Background
The aryl hydrocarbon receptor (AHR) is a ligand-activated transcription factor [
1] that has been shown to have important biological roles in ovarian function [
2‐
8]. Specifically, previous studies indicate that pre-pubertal aryl hydrocarbon receptor knockout (AHRKO) mice have slower antral follicle growth and a reduced capacity to produce estradiol compared to wild-type (WT) mice [
2‐
8]. Further, the slow growth and reduced estradiol production is only observed in pre-pubertal AHRKO mice, but not in sexually mature AHRKO mice [
8].
One possible explanation for the slow follicle growth and reduced production of estradiol in pre-pubertal AHRKO mice compared to pre-pubertal WT mice may be that pre-pubertal AHRKO mice have a reduced capacity to respond to follicle-stimulating hormone (FSH) compared to pre-pubertal WT mice. FSH is critically important for promoting estradiol production in granulosa cells, and estradiol is the hormone that directly stimulates growth in antral follicles [
9,
10]. Given that the levels of FSH are generally low prior to puberty and dramatically rise after puberty, it is possible that AHRKO mice are able to overcome their potentially low ability to respond to FSH prior to puberty by the presence of higher FSH levels after they become sexually mature. These possibilities are supported by previous studies that show that AHRKO follicles have reduced mRNA expression of FSH receptors (
Fshr) as well as a reduced number of FSH binding sites compared to WT follicles [
5,
8]. Further, they are supported by studies indicating that the AHR and FSH pathways may interact to regulate transcription of the
Fshr because the AHR binds to aryl hydrocarbon response elements in the promoter regions of the
Fshr and an E-box binding site in the mouse ovary [
5,
11].
If pre-pubertal AHRKO follicles are less responsive than pre-pubertal WT follicles to FSH, this may lead to reduced estradiol biosynthesis and slow follicle growth. This is because binding of FSH to FSHR promotes cytochrome P450, family 19, subfamily A, polypeptide 1 (
CYP19A1) expression [
12], the enzyme that converts theca cell-produced androgens (
i.e., dehydroepiandrosterone and androstenedione) into estrogens (
i.e., estradiol and estrone) [
9,
10,
13], and the estrogens then stimulate proliferation of granulosa cells to promote follicle growth [
9,
14]. Thus, the current study was designed to test the hypothesis that low FSH responsiveness is responsible for the slow growth and reduced estradiol production observed in pre-pubertal AHRKO
versus pre-pubertal WT antral follicles.
To test this hypothesis, antral follicle growth was compared in follicles from pre-pubertal WT and AHRKO mice in response to varying concentrations of FSH. As balanced steroid hormone synthesis is crucial for antral follicle growth [
9], the levels of steroid hormones and the expression of factors that regulate steroidogenesis were also compared in WT and AHRKO follicles cultured with FSH. Further, studies also show that FSH regulates the levels of inhibin beta-A (INHBA) in antral follicles during granulosa cell differentiation [
15‐
17]; and that reduced levels of
Ahr mRNA are related to high levels of
Inhba mRNA in antral follicles [
18]. Thus, the levels of
Inhba mRNA were compared in WT and AHRKO follicles cultured with varying concentrations of FSH.
Methods
Chemicals
Fetal bovine serum (FBS) was obtained from Atlanta Biologicals (Lawrenceville, GA). Human recombinant FSH was obtained from Dr. A.F. Parlow from the National Hormone and Peptide Program (Harbor-UCLA Medical Center, Torrance, CA). Penicillin, streptomycin, ITS (insulin, transferrin, selenium), antibiotic antimycotic solution, and ampicillin were obtained from Sigma-Aldrich (St. Louis, MO). Alpha-minimal essential medium (α-MEM) was obtained from Invitrogen (Carlsbad, CA).
Animals
The AHRKO mice were generated by Schmidt
et al.[
19] and are in C57BL/6J background, along with their WT littermates. Only homozygous mice (WT and AHRKO) were used from breeding colonies maintained by our laboratory at the University of Illinois at Urbana-Champaign, Veterinary Medicine Animal Facility. Mice were provided with food and water for
ad libitum consumption, and maintained in a temperature and light controlled room (24 ± 1°C, 12 h daylight/12 h dark cycle) with 35 ± 4% relative humidity. Genetic screening was performed using ear tissue punches as previously described [
5,
6]. Female WT and AHRKO mice were euthanized by carbon dioxide inhalation. The ovaries were removed and early antral follicles were isolated as described below. All animal care, euthanasia, and tissue collection were approved by the Institutional Animal Use and Care Committee at the University of Illinois.
