Pre- and Postnatal Developmental Exposure to the Polychlorinated Biphenyl Mixture Aroclor 1221 Alters Female Rat Pituitary Gonadotropins and Estrogen Receptor Alpha Levels

Abstract

Polychlorinated-biphenyls (PCBs) are industrial compounds, which were widely used in manufacturing of electrical parts and transformers. Despite being banned in 1979 due to human health concerns, they persist in the environment. In humans and experimental model systems, PCBs elicit toxicity in part by acting as endocrine-disrupting chemicals (EDCs). Aroclor 1221 (A1221) is a weakly estrogenic PCB mixture known to alter reproductive function in rodents. EDCs can impact hormone signaling at any level of the hypothalamic-pituitary-gonadal (HPG) axis, and we investigated the effects of A1221 exposure during the prenatal and postnatal developmental periods on pituitary hormone and steroid receptor expression in female rats. Examining offspring at 3 ages, postnatal day 8 (P8), P32 and P60, we found that prenatal exposure to A1221 increased P8 neonate pituitary luteinizing hormone beta (Lhb) mRNA and LHβ gonadotrope cell number while decreasing LH serum hormone concentration. No changes in pituitary hormone or hormone receptor gene expression were observed peri-puberty at P32. In reproductively mature rats at P60, we found pituitary follicle stimulating hormone beta (Fshb) mRNA levels increased by prenatal A1221 exposure with no corresponding alterations in FSH hormone or FSHβ expressing cell number. Estrogen receptor alpha (ERα) mRNA and protein levels were also increased at P60, but only following postnatal A1221 dosing. Together, these data illustrate that exposure to the PCB A1221, during critical developmental windows, alters pituitary gonadotropin hormone subunits and ERα levels in offspring at different phases of maturation, potentially impacting reproductive function in concert with other components of the HPG axis.

Graphical Abstract

1. Introduction

Since the time of the industrial revolution, chemical production has increased tremendously in the United States and worldwide. The ubiquity of human-made chemicals in the environment leads to daily exposure to compounds that can interfere with normal hormone-driven actions. These substances are classified as endocrine disrupting chemicals (EDCs) and in humans are known to cause cancer [1,2], metabolic disorders [3,4], neurological/behavioral changes [5-7], decreased fertility [8,9], and other disease states. Polychlorinated biphenyls (PCBs) are a well-known class of EDCs widely used in the electrical manufacturing industry during the 20^th century in products like hydraulic fluids and flame retardants [10]. Concerns about the toxic effects of PCBs on human health led to widespread banning of PCB use by the late 1970s. Unfortunately, due to their inherent stability and ability to accumulate in fatty tissues [11], PCB compounds persist in the environment and can concentrate in the food chain, especially in fish harvested from PCB contaminated waters [12-14]. Of concern are the negative consequences of PCBs on reproduction. Decreased fertility [15,16], accelerated reproductive senescence or early menopause [17,18] and ovarian follicle atresia [19], have all been documented following PCB exposure in rodents and humans. Beyond acute toxicity, PCBs can be passed from mother to offspring during pregnancy or lactation [20,21] leading to severe developmental consequences such as impaired growth in children both pre- and postnatally, as well as neurodevelopmental problems including difficulties with memory and verbal function [22-25]. Over 200 PCB molecules are known, and humans primarily encounter mixtures of PCBs (Rev in [26]). PCB mixtures, depending on their congeners, have been reported to act as dioxins [27,28], or through hormone receptors such as the androgen receptor (AR) [29,30] or estrogen receptors (ERs) [31,32]. Further, PCB exposures have been reported to affect serum levels of testosterone, estradiol and progesterone as well as AR, ER and progesterone receptor expression in the brain [33-35]. The ability to interfere with normal hormone signaling via cognate hormone receptors make PCB compounds especially disruptive to reproductive function, which relies on hormone signaling through the hypothalamic-pituitary-gonadal (HPG) axis. PCBs can disrupt the female HPG axis at multiple levels including hypothalamic secretion of gonadotropin releasing hormone (GnRH), pituitary gland release of luteinizing hormone (LH) and follicle stimulating hormone (FSH) or negative feedback of estradiol from the ovary [36-43]. Because the HPG axis is established and functional both embryonically and neonatally [41-43] each organ in the axis could be affected by developmental PCB exposure, and the effects on brain and ovary have begun to be explored.

