KPV (alpha-MSH 11-13): An Evidence Ledger

Three residues subtracted from a thirteen-residue hormone

Alpha-melanocyte-stimulating hormone is thirteen residues long and ends in lysine, proline, valine. Detach those three and the result is KPV, indexed in PubChem as CID 125672 with CAS 67727-97-3, molecular formula C16H30N4O4 and a molecular weight of 342.43. The same three residues occupy positions 11 to 13 of adrenocorticotropic hormone, so PubChem carries ACTH(11-13) as a synonym for the identical entry. Nothing about the fragment is exotic. It is three common amino acids in a row, inexpensive to synthesise, and small enough to be a substrate for the transporters that move dietary di- and tripeptides across the gut epithelium.

One detail in the identity record does real damage downstream. Within alpha-MSH the C-terminal valine is amidated, so the fragment as it exists inside the parent hormone is Lys-Pro-Val-NH2. The PubChem entry is the free acid, and PubChem marks the distinction in its own synonym list, which carries alpha-MSH(11-13) and alpha-MSH (11-13) (free acid) as separate strings on the same record. Published work uses both forms. Songok and colleagues (2018) state their material as H-KPV-NH2 and Ac-KPV-NH2. A large share of papers write only KPV and leave the C-terminus unspecified, so two results placed side by side may describe two compounds differing at that terminus.

Regulatory indexing is thin. A search of the FDA Global Substance Registration System for CAS 67727-97-3 returned no substance record, and no UNII appears anywhere in the PubChem synonym list, which is unusual for a molecule carrying two decades of published work behind it. ChEBI holds it as CHEBI:160254 and the EPA DSSTox identifier is DTXSID80987067. Both are chemical-database identifiers rather than regulatory ones. No approval for human use in any jurisdiction was located for this record.

What the founding experiments measured

Hiltz and Lipton published the founding experiment in FASEB J in 1989. Ear swelling was induced in mice with picryl chloride, graded doses of alpha-MSH(11-13) were compared against saline and against a large dose of corticosteroid, and the tripeptide inhibited swelling in a dose-related fashion. The abstract states neither the route nor the number of animals per group. The framing in that paper is worth keeping: the authors presented the result as evidence that endogenous alpha-MSH fragments may modulate host defence, and offered the tripeptide as a starting point for peptide drugs rather than as one.

A companion paper in Peptides the following year extended the finding to acute paw oedema and to contact sensitivity in mice, again without stated group sizes in the abstract. Stereochemistry came next. Hiltz, Catania and Lipton (1991, PMID 1788140) reported that Ac-[D-Val13]alpha-MSH(11-13)-NH2 raised anti-inflammatory activity roughly four-fold over the parent tripeptide amide in the picryl chloride ear model on intraperitoneal injection, and that D-Pro12 substitution abolished activity altogether. That four-fold statement is the origin of the D-valine claim usually attributed to Watanabe and colleagues (1993, PMID 8397057), who restate it as background rather than reporting it. Their own result is that alpha-MSH(1-13) and its D-Val13 form were equipotent against IL-1 beta-induced ear oedema, and they concluded the L-conformation of the parent molecule is maximally effective. Charnley and colleagues (2008) later found no potency difference between the two tripeptide amides. The literature filed under this heading already contains an analogue, KPdV, that is not the molecule identified above.

What the fragment was pursued for is as informative as what it did. Brzoska and colleagues, reviewing the field in Endocrine Reviews in 2008, describe the pigmentary action of alpha-MSH as the obstacle limiting its use in inflammatory disorders, and KPV as the alternative that retained the anti-inflammatory activity without that action. The tripeptide is, in origin, a subtraction problem rather than a discovery.

PepT1, and a mechanism the reviews do not agree on

Dalmasso and colleagues (2008, Gastroenterology) supplied the mechanism that most later work builds on. In human intestinal epithelial Caco2-BBE cells and Jurkat T cells, nanomolar KPV inhibited NF-kappaB and MAP kinase activation and reduced pro-inflammatory cytokine secretion. Uptake experiments with tritiated KPV gave a Km near 160 micromolar in Caco2-BBE cells and near 700 micromolar in Jurkat cells, which the authors noted were among the lowest values reported for hPepT1 substrates, against a figure of at least 1 millimolar for the standard substrate Gly-Sar.

The transporter dependence was tested rather than assumed. HT29-Cl.19A cells, which do not express PepT1 natively, showed no response to KPV when transfected with empty vector. In the same paper, female C57BL/6 mice received 100 micromolar KPV in drinking water in the dextran sodium sulphate and trinitrobenzene sulphonic acid models, and colitis incidence and pro-inflammatory cytokine messenger RNA were reduced relative to controls. The paper reports its own DSS group size two ways, ten per group in the methods and five per group in the figure legend, and that discrepancy is recorded in the ledger rather than resolved here.

