|Home | About | Journals | Submit | Contact Us | Français|
The luteinizing hormone receptor, LHCGR, is essential for fertility in males and females, and genetic mutations in the receptor have been identified that result in developmental and reproductive defects. We have previously generated and characterized a mouse model (KiLHRD582G) for familial male-limited precocious puberty caused by an activating mutation in the receptor. We demonstrated that the phenotype of the KiLHRD582G male mice is an accurate phenocopy of male patients with activating LHCGR mutations. In this study, we observed that unlike women with activating LHCGR mutations who are normal, female KiLHRD582G mice are infertile. Mice exhibit irregular estrous cyclicity, anovulation, and precocious puberty. A temporal study from 2–24 wk of age indicated elevated levels of progesterone, androstenedione, testosterone, and estradiol and upregulation of several steroidogenic enzyme genes. Ovaries of KiLHRD582G mice exhibited significant pathology with the development of large hemorrhagic cysts as early as 3 wk of age, extensive stromal cell hyperplasia and hypertrophy with luteinization, numerous atretic follicles, and granulosa cell tumors. Ovulation could not be rescued by the addition of exogenous gonadotropins. The body weights of the KiLHRD582G mice were higher than wild-type counterparts, but there was no increase in the body fat composition or metabolic abnormalities such as impaired glucose tolerance and insulin resistance. These studies demonstrate that activating LHCGR mutations do not produce the same phenotype in female mice as in humans and clearly illustrate species differences in the expression and regulation of LHCGR in the ovary, but not in the testis.
The luteinizing hormone receptor (LHCGR), a member of the G protein-coupled receptor family, is essential for fertility . In the ovary, LHCGR are present in theca cells lining the follicle, mature granulosa cells, stromal cells, and luteinized cells. Activation of LHCGR by luteinizing hormone (LH) stimulates androgen production in the theca cells, thereby providing the substrate for conversion to estradiol by follicle stimulating hormone (FSH) induced aromatase in granulosa cells . LHCGR activation is also essential for ovulation and subsequent progesterone production by the corpus luteum (CL) . The canonical signaling pathway mediated by LHCGR is the Gαs/cAMP/protein kinase A pathway. However, LHCGR can also activate additional pathways, including the inositol phosphate/protein kinase C, protein kinase B/Akt, and ERK1/2 pathways [1, 3–7].
An interesting feature of the human LHCGR is the large number of inactivating and activating mutations that have been detected in the gene with varying phenotypic consequences. Inactivating mutations are present throughout the molecule and lead to disorders of sexual differentiation in males and infertility in females. Activating mutations are found exclusively in exon 11 of the LHCGR gene resulting in amino acid substitutions in the transmembrane helices of the receptor protein or in the intracellular loops. The majority of the mutations are clustered in transmembrane helix 6 with aspartic acid at position 578 most commonly mutated to glycine (D578G) [8, 9]. Activating mutations are heterozygous and inherited in an autosomal male-limited pattern resulting in familial male-limited precocious puberty (FMPP). Signs of puberty are apparent in boys by age 4 and are caused by high levels of testosterone in the context of prepubertal LH levels. In contrast, no phenotype has been identified in females. The lack of an apparent phenotype in females with activating LHCGR mutations remains a paradox. Several explanations have been offered including low or absent LHCGR expression in prepubertal girls and the requirement for both LH and FSH to induce puberty .
We have previously reported on a mouse model of FMPP generated by introducing an aspartic acid to glycine mutation at amino acid residue 582 (D582G), which corresponds to the D578G mutation in humans . These mice (KiLHRD582G) are an accurate phenocopy of patients with FMPP as evidenced by precocious puberty, Leydig cell hyperplasia, and elevated testosterone levels. In the current study, we have examined the phenotype of female KiLHRD582G mice. Our results indicate that unlike women, mice with activating mutations exhibit precocious puberty, infertility, elevated steroid hormone levels, polycystic ovaries, and granulosa cell tumors.
