GnRH-deficient phenotypes in humans and mice with heterozygous variants in KISS1/Kiss1

Abstract

Context: KISS1 is a candidate gene for GnRH deficiency.

Objective: Our objective was to identify deleterious mutations in KISS1.

Patients and methods: DNA sequencing and assessment of the effects of rare sequence variants (RSV) were conducted in 1025 probands with GnRH-deficient conditions.

Results: Fifteen probands harbored 10 heterozygous RSV in KISS1 seen in less than 1% of control subjects. Of the variants that reside within the mature kisspeptin peptide, p.F117L (but not p.S77I, p.Q82K, p.H90D, or p.P110T) reduces inositol phosphate generation. Of the variants that lie within the coding region but outside the mature peptide, p.G35S and p.C53R (but not p.A129V) are predicted in silico to be deleterious. Of the variants that lie outside the coding region, one (g.1-3659C→T) impairs transcription in vitro, and another (c.1-7C→T) lies within the consensus Kozak sequence. Of five probands tested, four had abnormal baseline LH pulse patterns. In mice, testosterone decreases with heterozygous loss of Kiss1 and Kiss1r alleles (wild-type, 274 ± 99, to double heterozygotes, 69 ± 16 ng/dl; r(2) = 0.13; P = 0.03). Kiss1/Kiss1r double-heterozygote males have shorter anogenital distances (13.0 ± 0.2 vs. 15.6 ± 0.2 mm at P34, P < 0.001), females have longer estrous cycles (7.4 ± 0.2 vs. 5.6 ± 0.2 d, P < 0.01), and mating pairs have decreased litter frequency (0.59 ± 0.09 vs. 0.71 ± 0.06 litters/month, P < 0.04) and size (3.5 ± 0.2 vs. 5.4 ± 0.3 pups/litter, P < 0.001) compared with wild-type mice.

Conclusions: Deleterious, heterozygous RSV in KISS1 exist at a low frequency in GnRH-deficient patients as well as in the general population in presumably normal individuals. As in Kiss1(+/-)/Kiss1r(+/-) mice, heterozygous KISS1 variants in humans may work with other genetic and/or environmental factors to cause abnormal reproductive function.

Fig. 1.

RSV identified in probands. A, Genomic organization of KISS1. Light blue rectangles, noncoding portions of exons; purple rectangles, coding portions of exons; dark blue lines, introns. B, Amino acid sequence of the KISS1 gene product in humans, cow, mouse, and rat. Light blue, signal peptide; blue, mature kisspeptin, with the active C-terminal amino acids in dark blue; line above, putative PEST sequence; asterisks, residues altered by sequence variants. C, Variants identified in subjects, with sequencing chromatographs, nucleotide and amino acid changes, predictions of functional consequences by in silico programs, and activity tested in vitro. ↓↓, Likely damaging; ↓, possibly damaging; ↔, benign; NA, not applicable; ND, not determined.

Fig. 2.

Neuroendocrine profiles of subjects with RSV in KISS1. A, Baseline secretory patterns of LH as determined by frequent sampling (every 10 min) for 12 h in subjects 3, 4, 15, 16, and 17. Arrowheads indicate pulses. B, Gonadotropin responses to 7 d pulsatile GnRH. The response to a single GnRH pulse (time = 0 to +2 h) was monitored each day. Daily sex-steroid levels are shown above the graphs. T, Testosterone, in ng/dl; E2, estradiol, in pg/ml.

Fig. 3.

Characterization of KISS1 variants in vitro. A, IP accumulation induced by WT or F117L kisspeptin decapeptide at doses from 10⁻¹⁰ to 10⁻⁴ m; B, transcriptional activity of WT and g.1-3659C→T kisspeptin regulatory regions. Data from a representative experiment are shown. CPM, Counts per minute; RLU, radioluminescent units; CMV, cytomegalovirus enhancer.

Fig. 4.

Pedigrees of subjects with RSV in KISS1. Arrows indicate probands. HA, Hypothalamic amenorrhea; IHH, idiopathic hypogonadotropic hypogonadism.

Fig. 5.

Phenotypes of male mutant mice. A, Anogenital distance (AGD) in male mice over time of 11 animals were in each group; B, combined weights of testes; C, serum testosterone; D, serum testosterone as a function of number of WT alleles. K^+/−, Kiss1 heterozygote; G^+/− Kiss1r (Gpr54) heterozygote; K^+/−G^+/−, Kiss1 and Kiss1r double heterozygote; K^−/−, Kiss1 homozygote; G^−/−, Kiss1r homozygote; K^−/−G^−/− Kiss1 and Kiss1r double homozygote. Columns and error bars show mean ± sem. *, P < 0.01 compared with WT; †, P < 0.01 compared with K^−/−, as determined by two-way ANOVA; for clarity, not all differences are indicated. Columns in B and C with different letters (a, b, c) are statistically different by one-way ANOVA.

Fig. 6.

Phenotypes of female mutant mice and fertility of mutant mice. A, Percentage of female mice with vaginal opening (VO) over time. B, Length of estrous cycle. Daily vaginal smears were obtained over a period of 2–10 months. Mice in prolonged estrus (>10 d) were excluded. C, Length of estrous phase. D, Average number of litters per month for each mating pair. E, Average number of pups per litter. Genotype abbreviations as in Fig. 5. Columns and error bars show mean ± sem. Numbers indicate cycles or litters evaluated. Columns with different letters (a, b, c) are statistically different by one-way ANOVA. *, P < 0.01 compared with WT by one-way ANOVA.