F2α receptor compared to controls (75). Overall, the current literature suggests that BPA exposure selectively affects uterine cell proliferation and function, depending on the study model. These effects of BPA on uterine function could lead to adverse effects on fertility.
Estrous cyclicity is crucial for ovulation and the preparation of the uterus for potential implantation. Hence, chemical exposures that disrupt estrous cyclicity can impair fertility. Multiple experimental studies examined the effects of BPA exposure on estrous cyclicity (Table 8). Early neonatal low (80–82) and high (62) dose BPA exposure caused increased or decreased days in estrus and overall altered cycles in adult mice or rats (70–90 days) compared to controls. In contrast, low dose BPA exposure at earlier time points (51–54 days old) had no effect on estrous cyclicity (38). In rats, in utero low dose (64) and high dose (83, 84) BPA exposure resulted in irregular estrous cycles in the offspring compared to controls. Moreover, studies published by Wang et al. (36) and Ziv-Gal et al. (35) examined the effects of in utero exposure of low dose BPA on estrous cyclicity in subsequent generations of mice. Interestingly, BPA-induced altered cyclicity was observed in both the F1 and F3 generations, but not in the F2 generation compared to controls. In contrast, other studies reported no effect of in utero low dose BPA exposure or 50 mg/kg/day (i.e. LOAEL) on estrous cyclicity of rats and mice offspring (41, 46, 60). Differences in study design and timing of evaluation of estrous cyclicity may explain the disagreement between the results. Overall, neonatal BPA exposure may affect estrous cyclicity in older animals. However, the evidence regarding the effects of in utero BPA exposure on estrous cyclicity is inconclusive. Further studies that examine the effects of in utero BPA exposure on estrous cyclicity and that encompass multiple generations are needed to fully understand the effects of BPA on estrous cyclicity.
The ovary is required for normal production of ova for fertilization and for production of sex steroid hormones that regulate estrous cyclicity and fertility. Thus, BPA exposures that target the ovary can interfere with fertility. One epidemiological study examined the associations between BPA levels and ovarian volume and mature follicle counts (Table 9) (85). Results from this study indicate that urinary BPA exposure was negatively correlated with antral follicle counts in women undergoing IVF treatments (85).
Findings from experimental studies indicate that in utero or neonatal low and high dose BPA exposures resulted in abnormal ovarian morphology and histology compared to controls (Table 10). Specifically, BPA exposure increased the number of multi-oocyte follicles (86), inhibited germ cell nest breakdown (36, 45, 87), decreased the number of primordial follicles (45, 87), increased apoptotic oocytes (87), and increased primordial follicular recruitment (36, 88). It also affected follicle type distribution by reducing the number of antral follicles and increasing the numbers of primary and secondary follicles (89).
Molecular analysis revealed that low dose BPA exposure affected levels of genes related to apoptosis. Specifically, it BPA increased levels of B cell leukemia/lymphoma 2 (Bcl2), BCL2-like 1 (Bcl2l1) (36, 87). BPA exposure also decreased levels of BCL2-antagonist/killer 1 (Bak1), tumor necrosis factor receptor superfamily, member 11b (Tnfrsf11b), tumor necrosis factor receptor superfamily, member 1a (Tnfrsf1a), tumor necrosis factor (ligand) superfamily, member 12 (Tnfsf12), and lymphotoxin B receptor (Ltbr) (36, 87). Additionally, BPA exposure altered BCL2-associated X protein (Bax) levels by either increasing (87) or decreasing (36) its levels. Differences in the effects of BPA on Bax can result from differences in study designs (e.g., ovarian transplant versus excised neonatal ovaries).
Further, BPA exposure decreased expression of factors that control folliculogenesis such as NOBOX oogenesis homeobox (Nobox) (mRNA and protein), LIM homeobox protein 8 (Lhx8) (mRNA and protein), spermatogenesis and oogenesis specific basic helix-loop-helix 2 (Sohlh2), stimulated by retinoic acid gene 8 (Stra8), DNA meiotic recombinase 1 (Dmc1), REC8 meiotic recombination protein (Rec8), synaptonemal complex protein 3 (Scp3), and folliculogenesis specific basic helix-loop-helix (Figlα) (45, 87). In addition, BPA exposure prevented DNA methylation in CpG sites of Lhx8, indicating that BPA may impair normal processes of folliculogenesis (87) and ovarian dynamics.
