Abstract
Over the last 10 years, kisspeptins - peptide products of varying lengths encoded by the KISS1 gene - have been found to be key regulators of normal reproductive function throughout life in animals and humans. By activating the kisspeptin receptor [previously known as orphan G protein-coupled receptor 54 (GPR54)], they elicit an effect on the central gonadotropin-releasing hormone neurons. Administration of kisspeptin by either the subcutaneous or intravenous route potently stimulates endogenous gonadotropin hormone release in healthy men and women as well as in animals. Kisspeptin also stimulates endogenous release of gonadotropins in subfertile as well as healthy volunteers, and therefore it has potential as a novel therapeutic agent in reproductive disorders. Further human studies have shown that chronic, high-dose administration of kisspeptin causes desensitisation with rapid subsequent suppression of the hypothalamic-pituitary-gonadal axis, and therefore high-dose long-acting analogues may have a clinical role in treating sex hormone-dependent malignancies. By further elucidating the intricacies and mechanisms of the kisspeptin signalling system, and the tissues it acts on during different phases of the reproductive timeline (including during puberty, fertility, pregnancy and menopause), pharmacologic analogues could become clinically useful.
Introduction
Lee et al. [1] first identified the KISS1 gene in 1996 in Hershey, Pa., USA, naming it after a famous confectionary brand of their locality. The early literature of the gene, its receptor orphan G protein-coupled receptor 54 (GPR54; subsequently re-named kisspeptin receptor) and protein product (first named metastin, now re-named kisspeptin) focussed on oncological pathways after the observation that kisspeptin inhibited the metastatic potential of melanoma cell lines [1,2]. However, its potential role in the reproductive pathway was suggested when expression of the receptor was found in hormonally responsive tissues including the pituitary and placenta [3], and circulating plasma levels were found to be significantly increased in healthy pregnant women [4]. Furthermore, the kisspeptin receptor has also been shown to be expressed in other peripheral tissues including the gonads, adipose fat and pancreas [3,5].
The KISS1 gene encodes a number of peptide products, now collectively known as kisspeptins, which activate the kisspeptin receptor [3]. The initial polypeptide synthesised is 145 amino acids in length and is then cleaved to active shorter lengths. The nomenclature represents the number of amino acids: kisspeptin-54, -14, -13, and -10, and all lengths share the same C-terminal decapeptide sequence [3]. Kisspeptin-54 is the most abundant circulating isoform in humans, but both kisspeptin-10 and -54, when administered to humans and animals, have been shown to be potent stimulators of endogenous gonadotropin hormone secretion, albeit with slightly different pharmacokinetic properties [6,7,8,9,10,11]. Data from animal studies, however, shows that when pre-administered with a gonadotropin-releasing hormone (GnRH) antagonist, the stimulatory effect of kisspeptin is completely inhibited [12]. In light of such results, the historical understanding of the feedback mechanism that regulates the hypothalamic-pituitary-gonadal axis was revised: kisspeptin acts upstream of the GnRH neurons to drive the subsequent hormonal cascade, whereby the anterior pituitary becomes stimulated to release luteinising hormone (LH) and follicle-stimulating hormone (FSH), which then act on the gonads to release sex steroids (fig. 1). Kisspeptin, GnRH and LH/FSH are all secreted in a pulsatile manner. In a study in midpubertal female rhesus monkeys, it was found that a kisspeptin pulse was released approximately every 60 min, and almost 75% of such pulses correlated with a pulse in GnRH secretion [13], and in mice, the genomic control of GnRH pulse generation in hypothalamic neurones is reviewed in this issue [14]. There appear to be two populations of kisspeptin neurons in the human brain, one which also co-expresses neurokinin B and dynorphin (so-called KNDy neurons) and resides in the infundibular (also termed arcuate) nucleus of the hypothalamus, and one which is present in the pre-optic area; both populations project to GnRH neurons in the mediobasal hypothalamus (fig. 1) [15,16,17]. The exact circuitry is unknown, and there is age and sexual dimorphism as well as significant species variation, but recent evidence from immunolocalisation studies in post-mortem specimens suggest there are axo-somatic, axo-dendritic and axo-axonal contacts between kisspeptin and GnRH neurons [17,18,19,20].
Whilst infusion of kisspeptin to humans and animals acutely stimulates gonadotropin release, chronic, high-dose administration of kisspeptin can lead to desensitisation (tachyphylaxis) with rapid suppression of the hypothalamic-pituitary-gonadal axis and inhibition of gonadotropin secretion. This desensitisation appears to occur at the level of the kisspeptin receptor rather than the GnRH neurons or pituitary [8,21].