The generation and genotyping of KiLHRD582G mice has been described previously . Female KiLHRD582G mice were obtained by breeding male KiLHRD582G mice with female B6129SF1/J hybrid mice (The Jackson Laboratory). Mice were fed a standard laboratory chow (Purina Labdiet 5008) and tap water ad libitum and were maintained in the conventional colony with a 12L:12D cycle. Randomly cycling wild-type (WT) mice were used in all the studies. Mice were euthanized by CO2 asphyxiation, and blood immediately collected by heart puncture between 0900 to 1100. Blood was allowed to clot for 1 h at room temperature, and serum was collected by centrifugation. Desired tissues were collected, cleaned, and either fixed or quick-frozen in liquid nitrogen.
WT and KiLHRD582G mice at 3 wk of age were injected intraperitoneally with 4 international units of equine chorionic gonadotropin (eCG; Sigma Chemicals) to stimulate preovulatory follicle development followed 48 h later with 5 international units of human chorionic gonadotropin (hCG; Ayerst Laboratories) to stimulate ovulation and luteinization. Oocytes were collected by flushing the oviducts 24 h after hCG treatment. Ovaries were collected and fixed in 10% buffered formalin. All the animal studies were approved by the Institutional Animal Care and Use Committee at Southern Illinois University.
Vaginal openings were checked daily starting at Postnatal Day 10. Once the vaginal opening was detected, vaginal smears were performed daily for a period of 21 days. The vaginal smears were stained with 0.1% methylene blue and examined under the microscope. The stage of the estrous cycle was determined by the presence of distinct cell types. Diestrus consisted predominantly of leukocytes, proestrus showed both leukocytes and nucleated epithelial cells in approximately equal numbers, estrus was characterized by cornified squamous epithelial cells, and metestrus consisted of equal numbers of leukocytes and cornified epithelial cells with translucent nuclei.
Serum levels of androstenedione were measured using an enzyme-linked immunosorbent assay (ELISA) kit (Neogen Corporation). The lower detection limit was 0.01 ng/ml. Testosterone and progesterone levels were measured by enzyme immunoassay kits (Cayman Chemical Company) following the manufacturer's instructions. The lower limit of detection of the testosterone assay was 3.9 pg/ml and of the progesterone assay was 10 pg/ml. Estradiol levels were measured using the mouse/rat ELISA kit (Calbiotech) with a lower detection limit of 3 pg/ml. Serum LH and FSH levels were measured by the University of Virginia Center or Research in Reproduction Ligand Assay and Analysis Core. The reportable range for LH was 0.08–37.4 ng/ml and for FSH was 1.8–65.4 ng/ml.
RNA was extracted from homogenized ovaries using TRIzol (Invitrogen) according to the manufacturer's instructions and quantified by determination of absorbance at 260nm. Complementary DNA was synthesized using the RETROscript kit (Ambion) and amplified by PCR. The primers for quantitative RT-PCR were designed such that one of the primers spanned an exon-exon junction ensuring amplification of only the cDNA. The sequence of the primers used is listed in Supplemental Table S1 (Supplemental Data are available online at www.biolreprod.org). Ribosomal protein L19 (Rpl19) was used as the internal control because its expression did not change significantly with age or genotype. A template negative control and a calibrator sample, prepared by mixing equal amounts of cDNA from all the samples, were also analyzed. Amplification was performed by 3 min denaturation at 95°C followed by 40 cycles for 15 sec at 95°C and 1 min of annealing at 61°C. The difference in the CT value between the gene of interest and Rpl19 was defined as ΔCT. The difference in the ΔCT between the gene of interest and the calibrator was defined as ΔΔCT. The relative quantity was determined by 2−ΔΔCT .
