Although long considered weakly estrogenic, the specific mechanisms by which BPA interacts with molecular and cellular targets within the hypothalamus and elsewhere are not clearly established (vom Saal et al., 2007). Classically, BPA is thought to disrupt genomic pathways mediated by the two primary forms of the nuclear estrogen receptor (ER); ER alpha (ERα) and ER beta (ERβ). Compared to estradiol, BPA has a binding affinity approximately 10,000–100,000 fold weaker for both ER isoforms (Andersen et al., 1999, Barkhem et al., 1998, Blair et al., 2000, Gould et al., 1998) but appears to bind each with relatively equal affinity (Kuiper et al., 1998) indicating that it has the potential to interact with either.

It is well established that estrogen is masculinizing during perinatal development in the murine brain, and that exposure to even low levels of estrogen or xenoestrogens during this time can permanently alter neuroendocrine pathways critical for mediating steroid negative feedback, gonadotropin release, energy homeostasis and sexual behavior (Amateau et al., 2004, Bader et al., 2011, Faulds et al., 2011, Gore, 2008, Simerly, 2002). We have recently shown that the expression of ERα and ERβ is sexually dimorphic within numerous subregions of the neonatal rat hypothalamus (Cao and Patisaul, 2011). For example, within the AVPV, ERβ expression is higher in males on the day of birth while ERα expression is higher in females, suggesting that the two isoforms may play different functional roles in the sexual differentiation of this region. This sex difference is transient, and eliminated within 48 hours, suggesting that if BPA exposure during only the first two days of life alters expression levels, estrogen sensitivity during this critical period of sexual differentiation may be impacted. Here we tested the hypothesis that neonatal BPA exposure can affect the sex specific hypothalamic expression of ERα and ERβ across postnatal development, focusing specifically on the AVPV, ARC and two other subregions (the medial preoptic area (MPOA) and the ventrolateral division of the ventromedial nucleus (VMNvl)) known to be essential for the neuroendocrine control of gonadotropin release and other aspects of female reproductive physiology, metabolic regulation and behavior (Nance, 1976, Pfaff and Keiner, 1973, Pfaff et al., 1994, Simerly, 2002). Although previous studies have explored similar hypotheses in a variety of species, they primarily looked at adult expression in only a single sex and either employed a different exposure window, (Ceccarelli et al., 2007), focused on only one hypothalamic area (Mahoney and Padmanabhan, 2010, Monje et al., 2007), or employed techniques with less anatomical resolution such as QT-PCR (Khurana et al., 2000, Monje, Varayoud, 2007). Thus, the present studies build on this prior work by (1) characterizing the specific hypothalamic subregions vulnerable to disruption in both sexes (2) identifying this disruption early in postnatal development and (3) more carefully identify the critical window of exposure for ER disruption by BPA. Collectively, these studies test the hypothesis that perturbation of ER expression during the neonatal period of hypothalamic sexual differentiation could be a mechanism by which BPA induces a wide range of effects, some of which may not emerge until adulthood.

We also sought to determine if BPA exposure can disrupt the expression of Kiss1, a gene which codes for the kisspeptin family of peptides. Kisspeptin is now recognized to be the primary “gatekeeper” of gonadotropin release and essential for initiating pubertal onset (Oakley et al., 2009, Pineda et al., 2010). It also plays a role in energy homeostasis and is a putative effector of leptin actions on gonadotropin releasing (GnRH) neurons (Castellano et al., 2010). There are two populations of Kiss1 neurons in the murine brain, an anterior group which extends through the AVPV into the MPOA (a region now collectively referred to as the rostral periventricular area of the third ventricle (RP3V)) and a mediobasal hypothalamic group confined to the ARC. In adults, the RP3V population is sexually dimorphic, with females having higher expression levels than males (Kauffman et al., 2007), while the ARC population is not (Losa et al., 2010). These observations suggest that the anterior population participates in steroid positive feedback on gonadotropin release, while the posterior population plays a role in steroid negative feedback (Oakley, Clifton, 2009, Pineda, Garcia-Galiano, 2010). Nearly all Kiss1 neurons in the RP3V and ARC co-express ERα, and a subset of RP3V Kiss1 neurons co-express ERβ (Smith et al., 2005, Smith et al., 2006) suggesting that disruption of ER expression within this neuronal phenotype could affect its sex specific ontogeny and function. Gestational and/or postnatal BPA exposure has been shown to alter Kiss1 mRNA expression within whole hypothalamic micropunches (Navarro et al., 2009, Xi et al., 2010), and RP3V Kiss1-immunoreactivity (Bai et al., 2011, Panzica et al., 2009) during peripuberty and adulthood. Although RP3V Kiss1 expression is not apparent in either sex until approximately the second week of life (Cao and Patisaul, 2011, Clarkson et al., 2009) we have shown that neonatal exposure to estrogen or an ERα selective agonist (Bateman and Patisaul, 2008, Patisaul, Todd, 2009) can defeminize kisspeptin signaling pathways in the adult RP3V, suggesting that the first few days of life may be a critical window of BPA vulnerability, and that this disruption may be apparent prior to puberty. In contrast to the RP3V, Kiss1 mRNA in the ARC is robustly expressed from PND 0 through PND 19, with higher levels in females than males within the first week of life (Cao and Patisaul, 2011). This indicates that Kiss1 expression may be influenced by the neonatal estrogen and vulnerable to endocrine disruption by a chemical like BPA. Here we examined whether neonatal BPA exposure affects the sexual dimorphic ontogeny of Kiss1 mRNA expression in the RP3V and ARC.