In contrast, relatively little work has determined effects of PCBs at the pituitary level of the HPG axis. Six different hormone secreting cells in the pituitary gland derive from stem-like progenitor cells and proper development of all the pituitary gland cell lineages are dependent upon a host of transcription factors and hormonal influences [44-52]. There are 2 critical periods of pituitary gland differentiation and cell proliferation identified in rodents: prenatal and the first postnatal weeks [53,54]. Interestingly, EDCs can exert different influences on the pituitary gland depending on the timing of developmental exposure. For example, our laboratory has reported increased cell proliferation and an expanded gonadotrope population in the pituitary gland of mice exposed prenatally to the estrogenic EDC bisphenol A (BPA) at 0.5 and 50 μg/kg/day from embryonic day 10.5 (E10.5) to E18.5 [55]; whereas BPA exposure in mice from 0.05-50 μg/kg/day during the postnatal developmental window, postnatal day 0 [P0] to P7, had no effect on gonadotrope cell numbers or proliferation [56]. Aroclor 1221 is a lightly chlorinated PCB mixture reported to act as a weak estrogen [57], and perinatal dosing with A1221 affected reproductive development in two generations of offspring by altering serum LH concentration, GnRH secretion and steroid hormone levels [58]. However, little is known about how A1221 alters pituitary gonadotrope development and function, which depends upon appropriate hormone/receptor signaling for LH and FSH hormone synthesis and release.

In this study we exposed pregnant rats to A1221 during the prenatal and/or postnatal windows of pituitary expansion. We monitored mRNA and protein expression of gonadotrope genes and serum hormones as well as other pituitary gland hormone and hormone receptor genes in female offspring at 3 significant ages, postnatal day 8 (P8), P32 and P60. By focusing on discrete developmental exposures, we better defined the precise timeline of pituitary gland sensitivity to PCB disruption. This work also broadened our understanding of how estrogenic PCB compounds dysregulate reproductive function, with a focus on the contributions of the pituitary gland, adding to our knowledge of EDC action at all levels of the HPG axis.

2. Methods

2.1. Animals and Polychlorinated biphenyl (PCB) Dosing

Sprague-Dawley rats (3-4 months old) were purchased from Envigo and allowed to acclimate 2 to 3 weeks prior to breeding. Rats were housed in polysulfone cages on a 12 hr dark, 12 hr light cycle and were fed with a low phytoestrogen diet and provided with reverse osmosis water in glass bottles ad libitum. Dams were mated with males and embryonic day 0 (E0) was determined by a sperm-positive vaginal smear. Aroclor 1221 (A1221, Accustandard) dissolved in 3% DMSO in sesame oil, or the vehicle, was applied to a ‘Nilla wafer brand cookie (Nabisco) at 1 mg/kg dosage and fed to pregnant dams or lactating dams from 0900-1100 daily during the exposure period. This treatment dosage has been determined by the Gore laboratory to elicit developmental and reproductive alterations in the brain [33,58]. Vehicle consisted of DMSO in sesame oil. To differentiate effects of treatment during key prenatal and early postnatal developmental windows, the two exposure periods were prenatal (E8 to E18) and postnatal (day 1 (P1) to P21) (Figure 1A). This resulted in four A1221 exposure conditions that were assayed in this study (Figure 1B): A control group that received DMSO vehicle during the entire exposure period (DD); the postnatal exposure group (DA) which received DMSO prenatally and A1221 postnatally; the prenatal exposure group (AD) which received A1221 prenatally and DMSO postnatally; and the pre- and post-natal exposure group (AA) which was exposed to A1221 for the entire period. Exposure to A1221 did not have any observable effect on gestational period, offspring weight, sex ratio or litter size.

Figure 1: Experimental design for Aroclor 1221 (A1221) exposure.

A) Embryonic rat pups were exposed to A1221 prenatally from embryonic day 8 (E8) to E18, and postnatally from postnatal day 1 (P1) to P21. Tissues were collected from F1 offspring at ages P8, P32 and P60. B) A1221 exposure groups are identified as follows: DD received DMSO vehicle prenatally and postnatally, DA received DMSO prenatally and A1221 postnatally, AD received A1221 prenatally and DMSO postnatally and AA received A1221 both pre- and postnatally.

2.1.1. Euthanasia and tissue collection

Only female offspring were used in this study and euthanized at ages P8, P32 and P60. These ages are important because rats at P8 are still undergoing pituitary gland proliferation and lineage differentiation, while P32 rats are transitioning to puberty and initiating estrus cyclicity. In contrast, P60 rats are adult and reproductively mature. Both male and female offspring were kept to equalize litter size and sex ratio during the postpartum period prior to weaning. Litters were culled to 3 males and 5 females. The five females were randomly assigned to use at different ages (P8, P32, P60) and method of euthanasia (perfusion or decapitation) such that no more than one female from a litter was used for any unique endpoint. Litters were weaned into same-sex cages at P22. For consistency, adult female rats were euthanized on the proestrous stage of the estrous cycle, as close to P60 as possible, with slight variations due to the timing of proestrus onset. At those ages, half of the rats were euthanized by rapid decapitation, and the pituitary removed, placed in a chilled microfuge tube, and snap-frozen on dry ice and stored at −80C for mRNA extraction and analysis. The second set of animals was deeply anesthetized and perfused transcardially with 4% paraformaldehyde in PBS as published previously [59]. All animal procedures were performed in accordance with protocols approved by the Institutional Animal Care and Use Committee (IACUC), University of Texas at Austin.