Similar effects of exposure to both low and high doses of BPA during neonatal life on the ovary were observed at older ages/post weaning. Specifically, researchers reported findings such as BPA-induced multinucleated and hemorrhagic tissue (84), multi-oocyte follicles (90), altered follicle type distribution or numbers (41, 83, 89–95), reduced ovarian weight (41, 62), and ovarian cysts (49) compared to controls. Molecular analysis revealed that high dose BPA exposure decreased the expression of Figla and oocyte-specific histone H1 variant (H1f00), and increased the levels of anti-Müllerian hormone (Amh) genes (92). Additionally, Lee et al. reported that low dose BPA increased apoptosis in ovarian follicles that was coupled with increased levels of the apoptotic protein caspase-3 (81). Hence, in rodents, BPA affects ovarian development and dynamics via molecular pathways that involve apoptosis, folliculogenesis, and oocyte specific factors.
Similarly, in fish, low dose BPA exposure increased ovarian weight, increased levels of hydrogen peroxide, and decreased glutathione levels compared to controls, indicating that BPA may alter the oxidative stress mechanism in the ovary (96). Another study in fish found that low dose BPA exposure increased ovarian weight, atretic follicles, perinuclear oocytes, and expression of factors involved in folliculogenesis (Gdf9 and Bmp15) compared to controls (97).
Other studies in mice found no effect on follicle type distribution in the adult ovary following low dose BPA exposure (38, 46); however, it is possible that BPA did not affect ovarian follicle distribution because of the timing of BPA exposure and differences in study design. Further, some of the effects of BPA exposure on the ovary do not persist in subsequent generations. Wang et al. (36) reported that in utero low dose BPA exposure inhibited germ cell nest breakdown in the F1 generation of mice compared to controls; however, these changes were not observed in the subsequent generations examined by Berger et al. (98). Further, BPA exposure caused several generation-specific differences in gene expression, but not all were transgenerational (i.e., genes related to oxidative stress, autophagy, and apoptosis) (98). Interestingly, BPA-induced changes in steroidogenic genes and Esr1, androgen receptor (Ar), and insulin-like growth factor (Igf) family genes were suggested to be carried transgenerationally (98). It is possible that the changes in the genes related to oxidative stress and apoptosis are activated in an acute manner and thus, effects on these factors were not carried over to the subsequent generations. In contrast, steroidogenic factors are crucial to the function of the ovary (as an endocrine organ) and thus, some of BPA effects on these factors were carried over to the subsequent generations. Overall, these experiments provide strong evidence that BPA acts via mechanisms related to apoptosis, folliculogenesis, and oxidative status.
In mice, in vitro studies of isolated neonatal ovaries indicate that high dose BPA exposure inhibited germ cell nest breakdown and accelerated primordial follicle recruitment compared to controls (99, 100), similar to some of the observations in in vivo studies. Specifically, BPA exposure decreased levels antigen KI-67 (Ki67), tumor necrosis factor receptor superfamily member 6 (Fas), and caspases (Casp3 and 8) (99, 100). Further, BPA increased levels of Bcl2 and factors related to the phosphatidylinositol 3-kinase/thymoma viral proto-oncogene (PI3K/Akt) signaling pathway (99, 100). In sheep fetal ovaries, low dose BPA exposure resulted in an age-dependent increase in expression of steroidogenic genes, mammalian target of rapamycin (mTor), peroxisome proliferator-activated receptor (Pparα), and Igf1r compared to controls (101). The overall findings suggest that BPA may act via mechanisms that are related to follicle dynamics and apoptosis. Further, studies suggest that BPA exposure may act via mechanisms involving altered miRNA levels (101). BPA downregulated miR-137 that may decrease sex steroid hormone synthesis (102) and miR-765 that may be associated with premature ovarian failure (103, 104). Additional findings included variable levels of miRNAs related to insulin signaling, without changing levels of miRNA processing enzymes (101).