Tissue-specific expression of WT and D582G Lhcgr was determined by RT-PCR. RNA from various tissues was isolated using Trizol and treated with DNase 1 (DNA-free kit; Ambion) for 30 min at 37°C to remove potentially contaminating DNA. Complementary DNA was synthesized using the RETROscript kit (Ambion) and amplified by PCR with primers specific for Lhcgr or Gapdh as described previously . Annealing was performed at 55°C for 1 min, and PCR reactions were performed for 35 cycles. The PCR primers for Lhcgr generate a 600-bp product containing the mutated sequence in exon 11. A portion of the 600-bp PCR product was digested with Hinf1 to detect the mutated allele and generate bands of 485 and 115 bp.
Ovaries from WT and KiLHRD582G mice were fixed in 10% buffered formalin and embedded in paraffin. Sections (5 μm) were stained with hematoxylin and eosin (H&E) to visualize the morphology or were used to perform immunohistochemistry. Ovary sections were blocked with 1.5% normal goat serum for 1 h at room temperature followed by incubation at 4°C overnight with steroidogenic acute regulatory protein (StAR), LHCGR, or 17β-hydroxysteroid dehydrogenase type VII (HSD17B7) antibody at 1:1500, 1:500, and 1:1000 dilutions, respectively. A negative control was performed by incubation with preimmune rabbit serum. Biotinylated secondary antibody was applied to sections followed by incubation with the avidin-biotin complex method (ABC reagent; Vector Labs). The antibody-antigen complexes were visualized as brown deposits by incubation with diaminobenzidine and counterstained with hematoxylin. For detection of forkhead transcription factor, FOXL2, antigen retrieval was performed by boiling the slides in 10 mM citric acid, pH 6.0, for 10 min. Sections were blocked with tyramide signal amplification block and streptavidin-horseradish peroxidase (TSA kit; Perkin Elmer) was used instead of the avidin-biotin complex reagent. Slides were examined using a Leica DM5000B microscope, and images were obtained using a Retiga 2000R camera and QImaging software. Antibodies were kindly provided by: StAR (Dr. Dale Buck Hales, Southern Illinois University), LHCGR (Dr. Asgi Fasleabas, Michigan State University), HSD17B7 (Dr. Geula Gibori, University of Illinois, Chicago), and FOXL2 antibody was generated by Dr. Reiner Veitia, Université Paris, and provided by Dr. Buffy Ellsworth, Southern Illinois University, with permission from Dr. Veitia.
WT and KiLHRD582G mice at 16 wk of age were fasted for 12 h in a clean cage prior to the test. Blood samples were obtained from the tail vein. Blood glucose levels were measured at 0, 15, 30, 45, 60, and 120 min after glucose administration (1.5 g/kg of 20% D-glucose injected intraperitoneally) using the One Touch glucometer (Life Scan Europe). Blood samples were also collected at 0 and 15 min, and plasma obtained by centrifugation at 4000 rpm for 15 min followed by 12000 rpm for 5min. Five microliters of plasma from each sample was used for insulin assay using the Ultra Sensitive Mouse Insulin ELISA kit (Crystal Chem) following the manufacturer's instructions.
Body composition of 18-wk-old WT and KiLHRD582G mice was analyzed with a rodent quantitative nuclear magnetic resonance apparatus (EchoMRI Whole Body Composition Analyzer; Echo Medical Systems). The mice were placed without anesthesia in a plexiglass tube that was inserted into the nuclear magnetic resonance apparatus for body composition analysis . The procedure takes less than 60 sec to perform. Body fat and lean tissue were measured. Percent fat and lean tissue was determined by dividing the mass of fat or lean tissue by the body weight.
Data are expressed as mean ± SEM. Statistical significance of differences between genotypes was determined by the Student t-test using the Prism 5 software program (GraphPad Software, Inc.). Outliers were identified with the Grubbs test using the online GraphPad software (http:// www.graphpad.com/quickcalcs/Grubbs1.cfm). P < 0.05 was considered statistically significant.
