What Is Kisspeptin? The Gatekeeper of Human Puberty

Peptide science

Kisspeptin is the switch that starts puberty — and it was discovered by cancer biologists who were not looking for it. The gene was cloned in 1996 as a melanoma metastasis suppressor. Five years later three groups independently showed that its peptide products are the natural ligands of an orphan receptor called GPR54. Two years after that, two research teams working on opposite sides of the Atlantic found that losing that receptor means puberty never happens. Thirty years on, the gene named for a cancer experiment has no cancer drug and no kisspeptin drug — but it explains how the reproductive axis is switched on, and it led fairly directly to a class of menopause medicines that reached the market in 2023 and 2025.

What is kisspeptin?

KISS1 sits at chromosome 1q32.1 and has three exons. Note straight away that the ligand gene and the receptor gene are on different chromosomes — KISS1R (also called GPR54 or AXOR12) is at 19p13.3. Conflating the two loci is one of the commonest errors in secondary write-ups.

The precursor is cleaved into four peptides: kisspeptin-54 (also called metastin), kisspeptin-14, kisspeptin-13 and kisspeptin-10. All four share a C-terminal RF-amide, and the ten-residue fragment carries essentially the entire activity — UniProt marks residues 112 to 121 as “essential for receptor binding and receptor activation.” Kisspeptin-10 in its human form is YNWNSFGLRF with a C-terminal amide, PubChem CID 25240297, molecular formula C63H83N17O14, molecular weight about 1,302 g/mol.

You will see the precursor described as 145 amino acids in the founding literature and 138 in current UniProt. Both are right. NCBI records a polymorphism in the terminal exon: when an adenosine is present it forms the third position of a stop codon; when it is absent a downstream stop codon is used and the protein runs seven residues longer. (The figure of 164 residues still circulating online comes from the uncorrected 1996 abstract; the authors published an erratum in 1997.)

One detail matters more than any of the numbers: non-amidated kisspeptins are inactive. The C-terminal amide is load-bearing, not cosmetic. Related RF-amide peptides such as NPFF and prolactin-releasing peptide produce no response at GPR54 up to micromolar concentrations, so the family resemblance does not translate into cross-activity.

The name really does come from Hershey

The 1996 paper came out of the Jake Gittlen Cancer Research Institute at Penn State in Hershey, Pennsylvania. Transferring chromosome 6 into melanoma cell lines suppressed metastasis by at least 95% without affecting tumorigenicity, and the gene responsible was named KiSS-1. OMIM records the rationale as combining the laboratory’s nomenclature for putative suppressor sequences “with acknowledgment of the gene’s discovery in Hershey, Pennsylvania.” The chocolate-kiss link is repeated everywhere and is almost certainly the intended joke, but the authoritative sources say only “Hershey” — so treat the candy detail as folklore with a very plausible basis.

“Metastin” is a later name, coined in 2001 by the group that isolated the 54-residue peptide from human placenta as the ligand of an orphan receptor. “Kisspeptins” was coined the same year by a different group. Neither name predates KiSS-1.

The 2003 moment

In August 2003, de Roux and colleagues published a large consanguineous family with eight children, five affected. The index case was a twenty-year-old man with no pubertal development and a normal sense of smell — which rules out Kallmann syndrome and points at signalling rather than neuron migration. Homozygosity mapping put the locus on 19p13, and all affected siblings carried a homozygous 155-nucleotide deletion spanning a splice acceptor site, truncating the receptor in its third intracellular loop and deleting the last two transmembrane helices.

Two months later, in October 2003, Seminara and colleagues published in the New England Journal of Medicine under the title “The GPR54 Gene as a Regulator of Puberty,” combining human genetics with a purpose-built GPR54-deficient mouse. The ligand side was closed in 2012 with an inactivating KISS1 mutation (N115K) in four affected sisters.