Antral follicle isolation
Ovaries were removed from pre-pubertal WT and AHRKO mice on post-natal day (PD) 30–32 (3–6 mice per genotype per experiment), placed in α-MEM, and cleaned of both interstitial tissue and small follicles using fine watch maker forceps under a dissecting microscope. WT and AHRKO mice on PDs 30–32 were considered to be of a pre-pubertal age because of their lack of a vaginal opening and lack of regular estrous cyclicity. Approximately 20–30 early antral follicles (260–400 μm) were mechanically isolated per ovary, pooled, and randomly assigned by genotype to follicle culture for in vitro follicle growth evaluation as described below.
Measurement of follicle growth
Follicle growth was evaluated in cultured antral follicles as previously described [
6,
20]. Briefly, early antral follicles from WT and AHRKO mice on PDs 30–32 (3–6 mice per genotype per experiment) were placed individually in 96-well culture plates with 150 μL α-MEM, which was immediately replaced with 150 μL of supplemented α-MEM containing 5% FBS, 1% ITS (10 ng/mL insulin, 5.5 ng/mL transferrin, and 5.5 ng/mL selenium), 100 U/mL penicillin, 100 mg/mL streptomycin, and 0, 5, 10 or 15 IU/mL FSH. Follicles were then incubated for 7 days at 37°C in 95% air, 5% carbon dioxide, and humidity saturated. Follicle growth patterns were examined by measuring follicle diameters in perpendicular axes every 24 h using an inverted microscope equipped with a calibrator ocular micrometer as described by Miller
et al.[
20]. Culture media were collected on days 3 and 7 and subjected to measurements of steroid hormone levels as described below. Follicles were discarded from cultures if they were dark in appearance or if they could not retain their oocyte enclosed within the granulosa cell mass. At least three separate cultures per FSH treatment group in WT and AHRKO mice were performed. Each experiment contained 10–16 follicles per treatment. The sizes of follicles per treatment were averaged in each experiment and then data across separate cultures were averaged. At the end of culture, follicles were snap-frozen and stored at −80°C until quantitative PCR (qPCR) analyses as described below.
Measurement of steroid hormone levels
Culture media samples were subjected to enzyme linked immunoassays (ELISA) using kits for measurement of progesterone and androstenedione (DRG International Inc., Mountainside, NJ), dehydroepiandrosterone (DHEA; Alpco Diagnostics, Salem, NH), and estradiol (Calbiotech, Spring Valley, CA). All samples from 5–15 IU/mL FSH treatments were diluted 1:2 (v/v) for androstenedione and 1:5 (v/v) for progesterone assays. Samples collected on days 3 and 7 were diluted 1:10 (v/v) and 1:100 (v/v), respectively, for estradiol assays. Hormone levels were measured on day 3 because this is a time at which media need to be changed in the culture system to maximize follicle growth. Hormone levels were measured on day 7 because this was the last day of culture. Then, a sum of day 3 and day 7 hormone levels was made to represent total hormone production by follicles as they grew in culture. All ELISA procedures were performed according to the manufacturer’s protocol. α-MEM medium was used as background control. Lyphochek Immunoassay Plus controls (Bio-Rad laboratories, Inc.) containing known amounts of specific hormones were included as positive controls in every assay. All samples were run in duplicate, and values were calculated by multiplying by the corresponding dilution factor. The analytical sensitivities, as determined by the ELISA kit manufacturers, were 0.045 ng/mL for progesterone, 0.005 ng/mL for DHEA, 0.019 ng/mL for androstenedione, and 10 pg/mL for estradiol. No samples were below the limit of detection. Intra-assay and inter-assay coefficients of variation for all assays were <10%.