Ovarian expression of the WT and mutant allele was confirmed by RT-PCR (Supplemental Fig. S1). Expression of Lhcgr from the WT allele was higher than the mutant allele. Lower levels of expression of both alleles were detected in the uterus and brain, but not in the adrenal gland. To determine the age of puberty, mice were examined daily for vaginal opening from Postnatal Day 10. Vaginal opening was detected at a significantly earlier age of 15 ± 0.3 days (n = 8) in KiLHRD582G mice compared to 29 ± 0.6 days (n = 9) in WT mice (P < 0.001). WT mice exhibited normal 4–5 days of estrous cycle (Fig. 1A) while KiLHRD582G mice exhibited irregular estrous cyclicity, with several days in estrus, followed by an abnormal stage (other, O) with a mixture of leukocytes, nucleated, squamous, and cornified epithelial cells (Fig. 1, B and C). Mice were infertile and produced no litter when bred with male B6129SF1/J hybrid mice over a period of 6 mo.
Examination of the reproductive tract showed enlarged ovaries and uterus in KiLHRD582G mice compared to WT mice (Fig. 2, A and B). Large ovarian cysts were apparent in KiLHRD582G mice (Fig. 2B). Ovarian weights were increased in KiLHRD582G mice as early as 2 wk of age and remained elevated at 24 wk (Fig. 2C). The dip in ovarian weight at 12 wk of age is likely due to the decrease in the number of blood-filled cysts (Fig. 5). Uterine weights of KiLHRD582G mice increased dramatically at 3 wk of age (Fig. 2D) primarily due to accumulation of fluid in the uterine lumen coincident with the continuous state of estrus as shown in Figure 1. Although still elevated compared to WT mice, the uterine weights of KiLHRD582G mice decrease at 4–6 wk and are similar to WT mice at 12 and 24 wk of age. At these ages, the mice are not in estrus, and the uterine lumen is not fluid filled although it remains enlarged. Histological analysis of the uterus at 3 wk of age shows an enlarged lumen and lack of endometrial glands in KiLHRD582G mice compared to WT mice (Fig. 2, E and F). The finding of precocious puberty and irregular estrous cyclicity suggested altered levels of steroid hormones, which was confirmed by measuring serum levels of steroid hormones (Fig. 3). Serum progesterone level in 2-wk-old KiLHRD582G mice was significantly lower than in WT mice but significantly increased from 4 to 24 wk. Serum androstenedione, testosterone, and estradiol levels in KiLHRD582G mice were significantly higher than WT mice at prepubertal and adult ages. Due to negative feedback of the steroids, LH and FSH levels in KiLHRD582G mice were below the limit of detection at all ages except for FSH levels at 2 wk of age.
A temporal profile of steroidogenic gene expression was determined using total ovarian RNA (Fig. 4). Lhcgr was upregulated significantly in KiLHRD582G mice compared to WT counterparts at all ages. Follicle stimulating hormone receptor (Fshr), Star, cytochrome P450 side-chain cleavage (Cyp11a1), 3β-hydroxysteroid dehydrogenase I (Hsd3b1), and aromatase (Cyp19a1) were upregulated in the KiLHRD582G mice at prepubertal ages, which was consistent with the premature increase in steroid hormone levels in these animals. Fshr and Cyp19a1 showed a similar temporal pattern of expression in the KiLHRD582G mice. 17β-Hydroxysteroid dehydrogenase type I (Hsd17b1), is expressed in granulosa cells of developing follicles where it converts estrone to estradiol, and its expression was not significantly different between genotypes. However, Hsd17b7 is a luteal cell-specific marker responsible for the conversion of estrone to estradiol in luteal cells, and its expression was upregulated at younger ages in KiLHRD582G mice. Cytochrome P450 17α-hydroxylase/c17-20 lyase (Cyp17a1) showed an interesting pattern of expression in KiLHRD582G mice with greatly elevated levels at 3 wk of age followed by a decrease at 6 wk and increase at 12 and 24 wk of age.