Potential Clinical Uses of Kisspeptin
Can Kisspeptin Stimulate GnRH Release to Trigger Puberty?
It was known for many years that a number of environmental cues, which include energy status and overall health as well as predisposing genetic factors, influence the timing of pubertal onset. Puberty is the final event of a transition, whereby having been quiescent in childhood after fetal and neonatal development, the amplitude and frequency of GnRH pulses sufficiently increases in response to reduced inhibitory and increased stimulatory tone on the hypothalamic GnRH neurons. The subsequent release of gonadotropins from the anterior pituitary, and the resultant secretion of sex steroids from the gonads drive pubertal development. However, it is only since 2003 that the finer detail of this pathway, and the crucial role of kisspeptin, has been elucidated as a result of improved molecular techniques and the study of patients and families with idiopathic hypogonadotropic hypogonadism (IHH).
Two original articles, published in short succession by independent groups, described pathological mutations of the gene encoding the kisspeptin receptor in individuals with IHH who failed to enter puberty spontaneously but did so normally when administered GnRH agonists. The first publication, a seminal report by de Roux et al. [22], used genome-mapping techniques to demonstrate a large deletion in the kisspeptin receptor gene on chromosome 19p13 that resulted in the production of a truncated and non-functional receptor in 5 individuals in a consanguineous family with IHH. Seminara et al. [23] subsequently published their series including a family with numerous affected members in the same generation with IHH suggesting autosomal recessive inheritance. Linkage analysis mapped the region to the location containing the kisspeptin receptor gene and only affected members were found to be homozygous for a point mutation in exon 3 (L148S). Examination of another proband led to the identification of two other point mutations causing IHH, which could be effectively treated with GnRH agonist therapy; one resulted in a premature stop codon in the coding sequence of the kisspeptin receptor (R331X) and one which resulted in a stop codon being changed to an arginine (X399R) [23]. In the same publication, the authors also reported the same phenotype in transgenic kisspeptin receptor knockout mice (low-circulating gonadotropin levels despite normal hypothalamic GnRH content and pre-pubertal sexual characteristics). More recently, a patient with central precocious (early) puberty who had an activating mutation of the kisspeptin receptor has been reported [24]. Other mutations causing IHH have since been found, including null mutations in both the kisspeptin receptor and the KISS1 gene [25,26] (for a full review, see Silveira and Latronico [27]). IHH has an estimated incidence of 1-10 in every 100,000 births, occurring more often in males than females [28,29].
Non-human mammalian models have provided further evidence that kisspeptin has a central role in triggering the hypothalamic GnRH pulse generator to ‘kick start' puberty and then maintain reproductive function. mRNA expression of KISS1 and kisspeptin receptor is significantly upregulated at puberty, the percentage of GnRH neurons depolarised by kisspeptin increases through late childhood and into adulthood, and pulse frequency of kisspeptin also increases [13,30,31,32]. Furthermore, the timing of the onset of puberty in these animals could be pharmacologically manipulated by the administration of exogenous kisspeptin or kisspeptin antagonist to initiate or delay onset, respectively [33,34,35,36]. Furthermore, studies have shown that the link between nutritional status and timing of reproductive maturity as well as any subsequent period of transient hypogonadotropic hypogonadism related to malnutrition may also be explained through kisspeptin signalling: mRNA expression of KISS1-related genes alters during fasting [37], the kisspeptin receptor is present in peripheral tissues including adipose fat and the pancreas [3,5], and it has been shown by some groups that a percentage of kisspeptin neurons in the arcuate nucleus co-express the leptin receptor [38]. Others did not find this but instead suggested that afferent leptin signalling may be impacting on the kisspeptin/GnRH system [39,40].
Can Kisspeptin Stimulate Reproductive Hormone Release in Patients with Subfertility?
Kisspeptin-54 and -10 have now been administered to many men and women in varied dosing protocols, and by both the intravenous and subcutaneous route, in an attempt to elucidate the effects of peripheral administration of kisspeptin on the reproductive axis, and identify novel therapeutic uses in patients with subfertility. All studies have shown both isoforms to be well tolerated. As it is almost impossible to measure endogenous GnRH pulsatility, LH pulsatility is the gold standard surrogate marker for GnRH pulsatility.
To summarise the literature in healthy male volunteers: both isoforms acutely stimulate gonadotropin secretion approximately 2.5-fold in a dose-dependent manner, with a first LH peak occurring within 60 min of drug administration. FSH and testosterone also increase but to a lesser extent. A single intravenous bolus dose of kisspeptin can continue to stimulate the GnRH neurons for some time (approx. 17 min), and the timing of endogenous pulses created by the central GnRH pulse generator, or ‘clock', is reset. Furthermore, a prolonged infusion of kisspeptin can increase both the frequency and amplitude of LH pulses over the entire time course [9,11,41].
