Molecule guides
Tissue Repair Research Peptides: BPC-157, TB-500, GHK-Cu, KPV and ARA-290 Compared
How the main healing and recovery research peptides differ in mechanism, injury model and quality of evidence, and how to pick one for a repair study.
Four groups of compound, with genuinely different mechanisms, make up the literature on healing and recovery peptides: cytoskeletal and angiogenic repair peptides, matrikines that carry copper, anti-inflammatory fragments derived from the melanocortin system, and cytoprotective molecules acting at defined receptors. Nearly all the evidence comes from cell culture and rodent injury models; controlled human data are scarce and for most of these molecules do not exist at all. This overview connects each group with the injury models where it is actually used, compares the materials researchers most often request, and sets out how to select a reference compound for a given question.
Turning "recovery" into a measurable endpoint
Recovery almost never reduces to a single figure in a repair study. Reported outcomes vary by tissue: stiffness and load to failure in ligament and tendon work; closure rate and percentage re-epithelialisation in excisional skin wounds; the type I to type III collagen ratio measured histologically; capillary density from CD31 immunostaining; and panels of inflammatory cytokines, usually IL-1β, IL-6 and TNF-α. Better rodent studies also include functional readouts such as grip strength or gait analysis.
These outcomes are not coupled. Faster wound closure can come with a mechanically weaker scar, and higher capillary density need not alter tensile strength. Choosing the endpoint first, and the compound second, is what turns a screen into a designed experiment.
Peptide groups used in tissue repair research
BPC-157, the gastric pentadecapeptide
This 15-amino-acid sequence, derived from part of the human gastric juice protein BPC, has a mass of 1419.55 Da and no known endogenous receptor. Its literature describes an indirect mechanism: rodent work reports enhanced eNOS activity and VEGFR2 signalling, increased vessel formation within granulation tissue, and changes in nitric oxide handling. The animal evidence is remarkably broad — nerve, bone, gut mucosa, muscle, ligament and tendon — but nearly all of it originates with a single group in Zagreb across roughly three decades, and that concentration of authorship is a limitation worth naming. No completed human trial has been published.
Thymosin beta-4 versus the TB-500 fragment
These two are not the same reagent. The parent is a 43-residue protein of 4963.44 Da; TB-500 is a synthetic heptapeptide (LKKTETQ) reproducing its actin-binding region and weighing 889.02 Da. The fragment retains actin-sequestering activity and is far cheaper to produce, but other functions of the whole protein are absent from it. Any study intended to reproduce thymosin beta-4 biology in full should use full-length thymosin beta-4; papers that treat the two names as synonyms are in fact discussing two different compounds. Reported activities centre on angiogenesis, migration of cells, and reduced fibrosis in corneal and cardiac injury models.
Copper-binding matrikines
The tripeptide glycyl-L-histidyl-L-lysine complexed with copper(II) — GHK-Cu — was first isolated from human plasma in 1973 by Loren Pickart, and it is the best-documented member of this group. Gene-expression work in cultured fibroblasts reported that GHK affects a substantial fraction of the human genome, with the changes clustering around antioxidant response, metalloproteinase balance and collagen production. Because the copper is integral to activity, handling GHK-Cu differs from handling a normal peptide: pH affects the complex, and the free peptide will not substitute for it.
Anti-inflammatory and cytoprotective peptides
Residues 11-13 of α-MSH make up KPV (342.43 Da), which retains the parent hormone's anti-inflammatory properties while losing the melanocortin-receptor effects on pigmentation. Its proposed mechanism is intracellular inhibition of NF-κB signalling, consistent with the colitis models in which it reaches epithelial cells. ARA-290 (cibinetide) is a different case: an 11-residue peptide engineered to stimulate the innate repair receptor — a heterodimer formed from the EPO receptor and the β-common receptor — without driving erythropoiesis as erythropoietin does. It is one of very few compounds in this overview with published human trial data, from work on sarcoidosis-associated small-fibre neuropathy.
Most frequently specified repair compounds compared
Chain lengths and molecular weights given below are the specifications of the reference molecules and match the figures on the individual product pages; the model column shows where the most substantial published work lies and makes no claim about effect.