Ovarian sections of 2- to 24-wk-old mice were examined (Fig. 5). Ovaries from WT mice appeared normal with various stages of follicle growth apparent. Corpora lutea could be detected starting at 6 wk of age. Ovaries from KiLHRD582G mice appeared normal at 2 wk of age with several developing follicles similar to the WT ovary. Degenerating follicles and hemorrhagic cysts were apparent in the ovaries of KiLHRD582G mice starting at 3 wk of age and remained at the older ages examined. Follicles did not progress beyond the preantral stage (see higher magnification image at 4 wk). No CLs were observed in KiLHRD582G ovaries, indicating that ovulation did not occur. This result was consistent with the abnormal estrous cyclicity in KiLHRD582G mice. At 6 and 12 wk of age, all KiLHRD582G ovaries exhibited mild, infrequent (grade 2) or moderate, frequent (grade 3) stromal cell hypertrophy and hyperplasia with luteinization, clearly seen in the higher magnification images of the boxed areas. Numerous atretic follicles were present, and the granulosa cells in these follicles appeared luteinized. In 24-wk-old animals, in addition to the luteinized stromal cell hypertrophy/hyperplasia, granulosa cell tumors were evident in 50% of the mice and tubulostromal hyperplasia with stromal cell luteinization in 25% of the mice. The granulosa cells in the tumors were well differentiated and not luteinized. Immunohistochemistry of ovaries from 24-wk-old KiLHRD582G mice showed that the hyperplastic stromal cells and the granulosa cells in atretic follicles stained positively for LHCGR, StAR, and HSD17B7, confirming that these cells were steroidogenic and luteinized (Fig. 6). In WT ovaries, expression of LHCGR and StAR was detected in follicles, CL, and stroma while expression of HSD17B7 was limited to the CL, indicating the stroma was not luteinized. Staining with FOXL2, a granulosa cell marker, showed specific nuclear staining in the granulosa cells of follicles in WT mice and granulosa cell tumors of the KiLHRD582G mice.
WT and KiLHRD582G mice were superovulated at 3 wk of age. At this age, the ovaries of the KiLHRD582G mice contain numerous preantral follicles although a few cystic follicles are also seen. WT mice yielded an average of 23 ± 2.3 ovulated oocytes per mouse. In contrast, no oocytes were retrieved from the oviducts of KiLHRD582G mice (Fig. 7A). Histological analysis of the ovaries showed the presence of CLs in WT mice indicating ovulation (Fig. 7B). Preovulatory follicles with oocytes were present in the ovaries of KiLHRD582G mice, indicating that growth of the follicles occurred but the follicles did not rupture to form CLs (Fig. 7C).
As shown in Figure 8A, the total body weights of KiLHRD582G animals were significantly higher compared to WT mice as early as 3 wk of age. Because of this increase in body weight coupled with hyperandrogenism, irregular estrous cyclicity, anovulation, and polycystic ovaries seen in the KiLHRD582G mice, we wanted to determine if there were alterations in body fat composition and insulin resistance as seen in some experimentally induced hyperandrogenic rodent models of polycystic ovarian syndrome (PCOS) [13–15]. Body composition was determined in 18-wk-old mice. No difference was found in percent body fat between WT (20.1 ± 1.9, n = 9) and KiLHRD582G mice (20.2 ± 1.3, n = 4) or lean tissue (70.7 ± 1.5 for WT and 68.0 ± 1.1 for KiLHRD582G mice). The intraperitoneal glucose tolerance test performed at 16 wk of age did not show differences in the blood glucose levels (Fig. 8B) and the area under the curve, calculated by trapezoid analysis, was similar between the two genotypes (Fig. 8C). Plasma insulin levels measured at 0 and 15 min after glucose injection were not significantly different between the genotypes (Fig. 8D). These data suggest that the KiLHRD582G mice do not exhibit metabolic abnormalities such as glucose intolerance or insulin resistance.