The mirror image arrived in 2008: a heterozygous Arg386Pro change in the receptor’s C-terminal tail, found in a girl with idiopathic central precocious puberty and absent from 300 control chromosomes. The mechanism is subtler than it is usually reported. This is not a constitutively active receptor — basal signalling, binding affinity, dose-response and maximal effect were all unchanged. What changed was desensitisation: signalling declined more slowly, ERK phosphorylation stayed higher, and the receptor persisted longer at the plasma membrane. The authors make the point explicitly that a truly constitutive receptor would be expected to wreck pulsatility and delay puberty.

How kisspeptin is studied to work

KISS1R is a class A GPCR coupling to Gq. Ligand binding drives phospholipase C, IP3 turnover and a calcium rise, inhibits inward-rectifier potassium channels and activates TRPC-like cation channels — the net effect being depolarisation of GnRH neurons. There is a Gq-independent arm through beta-arrestin to ERK1/2 as well. The response is pertussis-toxin insensitive with no change in cAMP, which is how the coupling was assigned. For the general framework see what a GPCR is.

The circuit is where it gets interesting. Kisspeptin neurons in the arcuate nucleus co-express neurokinin B and dynorphin and are known as KNDy neurons; they project directly to GnRH neurons and are widely regarded as the GnRH pulse generator. A second population in the anteroventral periventricular region does the opposite job. Estradiol inhibits the arcuate population, giving negative feedback, and stimulates the periventricular population, giving the positive feedback that produces the preovulatory LH surge. One peptide, two populations, both directions of feedback.

Two constraints on how you describe this. First, kisspeptin does not act on the pituitary — its LH and FSH effect is abolished by pre-administration of a GnRH antagonist, so it works strictly through GnRH neurons. Our overview of the HPG axis puts it in the wider chain.

Second, the leptin story is more complicated than it is usually told. Kisspeptin is genuinely metabolically gated, but conditional-knockout work showed that deleting leptin receptors from kisspeptin cells causes no puberty or fertility deficit, and restoring leptin receptors only in kisspeptin cells of leptin-receptor-null mice produced no pubertal development and no metabolic rescue at all. Leptin’s permissive role is real but largely indirect. See what leptin is for the other half of that story.

Kisspeptin-10 versus kisspeptin-54

At the receptor, the fragments are essentially interchangeable. In the 2001 characterisation, kisspeptin-54, -14, -13 and -10 were described as equipotent at both rat and human receptors, with human binding constants of 1.45, 1.65, 4.23 and 2.33 nM and EC50 values clustered between 4 and 8 nM. That is a receptor-binding statement, not a clinical one.

There is a second constraint that shapes everything about how kisspeptin could ever be used: continuous exposure desensitises the system. In juvenile male rhesus monkeys, continuous infusion of kisspeptin-54 desensitised GnRH release. More kisspeptin does not mean more LH. Pulsatility is the entire point, which is why our note on tachyphylaxis and downregulation is directly relevant here.

The first human dosing study, in 2005, gave kisspeptin to six healthy male volunteers and found significant increases in circulating LH, FSH and testosterone. Primate work the same year showed KISS1 expression rising with puberty and kisspeptin-10 producing a GnRH-mediated LH surge in agonadal juveniles.

The IVF trigger trials

The most striking human data come from a Hammersmith group in London. The rationale is elegant: conventional IVF triggers either use hCG, which has a long half-life and drives ovarian hyperstimulation syndrome, or a GnRH agonist. Kisspeptin triggers the woman’s own GnRH-mediated LH surge and then clears.

The first trial, published in 2014, gave 53 women a single subcutaneous kisspeptin-54 injection with oocyte retrieval 36 hours later. Egg maturation occurred at every dose tested; 92% (49 of 53) had fertilisation and embryo transfer; biochemical pregnancy was 40% and clinical pregnancy 23%.

The second trial, published in 2015, deliberately enrolled 60 women at high risk of OHSS. Oocyte maturation occurred in 95%, embryos formed in 90%, and 51 of 60 had a fresh transfer. At the best-performing dose the biochemical, clinical and live-birth rates were 85%, 77% and 62%.