2.2. Serum Hormone Quantification

Blood was collected at time of euthanasia, allowed to clot at room temperature, centrifuged at 1500 x g for 5 minutes, and serum was collected and stored at −80. Serum LH and FSH were measured in 10 μl serum duplicates using the Milliplex Rat Pituitary Magnetic Bead Assay (RPTMAG-86K; Millipore) and were read on a Luminex Magpix (Austin, TX). Serum hormone concentrations were calculated using the Milliplex Analyst software. LH assay sensitivity was 4.9 pg/ml, and intraassay CV was 5.58%. FSH assay sensitivity was 24.4 pg/ml and intraassay CV was 3.87%.

2.3. Quantitative Polymerase Chain Reaction (qPCR) of mRNA

Pituitaries from P8, P32 and P60 rats were flash frozen and processed for RNA using Trizol reagent (Invitrogen) according to the manufacturer’s protocol. cDNA was prepared from 500 ng of total RNA using the Protoscript First Strand cDNA Synthesis Kit (New England Biolabs) according to the manufacturer’s instructions. For qPCR, 1 μl of 1:5 diluted cDNA was amplified per 25 μl reaction containing 1x platinum Taq reaction buffer (Invitrogen), 3 mM MgCl₂ (Invitrogen), 0.2 mM dNTPs (New England Biolabs), 1x SYBR green stain (Invitrogen), 10 nM Fluorescein (BioRad), 10 pmol rat gene specific primers (Integrated DNA Technologies or Invitrogen) and 0.1 μl of Platinum Taq DNA polymerase (Invitrogen). All data were normalized Gapdh as a housekeeping gene. The reaction was carried out using a Bio-Rad iQ5 real-time PCR instrument. Relative fold change versus control was calculated using the delta-delta Ct method (2^−ΔΔCt) [49]. Oliognucleotide primer sequences are listed in Table 1.

Table 1.

Primers used for qPCR

Gene	Accession number	Forward Primer	Reverse Primer
Gapdh	XM_032905640.1	CATAGACAAGATGGTGAAGGTCGG	GTCCCACTTTGTCACAAGAGAAGGC
Lhb	NM_012858.2	CCCAGTCTGCATCACCTTCAC	ACTCGAACCATGCTAGGACAGTAG
Fshb	NM_001007597.2	TGGTGTGAGGGCTACTGCTAC	ACAGCCAGGCAATCTTATGGTCTC
Nr5a1	NM_001191099.2	AAATTCCTGAACAACCACAGC	GCATCTCAATGAGAAGGTTG
Gnrhr	NM_031038.3	ATGATGGTGGTGATTAGCC	ATTGCGAGAAAACTGCTGGC
Gh	NM_001034848.2	AGGGCATCCAGGCTCTGAT	GCATGTTGGCGTCAAACTTG
Prl	NM_012629.1	CATCAATGACTGCCCCACTTC	CCAAACTGAGGATCAGGTTCAAA
Pomc	NM_139326.3	TGTACCCCAATGTCGCCGAG	AGGACGTGCTCCAAGCCATCA
Tshb	NM_013116.2	AGACTTCACCTACAGAACGGTGG	AGTTGGTTTTGACAGCCTCGTG
Esr1	NM_012689.1	GAGCACATTCCTTCCTTCCG	GCCGAGGTACAGATTGGCTT
Ar	NM_012502.2	TACGGAGCTCTCACTTGTGG	CATTTCCGGAGACGACACGA
Pgr	NM_022847.2	GGTGGAGGTCGTACAAGCAT	TGTGGGATTTGCCACATGGT
Mki67	NM_001271366.1	CCTGCCCGACCCTACAAAAT	TGCTGCTTCTCCTTCACTGG

2.4. Immunohistochemistry (IHC)

Chemical reagents/buffers for IHC were purchased from Fisher Scientific unless otherwise indicated. For IHC experiments, pituitary glands from P8, P32 or P60 rats were collected from rats following intracardial perfusion with 4% paraformaldehyde as described previously [59] and post fixed for 1 hr in 4% paraformaldehyde. Tissue was either cryoprotected in 30% sucrose/phosphate buffered saline (PBS) and frozen for cryosectioning or dehydrated in a graded series of ethanol washes and embedded in paraffin wax (paraplast plus, McCormack Scientific). Superfrost plus slides (Fisher) were mounted with frozen pituitary tissue sectioned at 10 μm using a cryostat (Leica) or paraffin embedded tissue sectioned with a microtome (Leica) to 6 μm thickness.