Quantitative real-time polymerase chain reaction (qPCR)
Total RNA (1–2 μg) from pooled antral follicles was extracted using the RNeasy Mini Kit (Qiagen, Inc., Valencia, CA) and then converted to cDNA using iScript cDNA synthesis kit (Bio-Rad Laboratories Inc., Hercules, CA) according to the manufacturers’ protocols. The cDNA was amplified by qPCR as previously described [
21] using a CFX96 Real-time System C1000 Thermal Cycler (Bio-Rad Laboratories Inc., Hercules, CA) and accompanying software according to the manufacturer’s instructions. To allow analysis of the amount of cDNA in the exponential phase, a standard curve from five serial dilutions was generated using cDNA from a pool of WT and AHRKO antral follicles. Specific qPCR primers (Integrated DNA Technologies, Inc, Coralville, IA) for the genes of interest and annealing temperatures are listed in Table
1. SsoFast EvaGreen Supermix (Bio-Rad Laboratories Inc., Hercules, CA) was used as dye for all qPCR analyses. A melting curve was generated at 55–90°C to confirm the generation of a single product, and PCR products were loaded in 3% agarose gel to confirm the product size according to Table
1. Beta actin (
Actb) was used for each sample as an internal control. Relative transcript amount was calculated by a mathematical model developed by Pfaffl [
22]. Briefly, the method calculates the relative expression ratio of the target gene based on the amplification efficiency of each amplicon and the ΔCt of the treated samples
versus the vehicle control. These ratios were then compared to the expression of the reference gene
Actb.
Table 1
Primers used in real-time qPCR analysis
Beta actin |
Actb
| F: ctggcaccacaccttctac | 55.0 | 238 | NM_007393 |
R: gggcacagtgtgggtgac |
Cytochrome P450, family 11, subfamily A, polypeptide 1 |
Cyp11a1
| F: agatcccttcccctggcgacaatg | 60.0 | 192 | NM_019779 |
R: cgcatgagaagagtatcgacgcatc |
Cytochrome P450, family 17, subfamily A, polypeptide 1 |
Cyp17a1
| F: ccaggacccaagtgtgttct | 56.0 | 250 | NM_007809 |
R: cctgatacgaagcacttctcg |
Cytochrome P450, family 19, subfamily A, polypeptide 1 |
Cyp19a1
| F: catggtcccggaaactgtga | 56.0 | 187 | NM_007810 |
R: gtagtagttgcaggcacttc |
Follicle stimulating hormone receptor |
Fshr
| F: gcagatgtgttctccaacctacc | 61.0 | 172 | NM_013523 |
R: ggagagactggatcttgtgaaagg |
Hydroxy-delta-5-steroid dehydrogenase, 3 beta- and steroid delta-isomerase 1 |
Hsd3b1
| F: caggagaaagaactgcaggaggtc | 59.5 | 280 | NM_008293 |
R: gcacacttgcttgaacacaggc |
Steroidogenic acute regulatory protein |
Star
| F: cagggagaggtggctatgca | 57.0 | 262 | NM_011485 |
R: ccgtgtcttttccaatcctctg |
Inhibin, beta-a |
Inhba
| F: tcacctttgccgagtcaggc | 59.0 | 97 | NM_008380 |
R: ccacacttctgcacgctcca |
Statistical analysis
The data were analyzed using SAS 9.2 (Statistical Analysis System Institute, Inc.). General linear models (GLM) for repeated measures were used for comparisons over time in follicle cultures. If the global tests from GLM were significant, Tukey’s tests were used for pairwise comparisons. Multiple comparisons between experimental groups were conducted on data obtained from at least three independent experiments using one-way analysis of variance (ANOVA) followed by Tukey’s post hoc test. Data were expressed as means ± SEM. Statistical significance was assigned at p ≤ 0.05.
Discussion
While previous studies indicate that pre-pubertal AHRKO antral follicles have slow growth and reduced estradiol production compared to WT follicles [
2‐
6], it was not known whether these alterations were due to low FSH responsiveness in AHRKO
versus WT follicles. To investigate this issue, we used an isolated follicle culture system to compare the direct effects of FSH on pre-pubertal WT and AHRKO antral follicles. Our main findings suggest that AHRKO follicles are less responsive to FSH-induced follicle growth than WT follicles because higher levels of FSH are required for AHRKO follicles to reach the same degree of growth as WT follicles. Surprisingly, the response of WT and AHRKO follicles to FSH in terms of steroid hormone production and steroidogenic regulator expression did not follow the response of follicle growth to FSH. Instead, FSH stimulated production of sex steroid hormones and steroidogenic regulators to a similar or greater degree in AHRKO follicles compared to WT follicles. These data suggest that the AHR may contribute to FSH-stimulated follicle growth, but it may not contribute to the ability of FSH to stimulate steroidogenesis.