Studies in healthy women revealed that even though kisspeptin does stimulate LH release with a peak approximately 40 min after administration, the size of the effect alters throughout the menstrual cycle: with the greatest effect on LH release being in the pre-ovulatory phase and the least effect in the follicular phase [7,9,41,42,43], thus suggesting that kisspeptin is required to stimulate sufficient LH release to induce egg maturation. This is in keeping with data from animals, which showed that kisspeptin activates the pre-oestrus LH surge [44,45] and data from primates, which showed that a spontaneous pre-ovulatory GnRH surge causes pituitary release of LH [46]. This is contrary to an earlier hypothesis that suggested that ovulation occurs due to an increased sensitivity of the pituitary to GnRH [47].
Since kisspeptin potently stimulated LH release in healthy volunteers, it was of great interest to determine whether it could do the same in patients with subfertility. Functional hypothalamic amenorrhoea (HA) is a reversible condition whereby transient failure of the secretion of gonadotropins occurs in response to a physiological insult, such as a dramatic change in body weight and in particular profound weight loss, psychological stress, chronic ill health, or excessive exercise. Jayasena et al. [48] reported the effects of kisspeptin administration in women with HA and found that such women acutely responded to subcutaneous kisspeptin to an even greater extent than healthy women (20-fold rise in LH), which is thought to be secondary to increased kisspeptin receptor expression as shown by a rodent model of hypogonadotropic hypogonadism caused by undernutrition [37], and/or reduced oestrogen feedback [49]. More recent studies have further shown that repeated administration of kisspeptin to such women with HA can stimulate LH pulsatility [50] but chronic, high-dose administration causes tachyphylaxis [8]. Therefore, in some women with HA, kisspeptin may become a novel physiological therapeutic agent if administered at a dosing regimen that does not cause tachyphylaxis, and if fertility is not immediately a priority, as otherwise FSH replacement is likely to be required to stimulate sufficient follicular growth.
Kisspeptin research has also been conducted in men with hypogonadotropic hypogonadism secondary to diabetes mellitus. Intravenous bolus and continuous infusion of kisspeptin-10 was shown to increase LH 2- to 5-fold, depending on the dose administered [0.3 mg/kg (0.23 nmol/kg) - 4 mg/kg/h (3.1 nmol/kg/h)], which is comparable to the response in healthy men [9,51]. LH pulse frequency also increased without tachyphylaxis over an 11-hour infusion period [51], and serum testosterone levels increased as well as gonadotropin levels [51]. Furthermore, recent animal research has shown that kisspeptin influences glucose and leptin sensitivity, and body weight, in a sexually dimorphic but by a partially sex steroid-independent mechanism. Studies of kisspeptin receptor knockout adult mice showed that female knockout mice had an increased body weight due to higher adiposity despite a reduced food intake, higher leptin levels, and markedly impaired glucose tolerance compared to wild-type littermates [52]. However, this phenotype was not seen in the male knockout mice but was still seen in the chronically ovariectomised female mice [52]. It is therefore possible that kisspeptin may have a role in the causation and treatment of human obesity and glucose intolerance. However, considerably more research will be required in this area as well as in individuals who do and do not have a pathological mutation in the kisspeptin signalling pathway.
Can Kisspeptin Be Useful in in vitro Fertilisation Treatment to Trigger Egg Maturation?
The data in healthy women that showed that kisspeptin was most effective in the pre-ovulatory phase of the menstrual cycle, and could precipitate advanced onset of the LH surge, suggested that kisspeptin-54 could be a novel therapeutic agent to trigger egg maturation in women undergoing in vitro fertilisation (IVF) therapy [7,42]. The effects of kisspeptin in women undergoing such treatment have recently been reported [53].
A single subcutaneous injection of kisspeptin-54 was used to trigger egg maturation after standard preparation with recombinant FSH injections and a GnRH antagonist to achieve superovulation and prevent premature ovulation. Four doses of kisspeptin were used with the majority of patients receiving the three highest doses (6.4, 9.6 and 12.8 nmol/kg) in an adaptive design for dose escalation. Serum kisspeptin, LH, FSH and progesterone levels all increased after the trigger injection. Egg maturation was observed with each dose of kisspeptin; however, the two higher doses improved the ease of egg collection. Biochemical pregnancy occurred in 40% and clinical pregnancy was achieved in 23% (12 of all 53 patients treated). Ten of these women had live births and 12 healthy babies were born.