Primary antibodies used were: LHβ, FSHβ (rabbit host, 1:100 dilution, National Hormone and Peptide Program) and ERα (rabbit host, 1:1000 dilution, Millipore Sigma). Prior to immunostaining, slides with paraffin embedded pituitary tissue were deparaffinized with xylene and rehydrated with ethanol followed by PBS. Frozen tissue slides were thawed to room temperature and postfixed with 4% paraformaldehyde for 10 min. For estrogen receptor alpha (ERα) staining, tissue sections were subjected to antigen retrieval in 0.1M citrate solution (10 min boil, followed by 20 min hot bath for paraffin slides and a 10 min hot bath for frozen sections).

All tissue samples were blocked for 1 hour with 5% normal donkey serum (Jackson ImmunoResearch), 3% bovine serum albumen (Jackson ImmunoResearch), and 0.5% Triton-X100 in PBS. Primary antibodies were applied to slides overnight at 4 °C. Paraffin embedded tissue slides were then incubated for 1 hour at room temperature with donkey anti-rabbit biotin conjugated secondary antibody followed by 1 hour incubation with streptavidin-cy3 tertiary (Jackson ImmunoResearch, 1:200 dilution each). For frozen tissue, slides were incubated for 1 hour at room temperature with donkey anti-rabbit-cy3 secondary antibody only (Jackson ImmunoResearch, 1:200). Sections were mounted using antifade medium (0.1M Tris pH 8.5, 20% glycerol, 8% polyvinyl alcohol, 2.5% 1,4-diazabicyclo[2.2.2]octane) containing nuclear stain 4’,6-diamidino-2-phenylindole (DAPI), and visualized with a fluorescent microscope (Leica DM2560).

2.5. Quantification of IHC

For P8 pituitary cell quantification, 4 individual 20 x images from 3 sections spanning the rat pituitary (dorsal, medial and ventral) approx. 95 μm apart were used for cell counting. For P60 quantification 2 individual 20 x images from 6 slides (2 each dorsal, medial and ventral) approx. 150 μm apart were selected to span the pituitary. For LHβ and FSHβ cell counting, positively stained cells were counted manually using ImageJ Fiji software (NIH) and normalized to pituitary tissue area. For ERα intensity quantification, Images were adjusted to equalize background staining across all sections and subsequently quantified for mean intensity per fixed pituitary area using Photoshop 2021 (Adobe).

2.6. Statistical analysis

Data are represented as mean +/− SEM. Statistical significance was determined using one-way ANOVA and Dunnett’s post hoc test as described in each figure legend. P values less than 0.05 were considered significantly different from control values. Significant outliers were removed from data sets by the Grubb’s test. All analyses were performed using GraphPad Prism 9.4.1.

3. Results

3.1. Prenatal Exposure to A1221 alters LH expressing gonadotropes in the pituitary of P8 rats

We first assessed the effects of A1221 pre- and postnatal exposure on neonatal offspring at P8. Pituitary glands from P8 mice were analyzed for Lhb mRNA expression changes by qPCR. We found that Lhb mRNA was significantly increased in the prenatal-only exposure group, AD relative to controls, DD (Figure 2A). An examination of LHβ expressing gonadotropes was conducted using IHC. Figure 2B shows representative images of LHβ stained pituitary cells for each A1221 dosing group. Cell counting revealed an increase in LHβ positive cells for the prenatal exposed rats (AD) (Figure 2C). Measuring LH hormone levels in P8 rat serum we discovered that LH was significantly reduced in both the postnatal (DA) and the prenatal (AD) exposure groups relative to the control rats (DD) (Figure 2D). Fshb mRNA levels in P8 rat pituitaries were unchanged in any of the A1221 exposure groups (Figure 2E). FSH serum hormone levels were quantified in P8 rats, and we also found no difference in circulating FSH hormone (Figure 2F). Other genes expressed in pituitary gonadotrope cells include the gonadotropin releasing hormone receptor (Gnrhr), which can fluctuate with changing levels of GnRH [60] and the gonadotrope specifying transcription factor Nr5a1. We found no change in Gnrhr mRNA levels in any A1221 treatment versus control pituitaries (Figure 2F). We also saw no significant differences in Nr5a1 relative to control pituitaries (Figure 2G). Taken together these data show that prenatal exposure to the PCB A1221 selectively affects LHβ expressing gonadotrope cells by increasing Lhb mRNA levels or the actual number of gonadotropes expressing LHβ mRNA and protein, but with an accompanying decrease in circulating levels of LH hormone.

Figure 2: Prenatal A1221 exposure alters LHβ expressing pituitary gonadotrope cells at P8.