Neonatal BPA exposure disrupted gene expression throughout the hypothalamus but in a dose, temporal, and region specific manner that did not consistently mirror that of EB exposure in either sex. As anticipated, EB exposure completely masculinized ER expression across the female hypothalamus except for RP3V ERβ expression on PND 4, which was largely unaffected by EB. By contrast, in the AVPV, ERα levels were markedly elevated in females following BPA exposure on PND 4, then fell to male-typical levels by PND 10. The drop in ERβ expression by PND 10 was even more robust, resulting in no detectable signal at either dose. In the VMNvl, however, ER levels were relatively unaltered by BPA but completely masculinized by EB. These observations are commensurate with prior studies by us and others showing no impact of neonatal BPA exposure on the number of ERα-immunoreactive cells in the adult female VMNvl (Adewale et al., 2011), or in ERα and ERβ expression in juvenile rats (Ramos et al., 2003). Similarly, EB exposure resulted in male-typical ERα and ERβ levels in the PND 4 and 10 female ARC but BPA had no significant effects on ER expression in this region. These divergent effects of BPA suggest that it alters hypothalamic organization by mechanisms other than by simply acting as an estrogen mimic, and demonstrate that some brain regions are more sensitive to disruption than others.

Our results are consistent with a prior, RT-PCR study, which also found a pronounced sex difference in ERα expression on PND 8 (Monje, Varayoud, 2007) and that sc injection of 50 µg/kg BPA every 48 hours from PND 1 to 7 abrogates PND 8 female ERα expression to male levels while sc injection of 20 mg/kg produces a modest increase. Collectively, the data suggest that acute administration of high dose BPA initially upregulates ERα throughout the POA of both sexes but this expression pattern changes over time once exposure ceases, declining in the AVPV and MPOA of the HBPA females, but only in the AVPV of the LBPA females. Interestingly, by adulthood, ERα levels are reportedly higher in the AVPV of females neonatally exposed to 50 µg/kg or 20 mg/kg BPA compared to same sex controls (Monje et al., 2010) but not the MPOA of females neonatally exposed to 50 µg/kg or 50 mg/kg (Adewale, Todd, 2011). Moreover, peripubertal BPA exposure increases female ER immunoreactivity in the MPOA and VMN on PND 37, but this effect is lost by PND 90 (Ceccarelli, Della Seta, 2007). Temporal changes in estrogen sensitive gene expression have also been reported for the selective estrogen receptor Tamoxifen (Patisaul et al., 2003) suggesting that this type of long term regulatory disruption may not be atypical for xenoestrogenic compounds. The mechanisms underlying this effect remain to be determined but altered expression of critical and cell specific ER coactivators or corepressors is one possibility.

The complicated biochemical character of BPA was also demonstrated by the observation that it can augment ERα expression but depress or eliminate ERβ expression within the same hypothalamic region. The mechanisms by which BPA, or even endogenous estrogen, can differentially regulate the two ER isoforms remains poorly characterized. One possibility is that epigenetic modification by DNA methylation (Kurian et al., 2010) and/or histone deacetylation (Matsuda et al., 2011) may alter the status of each ER promoter independently thus accounting for the different expression levels of the two ERs. It is now clear that methylation status is dynamic across development, and may play an important role in the mediation of hormone dependent neonatal brain organization (Nugent et al., 2011). Importantly, BPA has previously been shown to be capable of modifying DNA methylation patterns (Dolinoy et al., 2007) suggesting that it at least has the potential to affect ER expression through epigenetic mechanisms.

The specific functional roles ERα and ERβ play in the sex specific organization of the hypothalamus have not been fully characterized but it has been hypothesized that ERα is critical for the organization of reproductive neuroendocrine pathways while ERβ may facilitate the emergence of sex appropriate behavior (Rissman, 2008). Thus, the diminished ERβ levels observed in the anterior hypothalamus of the BPA exposed males may be a potential mechanism by which perinatal BPA exposure disrupts sexual and sociosexual behaviors (Farabollini et al., 2002, Farabollini et al., 1999, Patisaul and Bateman, 2008). Moreover, previous research has established that RP3V neurons expressing tyrosine hydroxylase (TH), a marker for dopaminergic neurons, are co-localized with ERα and, to a lesser degree, ERβ and that this co-localization is sexually dimorphic pre-weaning (Orikasa et al., 2002, Patisaul, Fortino, 2006). Both the sex specific level of TH and the co-localization of TH and ERα have been shown to be altered by neonatal BPA exposure in the murine brain (Patisaul, Fortino, 2006, Rubin, Lenkowski, 2006). In mice, this change was associated with the loss of sexually dimorphic locomotor behavior in the open field test. These observations further support the hypothesis that disruption of ER expression in the developing hypothalamus may impact the emergence of sexually dimorphic physiology and behavior. More definitive links, however, between disruption of early life gene expression and adult physiology are needed to confirm the biological significance of the findings reported here.

