Hypogonadotropic hypogonadism in mice lacking a functional Kiss1 gene

Abstract

The G protein-coupled receptor GPR54 (AXOR12, OT7T175) is central to acquisition of reproductive competency in mammals. Peptide ligands (kisspeptins) for this receptor are encoded by the Kiss1 gene, and administration of exogenous kisspeptins stimulates hypothalamic gonadotropin-releasing hormone (GnRH) release in several species, including humans. To establish that kisspeptins are the authentic agonists of GPR54 in vivo and to determine whether these ligands have additional physiological functions we have generated mice with a targeted disruption of the Kiss1 gene. Kiss1-null mice are viable and healthy with no apparent abnormalities but fail to undergo sexual maturation. Mutant female mice do not progress through the estrous cycle, have thread-like uteri and small ovaries, and do not produce mature Graffian follicles. Mutant males have small testes, and spermatogenesis arrests mainly at the early haploid spermatid stage. Both sexes have low circulating gonadotropin (luteinizing hormone and follicle-stimulating hormone) and sex steroid (beta-estradiol or testosterone) hormone levels. Migration of GnRH neurons into the hypothalamus appears normal with appropriate axonal connections to the median eminence and total GnRH content. The hypothalamic-pituitary axis is functional in these mice as shown by robust luteinizing hormone secretion after peripheral administration of kisspeptin. The virtually identical phenotype of Gpr54- and Kiss1-null mice provides direct proof that kisspeptins are the true physiological ligand for the GPR54 receptor in vivo. Kiss1 also does not seem to play a vital role in any other physiological processes other than activation of the hypothalamic-pituitary-gonadal axis, and loss of Kiss1 cannot be overcome by compensatory mechanisms.

Fig. 1.

Kiss1 gene targeting. (A) Gene targeting strategy. Kiss1 exons are indicated as shaded boxes with the coding sequence in black. The targeting vector replaces the complete Kiss1 coding region with an IRES-LacZ sequence. The location of primers used to confirm the correct targeting event and loss of Kiss1 coding sequence in the mutant mice are shown along with the size of the PCR products. (B) PCR analysis of Kiss1 locus in mutant mice. The identity of the PCR products was confirmed by restriction enzyme digestion. X, XcmI; E, EcoRI; M, Invitrogen 1-kb DNA marker. (C) RT-PCR of Kiss1gene expression confirming the null allele. (D) Immunohistochemical detection of KISS1 neurons in the ARC of the hypothalamus. (Scale bars: 300 μm.) (E) β-galactosidase activity in the ARC of mutant mice. (Scale bars: 300 μm.)

Fig. 2.

Reproductive system defects in female Kiss1^tm1PTL-null mice. (A) Reduced ovary size and thread-like uterus in mutant mice. (Scale bar: 1 mm.) (B and C) Histology of ovaries showing larger number of atretic follicles (arrows) in mutant mice (C) and absence of late-stage follicle maturation (f) present in wild-type mice. (Scale bars: 100 μm.) (D and E) Histology of uterus showing full development of glands (arrows) in the endometrial layer in wild type (D) in contrast to mutant mice (E). L, uterine lumen. (Scale bars: 25 μm.)

Fig. 3.

Reproductive system defects in male Kiss1^tm1PTL-null mice. (A) Reduced testes size in mutant mice. (Scale bar: 0.25 cm.) (B) Reduced development of seminal vesicle in mutant mice. (Scale bar: 0.25 cm.) (C and D) Histology of seminiferous tubules showing intact spermatogenesis and mature sperm in wild-type mice (C) and impaired spermatogenesis in mutant mice (D). (Scale bars: 50 μm.) (E and F) Histology of epididymis showing sperm in the lumen (l) in wild-type mice (E) and absence of sperm in the lumen in mutant mice (F). (Scale bars: 50 μm.) (G and H) Histology of adrenal gland showing retention of fetal zone X (arrow) in mutant male mice (H) and absence of this zone in mature male wild-type mice (G). (Scale bars: 300 μm.)

Fig. 4.

Hormone profiles of Kiss1^tm1PTL-null mice. (A) Plasma free testosterone levels in male mice. ∗∗, Testosterone concentration was below the limit of detection in the mutant mice. (B) Plasma 17-β-estradiol levels in female mice. Mutant mice had 17-β-estradiol levels close to the background limit of the assay (8 pg/ml). (C) Plasma FSH levels. Both male and female mutant mice showed significantly lower FSH levels than age-matched wild-type animals (P < 0.001, Student's t test). (D) Plasma LH levels. Mutant male mice showed significantly lower LH levels (P = 0.002) than age-matched wild-type males. In females, Kiss1^tm1PTL-null mice had lower plasma LH levels than proestrus (Pro) wild-type animals (statistically significant differences are shown, one-way ANOVA with Student–Newman–Keuls test). The number of mice used for each analysis is indicated in each histogram. Met/Di, metestrus and diestrus; Es, estrous.

Fig. 5.

GnRH neurons in the hypothalamus of Kiss1^tm1PTL-null mice and responses to kisspeptin injection. Photomicrographs of 50-μm-thick coronal sections showing GnRH immunoreactive neurons in the hypothalamus of wild-type (A and C) and mutant (B and D) mice. (A and B) GnRH-positive cell bodies in the preoptic region at the level of the organum vasculosum laminae terminalis. (Scale bar: 100 μm.) Frames in top corners are higher magnifications of respective dotted line squared areas. (Scale bar: 50 μm.) (C and D) GnRH-positive axonal projections and nerve terminals in the median eminence. (Scale bar: 50 μm.) 3V, third ventricle; ON, optic nerve. (E) GnRH content in hypothalami from both sexes. (F) Stimulation of LH release by kisspeptin-10 in Kiss1^tm1PTL-null mice. Kiss1-null (−/−) or wild-type (+/+) female mice at diestrus were injected i.p. with vehicle (PBS) or 1 nmol of kisspeptin-10 in PBS and killed after 30 min, and serum was measured for LH. a, b, c, and d indicate values significantly different from each other (P < 0.05, n = 6 per group, one-way ANOVA followed by Student–Newman–Keuls test).