A) Levels of mRNA for luteinizing hormone beta (Lhb) show a significant increase following prenatal A1221 exposure (AD) relative to DMSO exposed control animals (DD). N=5-9 pituitaries per group. B) Pituitaries immunostained for LHβ protein expression. A representative image of the anterior lobe is shown for each treatment group as indicated. Scale bar = 50 μm C) Quantification of LHβ immunostained cells in the pituitary gland shows a significant increase in LHβ positive cells/area in the prenatal A1221 exposure group (AD) versus the control pituitaries (DD). N=3-4 pituitaries per group. D) Serum LH levels were quantified for each A1221 exposure group. Both the postnatal (DA) and prenatal (AD) dosed cohorts have significantly lower serum LH relative to control animals (DD) N=12-19 animals per group. E) Follicle stimulating hormone beta (Fshb) mRNA is not affected by A1221 any exposure. F) Serum FSH levels were not altered by A1221 exposure, N=13-19 animals per group. Gonadotropin releasing hormone receptor (Gnrhr) and the gonadotrope specification factor Nr5a1 mRNA expression was not changed by exposure to A1221 (G and H respectively). N=5-9 pituitaries per group for panels E, G and H. * indicates p≤0.05 by One Way ANOVA and Dunnett’s post hoc test.

3.2. Other pituitary hormone and hormone receptor genes are not affected by any PCB exposures at P8

In the P8 pituitaries we also examined the mRNA levels of other pituitary hormones: prolactin (Prl), growth hormone (Gh), proopiomelanocortin (Pomc) and thyroid stimulating hormone beta (Tshb) (Figure 3A, B, C and D). Interestingly, we found no effect of A1221 dosing on any other pituitary hormone in P8 rats. This indicates that the effects of prenatal A1221 dosing we observe on Lhb mRNA and cell number appear to be confined to LH expressing gonadotrope cells. We also examined a panel of genes encoding steroid hormone receptors that might be affected by A1221 exposure, including estrogen receptor alpha (Esr1), androgen receptor (Ar), progesterone receptor (Pgr) (Figure 3E, F and G). We also assessed the cell cycle marker Mki67, a measure of active proliferation (Figure 3H). We found no changes in mRNA levels for any of the hormone receptor genes or indicators of increased cell proliferation during this period of pituitary gland expansion.

Figure 3: Non-gonadotrope pituitary gland hormone genes as well as hormone receptor and proliferation marker genes appear unaffected by A1221 exposure at P8.

A survey of hormone genes including A) prolactin (Prl), B) growth hormone (Gh), C) proopiomelanocortin (Pomc) and D) thyroid stimulating hormone beta (Tshb) showed no mRNA changes in the pituitary by any A1221 exposure. Nuclear hormone receptor genes involved in reproductive function of the pituitary gland are not altered by A1221 dosing as seen by qPCR for E) estrogen receptor alpha (Esr1), F) androgen receptor (Ar) and G) progesterone receptor (Pgr). H) Proliferation is not changed by A1221 exposure as measured by mRNA for the cell cycle gene Mki67. N=6-9 pituitaries per sample.

3.3. Pre- and Postnatal A1221 dosing does not affect pituitary gonadotrope mRNAs, hormone levels or hormone receptor genes at P32

Because we discovered alterations in Lhb mRNA, cell number, and serum hormone levels in P8 rat pituitaries, we next analyzed pituitary glands at age P32 to determine if these changes persist beyond the neonatal period and after the end of dosing. Interestingly, we found no differences in mRNA levels of Lhb or Fshb in P32 pituitaries in any A1221 treatment group relative to control samples (Figure 4A and B). Consistent with this finding, neither LH nor FSH serum concentration was altered by A1221 treatment at P32 (Figure 4C and D). A further interrogation of hormone receptor genes Esr1, Ar and Pgr and the proliferation marker Mki67 also showed no significant changes compared to the DMSO vehicle dosed animals (Figure 4E, F, G and H). These data suggest that the neonatal changes observed are no longer apparent at P32.

Figure 4: Pituitary gonadotrope cell genes are no longer altered by A1221 exposure at age P32.

Levels of mRNA for luteinizing hormone beta (Lhb) and follicle stimulating hormone beta (Fshb) (A and B) as well as LH and FSH serum hormone concentration (C and D) are not affected by A1221 in any dosing cohort. Nuclear hormone receptors and cell proliferation are not changed in pituitaries exposed to A1221 relative to controls as seen by qPCR for E) estrogen receptor alpha (Esr1), F) androgen receptor (Ar), G) progesterone receptor (Pgr) and H) the cell cycle gene Mki67. N=5-7 pituitaries per sample for panels A, B, E, F and G. N=16-24 animals per sample for panels C and D.

3.4. Neonatal gonadotrope changes in LH are due to prenatal A1221 are not sustained in adulthood

At P60, rats undergo estrous cycles and are reproductively mature. Once again, we screened gonadotrope specific gene expression in P60 rat pituitaries by qPCR. We saw no change in Lhb mRNA levels with PCB exposure (Figure 5A). LHβ positive cells were visualized by IHC and quantified by cell counting to complement mRNA data. No difference in LHβ positive gonadotrope number was observed in any A1221 dosed cohort relative to the control pituitaries (Figure 5B and C). Moreover, serum levels of LH hormone were not different in any PCB exposure group compared to control rats (Figure 5D). Based on these results the neonatal changes in LH positive gonadotrope cells that we found at P8 are not maintained in the adult pituitary gland.