Viennois and colleagues (2016) pushed the same logic further in azoxymethane/DSS colitis-associated carcinogenesis. Mice overexpressing hPepT1 in intestinal epithelium developed larger tumours and greater tumour burden, PepT1-knockout mice developed fewer and smaller ones, and colonic biopsies from patients with colorectal cancer showed increased PepT1 expression. KPV reduced tumorigenesis in wild-type mice and produced none of that effect when given to PepT1-knockout mice. Taken together the two papers make the transporter, not a receptor, the load-bearing element in the intestinal work.

Outside the gut there is one non-rodent animal study, and it points somewhere else again. Bonfiglio and colleagues (2006) denuded the corneal epithelium of both eyes in rabbits by mechanical abrasion and applied KPV topically at 1, 5 or 10 mg/ml, two drops four times daily for four days. The remaining epithelial defect was smaller than under vehicle at each follow-up, and at sixty hours eight of eight treated corneas were completely re-epithelialised against none of the placebo corneas. Pre-treatment with the nitric oxide synthase inhibitor L-NAME inhibited that effect; the authors reserved the stronger phrase, totally prevented, for what L-NAME did to the nitric oxide donor arm. Nitric oxide has no obvious connection to PepT1, and no study joining the two was located.

On receptors the literature contradicts itself, and this record does not resolve it. Luger and colleagues (2003) wrote that KPV was able to bind MC-1R and modulate antigen-presenting cell function. Brzoska and colleagues (2010) wrote that KPV lacks the entire sequence motif required for binding to any known melanocortin receptor and that its signalling mechanism is unknown. Kannengiesser and colleagues (2008) found that KPV rescued all treated animals in a group of mice carrying a non-functional MC1R from death during DSS colitis, which is consistent with the effect being at least partly independent of that receptor. Three statements, two of them from the same laboratory, seven years apart.

Colitis models, and the formulation that is usually being tested

Kannengiesser and colleagues tested KPV in two murine colitis models in 2008. In DSS colitis, treated mice regained body weight sooner, showed reduced inflammatory infiltrate on histology and lower colonic myeloperoxidase activity. In CD45RB-high transfer colitis, treated animals regained weight and showed reduced inflammatory change. Neither group sizes nor the route of administration appear in the abstract, and the full text was not retrievable for this record.

After 2010 the shape of the in-vivo literature changes. Laroui and colleagues loaded KPV into 400-nanometre nanoparticles encapsulated in an alginate-chitosan hydrogel and reported similar therapeutic effect in DSS colitis at a concentration 12,000-fold lower than free KPV in solution. Xiao and colleagues (2017) used hyaluronic-acid-functionalised polymeric nanoparticles of roughly 272 nanometres and a zeta potential near minus 5.3 millivolts, delivered orally inside the same class of hydrogel.

Formulation is not incidental here. Sun and colleagues (2021, PMID 34547895) state plainly that KPV solution is very unstable when rectally administered, which is why their rats received it from a self-cross-linked thiolated polyglutamic acid hydrogel. Li and colleagues (2024, PMID 38289234) built a thermosensitive hydrogel containing hyaluronic acid, epigallocatechin gallate, epidermal growth factor and KPV together and gave it rectally to colitis rats; the upregulation of the tight junction proteins that paper calls Zonula-1 and Claudin-5 belongs to the four-component system, not to the tripeptide alone.

The consequence is a reading problem. Most in-vivo results published under the KPV heading since 2010 are delivery-system results, in which the comparator is frequently the same peptide unformulated and the finding is a statement about the carrier. Secondary summaries routinely detach the number from the carrier and reattach it to the peptide, which converts a formulation-chemistry result into a pharmacology result it was never designed to be.

The antimicrobial claim, and the papers that disagree about it

Cutuli and colleagues reported in 2000 that alpha-MSH and its C-terminal tripeptide inhibited Staphylococcus aureus colony formation and reduced viability and germ tube formation in Candida albicans, with effects described across a broad concentration range including picomolar, and with killing activity partly reversed by an adenylyl cyclase inhibitor. That single paper is the origin of nearly every antimicrobial claim now made for KPV, and it is the only one located that reports killing in the picomolar range.

It did not survive one attempt at independent testing. Rauch, Holzmeister and Kofler published a letter in the same journal in 2009 reporting that they performed microplate growth-inhibition assays on C. albicans and could not observe any growth-inhibiting effect, and that on repeating the original assay with different strains they detected only a mild effect at 100 micromolar. The published abstract of that letter contains an apparent typographical error, writing alpha-MSH where the tripeptide is evidently meant in the second instance, and that ambiguity is recorded here rather than smoothed over. The journal printed an author reply alongside it.

Songok and colleagues (2018) ran into the same wall from a different direction, and only against the bacterium. Working on glycoalkylated lysine analogues, they assayed Ac-KPV-NH2 and its modified forms against S. aureus by agar diffusion, saw no inhibition zones where the ampicillin control inhibited growth, then obtained the peptide from the same commercial supplier used in the earlier positive study and repeated the assay following protocols similar to those of Cutuli and of Charnley, with the same result. They called the outcome surprising and disappointing. No Candida assay appears in that paper. The panel of S. aureus, Pseudomonas aeruginosa and Candida albicans returning minimum inhibitory concentrations above 100 micromolar for Ac-KPV-NH2 belongs to Lau and colleagues (2015), whom Songok cites.