In this study, we demonstrate that in a mouse model of FMPP, caused by an activating D582G mutation in the Lhcgr gene, female mice exhibit a phenotype of precocious puberty, infertility, ovarian cysts, elevated steroid hormone levels, stromal cell hyperplasia with luteinization, and granulosa cell tumors. In contrast to the male phenotype, which is an accurate phenocopy of the human condition , the female phenotype is distinct from women with activating mutations in the LHCGR gene who do not exhibit any ovarian dysfunction. Mothers and sisters of boys with FMPP who are carriers of the mutation revealed no infertility or other problems [16–18]. This is puzzling considering that constitutive activation of LHCGR in theca cells should result in gonadotropin independent production of ovarian androgens resulting in a condition similar to PCOS. Several explanations have been offered for the lack of phenotype, including low or absent LHCGR expression in prepubertal girls, the requirement for activation of both LHCGR and FSHR for puberty, and less efficient androgen synthesis in theca cells compared to Leydig cells [9, 18]. Studies of women with FSHβ gene mutations resulting in lack of bioactive FSH have provided another possible explanation into the lack of phenotype with activating LHCGR mutations and suggested a role for FSH in androgen biosynthesis. One patient with a compound heterozygous mutation in the FSHβ gene did not have measurable levels of immunoreactive or bioactive FSH. Although her LH levels were elevated, her testosterone levels were normal , leading to the hypothesis that FSH was necessary for LH-induced theca cell androgen synthesis. The mean LH level and pulse amplitude was elevated in this patient similar to women with PCOS. Unlike women with PCOS, treatment with hCG resulted in an increase in testosterone, but treatment with FSH and hCG resulted in a more dramatic increase in testosterone. This indicated that FSH is required for LH-mediated androgen synthesis perhaps by inducing LHCGR or growth factors . Additionally, cell culture studies indicate that the basal activity of D578G mutation in human LHCGR produces a modest 3-fold to 5-fold increase in basal cAMP activity and perhaps does not activate the receptor beyond the prepubertal level. The mutant receptor can still respond to hCG, suggesting that it will still be able to respond to high concentrations of LH required for ovulation and negative feedback systems regulating ovarian function are likely intact [9, 18]. In contrast, the basal cAMP levels in D582G mLHCGR is elevated by 23-fold although it is still able to respond to hCG . In rodents, synthesis and secretion of gonadotropins occurs by the end of gestation . FSH binding and FSH-stimulated cAMP production occur at about Postnatal Day 4 and precede the development of functional LHCGR that occurs at Postnatal Day 7 [20, 21]. Therefore, our results with KiLHRD582G mice suggest that the differences in the mouse and human phenotypes represent genuine species differences.
We had previously confirmed by genotyping tail biopsies that the neomycin cassette from the targeting vector had been deleted . To ensure that there was no residual neomycin expression in the ovaries, which could potentially alter Lhcgr expression and cause the infertility phenotype, we examined neomycin expression in the ovaries by RT-PCR. No expression was detected (data not shown), indicating that the mutant LHCGR was responsible for the phenotype of the KiLHRD582G mice.
The constitutive activity of LHCGR in theca cells results in an elevation in the levels of both androgens and estrogen in KiLHRD582G mice as early as 2 wk of age. Androstenedione synthesized by the theca cells is converted to testosterone by HSD17B1 in granulosa cells . Androstenedione is also the substrate for conversion to estrone by CYP19A1 and subsequent conversion to estradiol by HSD17B1 in the granulosa cells. Ovarian gene expression shows elevated levels of Lhcgr, Fshr, Star, Cyp11a1, Hsd3b1, Cyp17a1, and Cyp19a1mRNA at 3 wk of age consistent with the increased steroid hormone levels. Interpretation of these results is limited by the fact that the entire ovary was used in these measurements and the ovarian morphology between the genotypes is different. Therefore, it is difficult to ascertain if the differences in expression are due to increased expression per cell or due to increased numbers of theca and granulosa cells in the KiLHRD582G mice at this age. It is also not possible to distinguish if upregulation of Lhcgr is due to an equivalent increase in the expression of WT and mutant transcripts. The RT-PCR data in Supplemental Figure S1 suggests that the WT allele is expressed at a higher level than the mutant allele. Despite the higher expression levels, it is unlikely that the WT receptor can be sufficiently activated to contribute significantly to the phenotype of the KiLHRD582G mice because of the greatly suppressed LH levels in the mutant mice.