In vitro studies of isolated mouse ovarian follicles indicate that high dose BPA exposure selectively inhibited antral follicular growth (105–108), but increased preantral follicle growth (109) compared to controls. Molecular analysis revealed that BPA exposure affected the expression of genes related to the cell cycle, apoptosis, and steroidogenesis (105–108). Specifically, BPA exposure increased Bcl2, cyclin-dependent kinase 4 (Cdk4), cyclin E1 (Ccne1), transformation related protein 53 (Trp53), Bax, and downregulated cyclin D2 (Ccnd2). In short-term cultures (up to 24 hours) of Chinese hamster ovarian cells, high doses of BPA selectively increased cell viability, increased cytotoxicity, and induced DNA damage and the appearance of micronuclei (110). In short-term cultures (3–48 hours) of human ovarian cells, low dose BPA increased expression of VEGF-R2 (111). In short-term cultures (48 hours) of human granulosa-lutein cells, high dose BPA inhibited cell proliferation and decreased levels of IGF-1, aromatase, GATA binding protein 4 (GATA4), steroidogenic factor-1 (SF-1), and PPARγ (112). In contrast, in short-term porcine granulosa cell cultures (48 hours), high dose BPA did not affect cell proliferation or expression of oxidative stress genes (113). Similar concentrations of BPA over longer culture times (72 hours) decreased viability and disrupted matrix metallopeptidase 9 (MMP-9) secretion in human granulosa cells (114). It is plausible that some of BPA-mediated effects can be detected only at the end of longer culture times, or in a species specific manner.
Overall, current studies indicate that BPA affects the ovary, mainly during the ovarian developmental window as well as in early neonatal life via multiple pathways that include cell cycle, apoptosis, oxidative stress, and proliferation. More epidemiological studies are warranted to better understand the specific associations of BPA exposure and ovarian outcomes in women. Further, more experimental studies are warranted to better understand the specific mechanisms of action and the specific effects of low and high doses of BPA on the ovary.
Overall, reproductive function is dependent on the hypothalamic-pituitary-ovarian axis. Following sexual maturation, coordinated feedback loops along the hypothalamic-pituitary-ovarian axis control the ability of the mammalian female to ovulate and to prepare the reproductive organs to support potential pregnancy. In the hypothalamus, sex steroid hormones (estradiol and progesterone) activate the kisspeptin neurons that in turn, relay the secretion of gonadotrophic releasing hormone (GnRH). GnRH stimulates the anterior pituitary to secrete gonadotrophic hormones (FSH and LH). FSH and LH act on the ovary to support folliculogenesis. Increased levels of ovarian sex steroid hormones feed-back to the hypothalamic kisspeptin neurons to induce the LH surge that is needed for ovulation. Therefore, any alteration in proper levels/function of the hypothalamic-pituitary axis including the kisspeptin neurons can alter female fertility. The sections below describe the current data on the effects of BPA on the hypothalamus (Table 11), pituitary (Table 11), and gonadotrophic hormones (Tables 12 and and1313).
Few experimental studies have examined the effects of BPA on the hypothalamus, and findings vary between the studies. For example, neonatal BPA exposure increased (40, 46) or decreased levels of Kiss1 (115–117) in fish, mice, and rats. Differences in species and age of the animals in which observations were made can partially explain these opposite results rather than the doses that were used (i.e. low or high). Neonatal BPA exposure also decreased levels of gonadotropin releasing hormone (Gnrh) (40, 118), Esr1, and Esr2 (82, 117, 119, 120) in sheep and rats compared to controls. Interestingly, Wang et al. (118) reported that the effects of low dose BPA on Gnrh can be ablated by pretreatment with a specific blocker of the receptor of KISS1. These findings imply that the effects of BPA exposure are mediated via the kisspeptin signaling pathway. Future studies that utilize molecular techniques including kisspeptin knock-out mice can aid in further elucidating BPA-induced effects in the hypothalamus.