The safety signal is the headline. Three women (5%) had mild early OHSS and one (2%) mild late OHSS. There were no moderate, severe or critical cases, none required medical intervention, and none were hospitalised — against an expected rate of roughly 27% by published risk criteria for that population. Of 15 women with a previous hCG-triggered cycle, three had previously had severe OHSS requiring admission; none had any OHSS in the kisspeptin cycle, and two of those three achieved healthy pregnancies.

The authors called the result “promising” and said additional large randomised studies were warranted. That is the right register. It is a single open-label study with adaptive dose allocation and no direct comparison against a conventional trigger.

Why the drugs that reached the market are not kisspeptin

No kisspeptin or KISS1R product is FDA-approved. The commercial payoff from this circuit arrived from one door over — at the neurokinin B receptor on the same KNDy neuron.

Fezolinetant, approved in the US in May 2023 for moderate-to-severe vasomotor symptoms of menopause, is a neurokinin 3 receptor antagonist. Its own label describes it as blocking neurokinin B binding “on the kisspeptin/neurokinin B/dynorphin (KNDy) neuron” — the KNDy circuit is written into an FDA label. In September 2024 the FDA issued a drug safety communication about rare serious liver injury, and in December 2024 added a boxed warning; the current label requires hepatic laboratory testing before starting and at intervals thereafter. Elinzanetant, a combined NK1 and NK3 antagonist, was approved on 24 October 2025 for the same indication; its current label carries no boxed warning but does warn about CNS depressant effects and daytime impairment, and it is contraindicated in pregnancy.

So the honest framing is this: kisspeptin biology produced a real, approved therapeutic class — for hot flushes, through a receptor kisspeptin does not bind. Meanwhile a kisspeptin analogue appears in DrugBank and there is a kisspeptin-10 entry in our library, but nothing has been licensed.

The metastasis-suppressor side, honestly

The original cancer findings are real and were reproduced. KISS1 transfection into a human breast carcinoma line reduced metastatic potential by about 95% without suppressing tumorigenicity. Metastin inhibited chemotaxis and invasion in vitro and attenuated pulmonary metastasis in mice. Kisspeptin-10 strongly reduced proliferation of receptor-expressing cells without inducing apoptosis, and produced Rho-dependent stress fibres.

All of that is cell-line and mouse work. There is no approved kisspeptin-based cancer therapy, and there is a certain irony in a gene that earned its name and its first decade of fame from cancer biology whose actual clinical footprint today is reproductive and menopausal.

Frequently asked questions

Is kisspeptin a libido drug?

No. The study people are usually thinking of is a 2017 mechanistic imaging experiment in 29 healthy young men, which found that kisspeptin enhanced limbic brain activity specifically in response to sexual and couple-bonding stimuli. That is a brain-imaging and psychometric finding, not a treatment trial, and there is no efficacy endpoint, no approved product and no established dosing.

Does kisspeptin act on the pituitary?

No. Its effect on LH and FSH is entirely GnRH-dependent — pre-treatment with a GnRH antagonist abolishes it. Kisspeptin acts on GnRH neurons, which express the receptor; the pituitary responds to the GnRH those neurons release.

Is kisspeptin-10 the same as kisspeptin-54?

At the receptor they are close to equipotent, and kisspeptin-10 carries the essential C-terminal decapeptide. In a body they are not interchangeable — the human trials that produced the results above all used kisspeptin-54, and pharmacokinetics rather than affinity is what separates them. Both are inactive without the C-terminal amide.

Would more kisspeptin mean a stronger reproductive signal?

The opposite, past a point. Continuous exposure desensitises the receptor and lowers LH release; the 2008 precocious-puberty mutation causes early puberty by slowing switch-off rather than by increasing signal strength. The system encodes information in the rhythm, not the amplitude.

Informational only — not medical advice · 21+