At older ages, the steroid hormone levels remain elevated although there are few intact follicles in the ovary of KiLHRD582G mice by 6 wk of age. In adult mice, the luteinized, hyperplastic stroma is the likely source of the hormones. Consistent with this hypothesis is the detection of LHCGR, StAR, and HSD17B7 in the ovarian stroma and the upregulation of mRNA levels of these genes of KiLHRD582G mice. In WT mice, LHCGR and StAR staining can be detected in the follicles, stroma, and CL while HSD17B7 staining is only observed in the CL, indicating that the stroma is not luteinized. The upregulation of Hsd17b7, which converts estrone to estradiol only in luteal cells, maintains the elevated estradiol levels in the context of decreased Cyp19a1 and Hsd17b1 expression due to follicle loss.
Infertility in KiLHRD582G mice is primarily due to constant high levels of steroid hormones resulting in abnormal estrous cycles, lack of antral follicle development, and chronic anovulation. At 2 wk of age, the ovary of KiLHRD582G mice is similar to that of WT animals and developing follicles are apparent. The expression of the constitutively active LHCGR in the KiLHRD582G ovaries produces sufficient androgens and estradiol to cause the precocious vaginal opening. At 3 wk of age, ovarian cysts begin to appear and preantral follicles are present in KiLHRD582G mice. As a result of negative feedback from the elevated estradiol levels, there is insufficient FSH to support follicular development beyond the preantral stage. Previous studies have shown that chronic stimulation of LHCGR and intraovarian actions of estradiol receptor β are required for hemorrhagic cyst formation . However, Fshb-null and Fshr-null mice do not develop hemorrhagic cysts despite elevated plasma LH levels, suggesting that elevated estradiol is required for hemorrhagic cyst formation [24–26]. Consistent with this observation is the presence of hemorrhagic cysts in the ovaries of FSHβ-overexpressing mice and in mice expressing a constitutively active FSHR that have elevated estradiol, but normal or low LH levels [27, 28]. A striking phenotype is the luteinized stromal or interstitial cell hyperplasia/hypertrophy that is present in all KiLHRD582G mice starting at 6 wk of age. It has been suggested that in rodents, luteinized interstitial cell hyperplasia is a consequence of increased follicular atresia due to apoptosis of the follicle and surrounding granulosa cells, leaving behind the theca cells that remain steroidogenic .
The inability to rescue the anovulatory phenotype following exogenous gonadotropin stimulation of 3-wk-old KiLHRD582G mice is interesting. Follicles developed to the preovulatory stage, indicating that FSHR function was not affected. The results suggest that in the ovaries of the heterozygous KiLHRD582G mice, the D582G LHCGR inhibits the signaling of the WT receptor sufficiently to prevent ovulation. We have previously shown that in cell culture, the D582G LHCGR was responsive to additional hCG stimulation . However, this superovulation study suggests that the mutant receptor is unable to or responds poorly to additional hCG in vivo. Previous studies have shown that ovulation requires the LH-dependent activation of both the Gαs and Gαq/11 pathways in granulosa cells [3, 30]. The Gαs/cAMP-dependent activation of the EGF network and ERK1/2 cascade is required for oocyte maturation and cumulus expansion whereas follicular rupture is dependent on both Gαs and Gαq/11 [3, 30]. Shinozaki et al.  demonstrated that when a constitutively active hLHR (L457R), that was unresponsive to additional hormonal stimulation, was coexpressed with the WT hLHR there was no decrease in the cell surface expression of the WT receptor. However, coexpression caused an attenuation of the hCG/Gαs-stimulated cAMP production by WT hLHR. This study also demonstrated that the attenuation was due to the activation of phosphodiesterase (PDE)4D3. Whether a similar mechanism of cAMP attenuation occurs in vivo in the granulosa cells expressing both WT and D582G LHCGR remains to be determined. We have also not examined if the D582G LHCGR constitutively activates Gαq/11resulting in elevated basal levels of inositol phosphate (IP). However previous studies have shown that the corresponding mutation in hLHR does not increase the basal level of IP, but responds to hCG with an increase in IP production [32, 33].