Figure 5: LHβ expressing gonadotrope cells in adult (P60) pituitary glands are not impacted by pre- and postnatal A1221 exposure.

A) Luteinizing hormone beta (Lhb) mRNA levels showed no difference in any A1221 exposure cohort relative to controls. N=6-7 pituitaries per sample. B) Pituitary gland tissue immunostained for LHβ protein expression. A representative image of the anterior lobe is shown for each treatment group as indicated. Scale bar = 50 μ C) Quantification of LHβ immunostained cells in the pituitary gland shows no difference in the number of LHβ positive cells/area in A1221 dosed animals versus controls. N=3 pituitaries per sample. D) LH serum concentration is not changed in A1221 P60 rats compared to control animals. N=16-20 animals per sample.

3.5. Prenatal A1221 exposure leads to increased Fshb mRNA at P60

Although it was not altered neonatally, we also examined gene expression of FSHβ in the P60 pituitary gland. Surprisingly, we found a significant increase in Fshb mRNA in the prenatal-only A1221 group (AD) relative to the control samples (Figure 6A). We subsequently performed IHC staining of FSHβ positive gonadotrope cells to determine if gonadotrope number might be altered. Quantification of FSHβ positive cells revealed a slight, but significant decrease in cell number in the postnatal A1221 exposure group (DA) and the pre- and postnatal cohort (AA) relative to the control pituitary. However, the prenatal A1221 treatment group (AD) showed no change in FSHβ cell number versus the controls (Figure 6B and C). Measurement of FSH hormone in P60 serum revealed no change in circulating FSH (Figure 6D). Finally, we measured mRNA expression of the gonadotrope cell factors Gnrhr and Nr5a1 by qPCR. No difference was observed in expression levels for either gene compared to DMSO vehicle treated pituitary gland (Figure 6E and F). These data imply that in the adult rat pituitary gland, prenatal exposure to the PCB A1221 may alter Fshb mRNA production, but does not show a corresponding impact on hormone release or FSHβ positive cell number. This suggests that the full effects of postnatal A1221 on Fshβ expressing cells might not manifest until adulthood.

Figure 6: FSHβ expressing gonadotrope cells in adult (P60) pituitary glands show increased Fshb mRNA expression following prenatal A1221 exposure.

A) Follicle stimulating hormone beta (Fshb) mRNA is significantly increased in the prenatal A1221 exposure group (AD) N=5-7 pituitaries per sample. B) Pituitary gland immunostained for FSHβ protein expression. A representative image of the anterior lobe is shown for each treatment group as indicated. Scale bar = 50 μm C) Quantification of FSHβ immunostained cells in the pituitary gland shows a significant decrease in the number of FSHβ positive cells/area in the postnatal (DA) as well as the pre- and postnatal (AA) A1221 exposure groups relative to control samples (DD). N=3 pituitaries per sample. D) FSH serum concentration is not changed in A1221 exposed rats compared to control animals. N=18-21 animals per sample. Gonadotropin releasing hormone receptor (Gnrhr) (E) and gonadotrope specification gene Nr5a1 (F) are not changed by A1221 exposure. N=6-7 pituitaries per sample. * indicates p≤0.05 by One Way ANOVA and Dunnett’s post hoc test.

3.6. Other pituitary hormone and receptor mRNA levels are not affected by PCB exposure at P60

To assess whether developmental PCB exposure affects other pituitary hormone cell types in the adult we analyzed Prl, Gh, Pomc and Tshb gene expression at P60. No differences were observed in mRNA levels of these other pituitary cell hormone genes in any dosing cohort relative to controls (Figure 7A, B, C and D). We also examined the hormone receptors Ar and Pgr as well as the proliferation marker Mki67 for any changes in mRNA level and no found no significant differences as compared to control pituitary glands (Figure 7E, F and G). These findings suggest that the induction of Fshb mRNA observed with prenatal PCB dosing does not affect other pituitary hormone lineages or alter cell replication.

Figure 7: Non-gonadotrope pituitary gland hormone genes as well as hormone receptor genes Ar and Pgr and cell proliferation appear unaffected by A1221 exposure at P60.

A survey of hormone genes including A) prolactin (Prl), B) growth hormone (Gh), C) proopiomelanocortin (Pomc) and D) thyroid stimulating hormone beta (Tshb) showed no mRNA changes in the pituitary by any A1221 exposure. Nuclear hormone receptor genes androgen receptor (Ar) (E) and progesterone receptor (Pgr) (F) are also not changed by A1221 dosing relative to control pituitaries. G) Proliferation is not altered by A1221 exposure as measured by mRNA for Mki67. N=5-7 pituitaries per sample.