Infertility is common, affecting 10.9% of women in the US aged 15-44 years [54,] and 1 in 7 couples in the UK [55]. It is associated with significant morbidity in the form of social consequence and psychological distress, and mortality from complications of IVF treatment including ovarian hyperstimulation syndrome [56]. These results are therefore very exciting as they show that a single injection of kisspeptin can stimulate egg maturation and result in successful live births in subfertile women requiring IVF treatment. Since kisspeptin stimulates gonadotropin release by stimulating the release of an individual's endogenous GnRH, it may provide a more physiological trigger for egg maturation during IVF treatment than currently used alternatives such as hCG and GnRH agonists. However, this was a proof-of-concept study, and therefore, before kisspeptin can be widely utilised in clinical practice further data will be required to ensure it is as good as, or superior to, current approaches using hCG or GnRH agonists.
Furthermore, the improved ease of egg collection with the two higher doses may be a reflection of the dose-dependent rise in LH achieved, and/or a demonstration of a direct peripheral effect of kisspeptin on the ovary. A recent study by Gaytan et al. [57] has also provided evidence from a transgenic mouse model that kisspeptin signalling occurs within the ovary, where associated defects in the kisspeptin receptor advances premature ovarian failure, whilst central kisspeptin signalling remains intact as evidenced by the elevated circulating level of gonadotropins. This early data is fascinating but further work will be needed in order to comment further on specific peripheral signalling pathways, or potential future clinical relevance.
Can Kisspeptin Be Useful in Failing Pregnancies?
Based on the finding that kisspeptin increases 7,000-fold in healthy pregnancies [4], is expressed in the placenta [3], and is likely important for trophoblast invasion and angiogenesis [58,59], one might consider whether failing pregnancies have lower levels of circulating or intra-placental kisspeptin and could therefore be supplemented with exogenous replacement. A number of studies in the literature have examined this with slight discrepancies in results. Most found that women with early bleeding in pregnancy [60], recurrent spontaneous abortion [61,62], metabolic conditions including diabetes and hypertension during pregnancy [63], development of pre-eclampsia, and intra-uterine growth restriction [64,65,66] had lower levels of circulating kisspeptin and decreased levels of kisspeptin receptor on placental immunohistochemistry. However, others found no difference in circulating levels of kisspeptin [67] and/or higher levels of placental expression of kisspeptin receptor at the time of pre-term labour [68]. A recent study by Calder et al. [69], however, showed that in a transgenic mouse model, intact uterine kisspeptin signalling was essential for successful implantation of the embryo. These small studies suggest that kisspeptin may be useful as one of a number of biomarkers for identifying at-risk pregnancies [66], but as yet no study has administered kisspeptin isoforms to women to see if complications such as miscarriage could be prevented.
Development of Pharmacological Agents for Clinical Use
Over the last 5 years, kisspeptin analogues and kisspeptin receptor antagonists have been under development to provide new therapeutics for some of the clinical scenarios outlined in this review as well as other uses in oncology. The potential additional benefit of such analogues compared to the currently widely available GnRH agonists is the rapid effect on suppressing the hypothalamic-pituitary-gonadal axis, which therefore limits the initial stimulation on gonadotropin release prior to downregulation that is typical of current therapies including GnRH agonists [70,71,72]. The first phase 1 randomised double-blind placebo-controlled clinical trial of one of the kisspeptin analogues (TAK-683) in healthy human males reported in 2013 [70]. A second phase 1 clinical trial using TAK-448 in healthy human males, and in those with prostate cancer, published this year [71]. Both trials showed the analogues to be effective and well tolerated in humans, in both health and clinically relevant disease, although the 1-month depot formulation used in the cancer patients has since been withdrawn from further development. Further work is needed to test their full clinical utility, and given that kisspeptin was initially noted to have an anti-metastatic effect, longer-term studies will be required to confirm that a protective effect is not lost by chronic desensitisation.
Conclusion
The study of kisspeptin in reproduction has not only provided the detail and understanding of the basic science but has highlighted possible areas of future clinical use throughout the reproductive timeline. With improving molecular techniques that can truly identify the pathogenic lesion in reproductive disorders, and new therapeutic analogues, perhaps such devastating conditions like infertility and sex hormone-dependent malignancies could be more successfully and more safely treated in the future.
Acknowledgements
J.K.P. is an NIHR Academic Clinical Fellow and W.S.D. is funded by an NIHR Career Development Fellowship.
Disclosure Statement
The authors have no conflicts of interests.