Similar to the prominent effects of BPA on neonatal animals, in utero BPA exposure (low doses and LOAEL) that continued until the age of weaning increased levels of Kiss1 and GnRH on postnatal day 50 compared to controls in mice (46). Interestingly, the same exposure levels at a later age (postnatal days 21–49) did not affect hypothalamic gene expression (46), indicating that the timing of exposure has a large influence on the outcome. Other researchers reported that in utero low dose exposure to BPA disrupted methylation patterns in the hypothalamus that can affect the expression of Esr1 (121). In contrast, low dose BPA exposure did not affect the density of tyrosine hydroxylase cells that are involved in sexual dimorphism of the brain (122). Similarly, Abi Salloum et al. reported that sheep that were exposed to low dose BPA in utero still maintained a well-functioning neuroendocrine system in response to steroid feedback at adulthood (123). Overall, the effects of BPA exposure on the hypothalamus are likely to be dependent on the timing of exposure and type of animal model that is used. Potential factors that may be mediating BPA effects are kisspeptin and GnRH.
Limited data on the association between BPA exposure and pituitary outcomes are available from human studies (Table 12). Miao et al. reported a positive association between creatinine adjusted urine BPA levels and PRL, and a negative association with FSH levels in women exposed to BPA in their work place (124). In contrast, Souter et al. found no association between specific-gravity adjusted BPA levels and day-3 FSH levels in women undergoing IVF treatments (85). The differences between the two cohorts may explain the inconsistent results.
Few experimental studies have examined the effects of BPA exposure on the pituitary (Table 13). The majority of the studies examined the effects of low dose BPA, whereas only two studies included high doses as well (46, 125). In utero low dose BPA exposure increased proliferation of pituitary cells, and increased gonadotroph cell number (LHβ, FSHβ positive) compared to controls in the pituitaries of mice (125). In the same study, both BPA doses (0.5 and 50 μg/kg/day) decreased gonadotropin releasing hormone receptor (Gnrhr) levels, whereas 0.5 μg/kg/day of BPA increased the levels of Lhβ and Fshβ, and 50 μg/kg/day of BPA decreased the levels of Lhβ, Fshβ, and Nr5a1 compared to controls (125). Further, longer BPA exposure (25 and 50 mg/kg/day) from gestation through weaning increased Fsh levels in mice compared to controls (46). These findings indicate that BPA exposure can affect the pituitary, but it is likely to be dependent on the timing of exposure and dose.
Several of the BPA-induced changes in pituitary gene expression are accompanied by altered serum levels of the pituitary hormones. Specifically, low and high dose BPA exposure decreased serum FSH and LH levels in adult rats and fish compared to controls (80, 126, 127). Low dose BPA exposure also increased serum LH levels compared to controls in mice and rats (81, 89, 118). Lastly, early neonatal low dose BPA exposure resulted in a dampened LH surge compared to controls in adult rats (82). Differences in study design including BPA doses may contribute to the variability in the results. Nevertheless, additional studies are needed to examine if the effects on FSH, LH, and LH surge can explain some of the effects on ovarian function and estrus cyclicity.
In contrast to the effects of BPA described above, a few studies reported that BPA exposure does not affect serum levels of FSH (38, 46, 80, 81, 89) or LH (38, 118, 126) in mice and rats. Similarly, one study found that low dose BPA exposure did not affect the LH surge in adult sheep (95). Overall, the majority of the current studies suggest that BPA exposure affects the function of the anterior pituitary. However, the scientific evidence needs to be supported with additional studies. The secretion of LH and FSH may need to be measured over several time points to delineate the mechanisms through which BPA selectively targets impairs its function.
The current literature fairly consistently shows the following:
Despite the number of studies that have examined potential BPA effects on female fertility in the past years, it is still difficult to layout the exact mechanism of BPA action. BPA is a model endocrine disrupting chemical with a complicated mechanism of action that is yet to be fully elucidated. BPA effects are highly dependent on various factors in the study design such as timing of exposure, species, dose, route of exposure, and mode of quantification/assessment. Moreover, other modifying factors including co-exposures, study location/setting (e.g. hospital), and study sample (e.g. potentially unhealthy/previously exposed participants) should also be taken into consideration when possible as proposed by Teeguarden et al., (24, 25).
In summary, further studies are needed to better understand the mechanisms of action of BPA on female fertility. These studies will need to include some integrated endpoints to assure reproducibility of previous findings while taking into account relevance of the doses to human exposure, the ability of BPA to bind certain receptors, and study design/setting. Because female fertility relies on several organs and feedback loops, studies that utilize in vivo methods should aim for a multi-organ/disciplinary approach.
This work is supported by NIH P01 ES022848 and EPA RD-83459301.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.