Based on lower mRNA expression of
Fshr as well as reduced binding sites for FSH in AHRKO ovaries compared to WT ovaries already shown in previous studies [
5,
8] and the higher FSH levels needed in our experiments for AHRKO follicles to reach WT growth levels, we initially expected that FSH would not be able to stimulate expression of
Fshr to the same degree in AHRKO follicles as it does in WT follicles. However, our data indicate an opposite scenario in which FSH increases expression of
Fshr to a greater level in AHRKO follicles compared to WT follicles. Further, our data indicate that the down-regulation of
Fshr expression upon rising levels of FSH occurred in WT follicles, but not in AHRKO follicles. Because previous studies indicate that FSH can increase its own receptor, leading to increased ability of follicles to respond to FSH [
9,
27‐
30], our data suggest that the reduced levels of
Fshr, and thus, the reduced capacity of AHRKO follicles to respond to FSH are not due to an inability of FSH to stimulate expression of its own receptor.
It is unclear how rising levels of FSH down-regulate expression of the
Fshr in WT follicles, but not in AHRKO follicles. Previous studies indicate that low levels of FSH may favor increased
Fshr expression in granulosa cells, whereas high levels of FSH suppress
Fshr expression in granulosa cells [
29]. It is possible that differences in
Fshr transcript levels in WT and AHRKO follicles treated with rising levels of FSH may be explained by findings suggesting that the AHR is recruited to the
Fshr promoter to transcribe the
Fshr in WT, but not AHRKO follicles [
5]. As FSH levels increase in WT follicles, the high FSH levels may lead to down-regulation of the
Fshr. In AHRKO follicles, however, there is no active AHR; therefore,
Fshr transcription may occur independently of the AHR. The AHR has been suggested to be indispensable for proper transcription of
Cyp19a1 in the ovary [
4]. In our experiments, however,
Cyp19a1 as well as
Inhba transcription followed a very similar pattern in response to FSH in AHRKO follicles compared to WT follicles, suggesting that FSH regulation of
Cyp19a1 and
Inhba may not directly require the presence of the AHR. Instead, they may require proper transcription of
Fshr.
Our experiments used a minimal concentration of 5 IU/mL FSH, which has been shown to be an essential dose for sustaining
in vitro follicle growth [
6,
20]. Because the physiological doses of FSH that are required to promote growth in antral follicles are unknown, our use of a range of doses of FSH allowed us to compare FSH responsiveness in WT and AHRKO follicles. Unexpectedly, follicles were still able to grow to some degree when FSH was omitted from the culture. This finding is supported by previous studies that have demonstrated that FSHR knockout mice can develop pre-antral follicles up to the early antral follicle stage, but not further [
31‐
33]. In addition, follicles cultured with no FSH treatment during the antral phase have limited granulosa cell differentiation in terms of their ability to produce estradiol [
34]. As shown in our experiments, estradiol production and hormone precursors in follicles cultured with no FSH treatment were lower than those cultured with FSH treatment regardless of genotype. Since our
in vitro culture uses a luteinizing hormone (LH)-free culture medium, androgen levels may be the result of constitutive androgen production by the theca cells, which provided sufficient substrate for estradiol production [
34].
Conclusions
In summary, our data indicate that higher levels of FSH are required to stimulate follicle growth in AHRKO follicles compared to WT follicles. In contrast, FSH is able to stimulate expression of steroidogenic factors, Fshr, and Inhba as well as production of some sex steroid hormones to a similar or greater degree in AHRKO follicles compared to WT follicles. Collectively, these data suggest that the AHR may contribute to the ability of FSH to stimulate proper follicle growth, but it may not contribute FSH-stimulated steroidogenesis.
Acknowledgements
The authors acknowledge Sharon Meachum for her technical assistance. This study was supported by NIH R01HD047275 (JAF), the NIEHS ES07326 Research Training Program in Endocrine, Developmental and Reproductive Toxicology (BNK), an Environmental Toxicology Fellowship from the Interdisciplinary Environmental Toxicology Program at UIUC (MSB), and an Eli Lilly Fellowship in Toxicology/Pharmacology for graduate students (TP).
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
IHO designed the experiments, conducted the experiments, and wrote the initial draft of the manuscript. LG, JP, MB, SB, BK, and TP helped conduct the follicle isolations, cultures, and qPCR. BK also helped make the figures. JF helped design the experiments, edited drafts of the manuscript, and helped with data analysis and interpretation. She also obtained funding for the project. All authors read and approved the final manuscript.