The phenotype of the KiLHRD582G mice is similar to transgenic mice overexpressing LH (LHβCTP) or human chorionic gonadotropin (hCG) [34–36] and the yoked hCG-LH receptor complex . A common feature in these models is the detection of precocious puberty, elevated steroid hormone levels, lack of estrous cyclicity, hemorrhagic cysts, and infertility. The LHβCTP mice also develop stromal cell hyperplasia with luteinization [23, 38] similar to the KiLHRD582G mice, suggesting that chronic activation of the LHCGR is the cause of this phenotype. However, differences also exist between the mouse models. The anovulatory phenotype of the LHβCTP can be rescued by the administration of exogenous gonadotropins . Additionally, the LHβCTP mice develop granulosa cell tumors with 100% penetrance only in the genetic background of CF1 mice . KiLHRD582G mice also develop granulosa cell tumors in the mixed B6129SF1/J background, but only in about 50% of the mice. The hCG-overexpressing mice develop luteomas and subsequently teratomas [36, 41] and thecomas , which are absent in the KiLHRD582G mice. Both LHβCTP and hCG-overexpressing mice develop mammary gland tumors and pituitary adenomas in older mice of 10–12 mo of age [36, 42]. We do not know if KiLHRD582G mice would exhibit a similar phenotype because we have not examined mice older than 6 mo of age. No apparent morphological changes in the mammary gland or pituitary have been noted in the younger mice. The metabolic phenotype of the KiLHRD582G mice is also different from the LHβCTP and hCG-overexpressing mice. The increase in body weight of KiLHRD582G mice compared to WT is significant, but the difference is not as great as that seen in the LHβCTP or hCG-overexpressing mice [36, 43]. KiLHRD582G mice did not show changes in body fat composition or metabolic abnormalities such as impaired glucose tolerance and insulin resistance. This is in contrast to the LHβCTP mice that had increased body fat and serum leptin, but not insulin resistance . The differences in the phenotypes between the overexpressing models and KiLHRD582G mice are likely due to the high levels of LH/CG secreted by transgenes that were expressed ubiquitously under the control of promoters that do not mimic the spatial or temporal expression of LH.
In summary, female KiLHRD582G mice exhibit a distinct reproductive phenotype with pathological changes in ovarian function that is not seen in women with activating LHCGR mutations. The phenotype of the female KiLHRD582G mice are similar to that exhibited by mice overexpressing LH or hCG. Considering that the male KiLHRD582G mice exhibit the same phenotype as boys with FMPP, these studies indicate major species differences in LHCGR ovarian function.
We thank Drs. April Strader and Ping Zhao for assistance with the glucose tolerance test, Dr. Buffy Ellsworth and Jyoti Kapali for assistance with the FOXL2 immunohistochemistry, and Dr. James MacLean for the generous gift of Rpl19 primers.
1Supported by NIH grant HD044119 (to P.N.). The Ligand Assay and Analysis Core at the University of Virginia Center for Research in Reproduction is funded by the Eunice Kennedy Shriver National Institute of Child Health and Human Development/NIH Grant U54-HD28934. Presented in part at the 45th Annual Meeting of the Society for the Study of Reproduction, 12–15 August 2012, State College, Pennsylvania.