Kiss1 expression is also sexually dimorphic (Kauffman, 2009, Kauffman, Gottsch, 2007) and tightly coordinated by estrogen (Smith, Cunningham, 2005). Nearly all Kiss1 neurons, in both the anterior and mediobasal hypothalamus, co-express ERα, and more than 20% of RP3V and 10% of ARC Kiss1 neurons co-express ERβ (Smith, Popa, 2006). Although neonatal EB exposure decreased Kiss1 expression in the female ARC to male-typical levels, BPA had no effect in either sex, with the exception of a slight increase in the HBPA males. Expression in the RP3V, however, was disrupted in both sexes suggesting that this population may be more vulnerable to endocrine disruption. This postulate is supported by our prior work showing that the ontogeny of Kiss1 signaling pathways can be disrupted in females by neonatal administration of estradiol benzoate (EB) or the phytoestrogen genistein (Bateman and Patisaul, 2008, Losa, Todd, 2010). The results from the present study further reveal that both doses of BPA were sufficient to eliminate RP3V Kiss1 expression in the PND 10 females, and elevate it in the PND 10 males. Female expression appears to ultimately recover, however, as we have not found adult female RP3V levels of Kiss1 immunoreactivity to be significantly affected by either dose of BPA (Patisaul, Todd, 2009). Moreover, the capacity of GnRH neurons in ovariectomized females to respond to hormone priming does not appear to be affected by neonatal BPA exposure (Adewale et al., 2009), indicating that the organization of the neuroendocrine pathways required to mediate this response appear to be functionally intact. Numerous studies have shown, however, that developmental exposure to BPA results in irregular or absent estrous cycles (Adewale, Jefferson, 2009, Cabaton, Wadia, 2011, vom Saal, Akingbemi, 2007) and compromised fertility (Cabaton, Wadia, 2011) suggesting that HPG regulation may be altered in gonadally intact animals that are not primed with exogenous hormones. This may result from the altered distribution of ER within Kiss1 neurons. Recent studies have revealed that, within the RP3V, the two isoforms play functionally different roles in adult females as antagonism of ERα blocks LH responses to kisspeptin, thereby eliminating the preovulatory LH surge, while antagonism of ERβ fails to block the LH surge but instead augments acute LH responses to kisspeptin (Roa et al., 2008). BPA significantly reduced, and completely eliminated, RP3V ERα and ERβ expression respectively by PND 10, suggesting that ER expression within Kiss1 neurons is decreased or absent. Future studies will be needed to confirm this possibility and establish how long the effect persists.

The present study provides evidence that the sex specific ontogeny of hypothalamic pathways important for reproductive physiology, behavior and energy balance may be disrupted by BPA exposure. Uncovering the potential mechanisms by which BPA affects neuroendocrine development is important when evaluating whether or not effects observed in rodents can be extrapolated to humans. The possible health consequences of BPA exposure remain controversial (Beronius, Ruden, 2010, Goodman et al., 2009, Vandenberg, Maffini, 2009). Importantly, the health effects of low dose oral exposure remain the subject of considerable interest because human exposure is presumably low but constant and from a variety of sources including food, beverages, the handling of paper receipts, and dust (Biedermann, Tschudin, 2010, Lakind and Naiman, 2010, Vandenberg, Hauser, 2007). Although numerous studies have reported health effects in rodents consistent with disrupted neuroendocrine function including altered pubertal timing, irregular or absent estrous cycles and reduced fertility (Adewale, Jefferson, 2009, Cabaton, Wadia, 2011, Howdeshell, Hotchkiss, 1999, Vandenberg, Maffini, 2009), others have found no effects at all (Howdeshell et al., 2008, Ryan et al., 2010, Tyl et al., 2008, Tyl et al., 2002). Why some studies find significant effects while others do not remains the subject of intense debate but issues like dose, timing of exposure, route of administration, strain, housing conditions, and diet have all been purported to be contributing factors (Beronius, Ruden, 2010, Goodman, Witorsch, 2009, Myers et al., 2009, Thigpen, Setchell, 2007, Vandenberg, Maffini, 2009). Drawing conclusions about safety from this fragmentary and discordant literature has proven to be difficult and contentious. Weight of evidence assessments have been conducted by numerous groups but the results have been inconsistent in their conclusions about the level of concern consumers should have about BPA (Hengstler et al., 2011, NTP, 2009, vom Saal, Akingbemi, 2007). The present, largely mechanistic, study sought to inform this debate by providing information regarding the potential for BPA exposure, during the critical period in which the hypothalamus is undergoing steroid hormone directed sexual differentiation, to alter the expression of genes known to be critical or organizing sex differences.