3.7. Postnatal PCB exposure increases Esr1 mRNA and ERα protein expression in P60 adult rat pituitary gland

Estrogen hormone feedback is an essential element of the HPG axis and ERα signaling is important in multiple cell types of the pituitary gland including the gonadotropes. We measured Esr1 levels in pituitaries at P60 by qPCR and discovered a potent increase in Esr1 mRNA in the postnatal dosing cohort (DA) compared to DMSO exposed controls (Figure 8A). Building upon the mRNA data for Esr1 at P60, we immunostained pituitaries for ERα and quantified the fluorescence intensity of each PCB dosed group relative to control levels. We found that the postnatal A1221 cohort (DA) exhibited significantly increased ERα staining relative to the DMSO vehicle dosed rats (Figure 8B and C). Taken together, postnatal exposure to A1221 alters ER levels and possibly E2 hormone feedback signaling in the pituitary at PND 60.

Figure 8: ERα mRNA and protein level is elevated by A1221 postnatal exposure in P60 rat pituitary.

Levels of mRNA for estrogen receptor alpha (Esr1) are significantly increased following postnatal A1221 exposure (DA) relative to control pituitaries (DD). N=6-7 pituitaries per sample. B) Pituitary gland immunostained for ERα protein expression. A representative image of the anterior lobe is shown for each treatment group as indicated. Scale bar = 50 μm C) Quantification of ERα immunostained cells in the pituitary gland shows a significant increase in ERα staining intensity in the postnatal A1221 exposure group relative to control pituitary gland. N=4 pituitaries per sample. * indicates p≤0.05 by One Way ANOVA and Dunnett’s post hoc test.

4. Discussion

In this study, we found that the PCB mixture A1221 elicits very specific effects on pituitary gland gonadotropes and hormone receptor expression, depending on the timing of exposure and age at which pituitaries are examined. Multiple impacts of prenatal A1221 exposure were observed in P8 neonate rats including increases in pituitary gland Lhb mRNA and gonadotrope cell number as well as a decrease in serum LH concentration. By P32, offspring showed no significant changes resulting from pre- or postnatal A1221 dosing. However, adult pituitaries, aged P60, had increased Fshb mRNA in gonadotrope cells following prenatal A1221, and elevated Esr1 mRNA and protein levels with postnatal A1221 exposure. Our data help to define developmental periods when the pituitary is more vulnerable, or resistant, to the effects of EDCs and the changes that occur over time. Moreover, our results indicate that some of the reproductive deficits caused by PCB exposure may be due to changes at the level of the pituitary gland.

P8 is a significant age in pituitary gland development as it occurs during the period of neonatal differentiation and proliferation of the hormone secreting cell populations. Following prenatal A1221 exposure the Lhb expressing gonadotrope cell population appears to be expanded based on Lhb mRNA increases and, more telling, quantification of LHβ IHC. This could indicate an increase in the actual LHβ expressing gonadotrope number or an elevated level of LHβ protein making the cells more visible during cell counting. At P8, we also observed a significant decrease in LH serum concentration following prenatal A1221 dosing that was coupled with high pituitary tissue expression of the LHβ mRNA and protein. This effect was previously reported in adult rats acutely exposed to PCBs 126 and 153 and may indicate an impairment of LH release from pituitary gonadotrope cells [61]. Notably the postnatal A1221 (DA) dosing cohort also shows a significant decrease in serum LH but no corresponding increase in pituitary Lhb mRNA or gonadotrope cell number. Overall, these data suggest that LH expression and release is primarily impacted by prenatal A1221 exposure. This could be caused by direct effects at the pituitary or influences on hypothalamic GnRH release that have a downstream impact on the pituitary.

Strikingly, the increase in Lhb mRNA and LHβ expressing gonadotrope cells following prenatal, but not postnatal A1221 exposure is similar to what has been reported with prenatal versus postnatal exposure to BPA [55,56]. These results suggest there might be differences in pituitary gland vulnerability to EDC exposure during the prenatal and postnatal developmental windows. One commonality between BPA and A1221 is that both compounds can act as weak estrogens and signal through hormone receptors [31,62]. The postnatal pituitary gland appears to be more resistant to assaults from EDC exposure, which may be due in part to low hormone responsiveness of the neonatal pituitary gland. Despite a functional HPG axis, the pituitary gland is unresponsive to GnRH during the early neonatal period [56,63]. Conversely, our data indicate that the prenatal window of pituitary development is extremely sensitive to hormone disruption. While the HPG axis is established prenatally and hormone receptors are expressed in each organ, the fetus is relatively protected from maternal estrogens because of alpha fetoprotein (AFP) [64,65]. It is possible that during embryonic development the pituitary gland directly, or through the HPG axis, has fewer defenses against EDC exposure, including bypassing AFP binding, making the gland more susceptible to hormone disruption.