Two other laboratories reported the opposite. Charnley and colleagues (2008) tested Ac-Lys-Pro-Val-NH2 and Ac-Lys-Pro-D-Val-NH2 against Staphylococcus aureus and Escherichia coli, found both antibacterial over a broad range of concentrations against a control tripeptide, and noted that the potencies of alpha-MSH and the two tripeptide amides did not differ and that the cationic lysine charge was not required. Singh and Mukhopadhyay (2011) reported that alpha-MSH(11-13) killed more than 90 per cent of methicillin-sensitive and methicillin-resistant S. aureus in the micromolar range and about 50 per cent in the nanomolar range, with depolarisation and permeabilisation of the cells. Neither group is the original laboratory. Three positive reports stand against two negative ones, the negatives falling on Candida in one case and on S. aureus in the other, and nothing located reproduces picomolar killing specifically. The literature does not resolve, and this record does not resolve it either.

A separate molecule complicates the picture further. (CKPV)2 is a dimer built from two KPV sequences joined by a cysteine-cysteine linker, and it is a different compound with different reported behaviour. Gatti and colleagues (2006) found it more potent than KPV at suppressing LPS-induced TNF-alpha in human peripheral blood mononuclear cells, and later work described anti-fungal and macrophage-polarising effects in a Candida vaginitis model (Ji 2013, PMID 23457491). Results obtained with the dimer are frequently repeated as results for the tripeptide.

The human record, and what 2026 added

No registered human study of KPV was located. Queries to the ClinicalTrials.gov API for KPV, KPV peptide, alpha-MSH tripeptide, melanocortin tripeptide and CKPV each returned zero studies on 18 August 2026. A query for Lys-Pro-Val returned 58 records, none of which involves the tripeptide as an intervention; the leading hits concern tryptophan depletion, a meal replacement and plasma amino acid measurement, and the term is matching amino-acid text rather than the compound. Only ClinicalTrials.gov was searched for this record. The European, Chinese, Japanese and UK registries were not, so this is a single-registry negative rather than a global one.

One trail runs cold in a specific way. Gatti and colleagues wrote in 2006 that the dimer (CKPV)2 was then under clinical investigation for antimicrobial use. No corresponding registry record was found: ClinicalTrials.gov returns nothing for CKPV, and no completed or terminated trial of the dimer was located for this record. Whether that investigation was registered elsewhere, or abandoned, is not established here.

Nothing resembling human pharmacokinetics exists for the tripeptide. A PubMed search combining KPV or Lys-Pro-Val with pharmacokinetics, half-life and bioavailability returned ten records, none of which measures plasma concentrations of the peptide in any species, and no dose-ranging or exposure-response work in humans was found. The closest is a transdermal iontophoresis study of permeation across human skin in vitro (Pawar 2017, PMID 28343991). No published stability time-course in aqueous solution was located either, which leaves the handling and shelf-life figures in circulation with no measurement behind them.

Recent work has moved away from inflammation. An and colleagues reported in 2026 that KPV suppressed adipocyte differentiation in 3T3-L1 preadipocytes, with roughly 55 per cent lower Oil Red O staining and 38 per cent lower triglyceride content at 100 micrograms per millilitre, and that oral administration reduced weight gain in a high-fat-diet mouse model; a companion paper in HepG2 cells reported reduced lipid accumulation under oleic acid loading (Lee 2026, PMID 42064835). Zeng and colleagues (2026, PMID 40935835) used the peptide differently again, as a targeting ligand on deformable liposomes carrying Nlrp3 shRNA to melanocytes in a vitiligo model. In that study KPV is the address, not the cargo.

What is not known

No study of KPV in humans is registered on ClinicalTrials.gov, which returned zero records under five search terms on 18 August 2026; no other national or regional registry was searched for this record, so the human-trial gap is documented for one registry rather than globally. Within that limit there is no human pharmacokinetic data, no dose-ranging, no exposure-response relationship, no adverse-event profile and no route characterisation in humans. The gap extends further: no pharmacokinetic measurement was located in any species, by any route, which means the animal work rests on administered concentrations rather than on achieved exposures, and the oral results in particular cannot be connected to a circulating level. Route attribution is unreliable across the older literature, since the two founding papers state neither route nor group sizes in their abstracts and the full texts were not retrievable for this record. Group sizes are not always stable even where full text exists: Dalmasso 2008 reports its DSS group size as ten per group in the methods and five per group in a figure legend. Most in-vivo work published since 2010 tests a carrier as much as a peptide, and several of the most-quoted numbers are properties of nanoparticles or hydrogels rather than of the molecule. The identity record is itself ambiguous in a way that propagates: the free acid and the C-terminal amide are both in use and papers often do not say which. The antimicrobial finding that made the compound interesting is unresolved in print, with three positive reports and two negative ones and no reproduction of the picomolar figure. No study of KPV and mast cells was retrieved by a PubMed search using four spellings of the compound. No stability time-course exists. Sex distribution in the animal work is uneven and frequently unstated, no long-term exposure data exist in any species, and no toxicology package was located.

Questions