At P32, the differences we observed for pituitary LH gene expression neonatally are no longer apparent. This could indicate that the pituitary is able to correct aberrations resulting from prenatal A1221 exposure. The neonatal pituitary gland is known to be highly plastic, containing both stem cells and highly proliferative lineage specified progenitor cells [66-68]. These progenitor cells are sensitive to hormone feedback and it is likely that they sense both intra- and extra- pituitary hormone levels and adjusted cell number accordingly. Our data indicate that the developing pituitary gland has the capacity to restore balance in its gonadotrope cell populations possibly lending it some resistance to effects of EDC exposure by the time of puberty onset.

At age P60, we observed changes in FSHβ expressing pituitary gonadotrope cells following prenatal A1221 exposure not apparent prior. Levels of Fshb mRNA were significantly increased in the prenatal A1221 cohort, but with no concomitant increase in FSHβ expressing gonadotrope cells. In fact, cell quantification showed a decrease in FSHβ expressing cells in the postnatal and pre- and postnatal exposure groups relative to controls. It is conceivable that A1221 exposure reduces FSHβ expressing gonadotrope cells in all adult cohorts, but the prenatal exposure group can compensate by increasing FSHβ mRNA and protein expression. Despite the alterations in Fshb mRNA and gonadotrope cells we did not find any differences in serum FSH hormone levels at P60 following any A1221 exposure. Of note, FSH is notoriously difficult to quantify in serum and potentially a more sensitive assay or specific sampling time could produce different results [69].

At both P8 and P60, prenatal A1221 appears to affect discrete populations of gonadotrope cells, Lhb expressing at P8 and Fshb positive at P60. While most gonadotrope cells express and secrete both LH and FSH there are mechanisms for differential expression and hormone release. Fshb mRNA levels at P60 could be affected by inhibins produced by the ovary. These proteins can directly suppress Fshb transcription independently of Lhb [70]. Additionally, the timing and pulsatility of GnRH from the hypothalamus selects for either LH or FSH release [71,72]. The perceived impairment of LH secretion in P8 pituitary glands following prenatal and postnatal A1221 exposure might also indicate changes in GnRH released from the brain. There is evidence, for example, that A1221 can directly impact GnRH secretion in vitro, as A1221 potentiated GnRH release in cultured neurons [73]. Altered GnRH dynamics have also been seen in vivo in adult lambs exposed pre- and postnatally to two different estrogenic PCB congeners (PCB 118 and 153), with higher levels of LH induced in PCB dosed lambs relative to controls [74]. Another recent survey found high serum FSH:LH ratios during the menstrual cycle of a group of indigenous women living near a PCB contaminated portion of the St. Lawrence river that could possibly involve impaired GnRH signaling [75]. More detailed electrophysiological studies would be required to investigate whether prenatal A1221 can change the pulsatility of GnRH secretion at P8 or P60 or if A1221 exposure causes direct effects on LH or FSH at the level of the pituitary.

An interesting consequence of A1221 exposure was the increase in pituitary ERα levels in the postnatal exposure cohort. ERα is widely expressed in most of the cells of the anterior pituitary gland, including the gonadotrope cells. A1221 is reported to act as a weak estrogen, so early postnatal exposure could result in long term disruption of ERα action via A1221 signaling at a time when the HPG axis does not normally receive E2 feedback. In addition, ERα signaling has been shown to control the level of its own receptor in some cell and tissue contexts [76,77], so the increase in pituitary ERα could also result from changes in E2 levels possibly due to impaired synthesis in the ovary. Aromatase is an enzyme essential for the biosynthesis of E2 and it has been previously reported that A1221 can decrease aromatase activity In vitro. [78]. More analysis would be required to determine the exact mechanism by which A1221 alters ERα expression in the pituitary gland, but our data show that postnatal A1221 exposure can perturb a critical hormone receptor involved in HPG signaling with potentially negative consequences to reproduction in mature adults.

Ultimately, this study defines PCB critical exposure windows for the developing pituitary gland. Pituitary LH expression is highly sensitive to A1221 during prenatal developmental, but with some ability to self-correct. However, even short-lived changes in pituitary gland hormone secretion or gonadotrope population could have long term consequences. For example, inappropriate hormone signaling from the neonate pituitary might affect the brain and/or ovary during critical periods of their development resulting in disruption of reproductive function or decreased fertility. The concept of developmental origins of health and disease (DOHaD) posits that EDC exposures during critical developmental periods can have profound effects even at low concentrations, and the consequences of developmental exposures may not become apparent until later in life [79,80]. This is precisely what we observed in our dosing study, as alterations in Fshb mRNA and ERα mRNA and protein did not manifest until adulthood even though the A1221 exposures that elicited those changes occurred during prenatal and early postnatal periods respectively. A further concern is that developmental PCB exposures can induce epigenetic modifications which are subsequently passed on to offspring [81,82]. Given this, the damage caused by exposure to A1221 could persist for generations.

Highlights:

• Pituitary gonadotrope cells are altered by prenatal exposure to Aroclor 1221

• Developmental exposure to Aroclor 1221 affects FSHβ and ERα in adult pituitary

• The pituitary is more susceptible to disruption by Aroclor 1221 during prenatal development