What the evidence really shows
Cell and tissue studies
The culture literature is reasonably consistent. Increases in collagen and glycosaminoglycan production in dermal fibroblasts treated with GHK-Cu have been reported by several independent laboratories, and the transcriptional data are the most reproduced findings in the whole area. Both thymosin beta-4 and its actin-binding fragment reliably accelerate endothelial and keratinocyte migration in transwell and scratch assays. In activated immune and epithelial cell lines, KPV suppresses cytokine release driven by NF-κB. BPC-157's cell-culture record, in contrast, is thinner than its animal record — an unusual pattern that no honest review should gloss over.
In vivo models
Rodent experiments dominate. Excisional skin wounds, colitis, muscle crush and Achilles tendon transection recur throughout the BPC-157 literature, with reported improvements in closure rate and tensile strength. Thymosin beta-4 has corneal and cardiac infarct studies from more than one group, and ARA-290 has rodent data in islet transplantation and neuropathy. The methodological problems are familiar: small group sizes, inconsistent blinding and — for BPC-157 — the narrow authorship already noted.
Clinical data
ARA-290 stands apart here, having gone through controlled human trials that reported changes in patient-reported outcomes and small-fibre function. GHK-Cu has controlled dermatology and cosmetic trials of topical formulations. For full-length thymosin beta-4, TB-500 and BPC-157 there are no published, completed human efficacy trials, and not one compound covered in this guide is approved for any repair indication in people.
Selecting a peptide for a healing or recovery study
- Which tissue is involved, and what limits its repair? Where vascularisation is the bottleneck, the angiogenic compounds are indicated; where migration is limiting, the actin-binding peptides; and where inflammation drives the pathology, ARA-290 or KPV.
- Does your readout depend on a known receptor? Defined receptors exist for ARA-290 and the IGF-family reagents, which makes knockout or antagonist controls possible. BPC-157 has none, so mechanistic claims about it are correspondingly harder to establish.
- Do you need local or systemic exposure? Cream and topical formats exist for BPC-157 and the copper peptides, which suits questions restricted to skin; vials are the format when systemic exposure is intended.
- Does the ratio between components matter? Products freeze-dried together, the KLOW blend among them, fix four molecules in a set proportion. That is fine for screening a fixed combination, but it rules out concentration–response work on any one of them.
Every compound associated with this goal appears in the healing and recovery collection, with the tighter mechanistic subset listed under tissue repair peptides. The comparison researchers request more than any other is covered in our detailed BPC-157 versus TB-500 analysis.
Purity, identity and handling
Short peptides — GHK, KPV, TB-500 — are straightforward to synthesise and should arrive at 98% HPLC purity or better with a clean mass-spectrometric identity match; at 15 residues, BPC-157 is much the same. Full-length thymosin beta-4 is a different proposition: 43 residues make the synthesis considerably harder and deletion sequences correspondingly more probable, so the chromatogram warrants closer attention than the purity figure on the front page.
GHK-Cu calls for extra care. What you want is the copper complex, not free GHK, and although a blue tinge in the reconstituted solution is a useful first indication, it proves nothing on its own — the COA should give copper content. The complex is sensitive to pH and must be kept away from chelating buffers. Store lyophilised material from this class sealed, dark and at -20 °C, and aliquot reconstituted solutions immediately so they are never frozen and thawed more than once.
Supply formats and their purpose
No other research goal is offered in as many formats as repair peptides, and format is not cosmetic: it decides which question the material is able to answer.
Lyophilised vials remain the default. You set the concentration when you reconstitute, which is a precondition for any concentration–response design, and it is the only format offering a straightforward path from the vial's mass to a molar concentration.
Serums and creams put the compound directly onto skin and dermal fibroblasts without systemic exposure. That is the more relevant route for cosmetic-adjacent research and dermal wound models, and it is the format from which GHK-Cu's controlled human data came. It is useless for gut or tendon questions, because the material never arrives at those tissues.
Troches and oral capsules are available for BPC-157 and for its arginate salt. The purpose is exposure within the gastrointestinal tract rather than systemic delivery: a peptide of this size is largely digested in the gut, so the oral format is a means of putting the compound where mucosal and colitis models require it — not a substitute for a vial.
Nasal sprays arrive pre-diluted at a fixed concentration. That removes reconstitution error, but it also removes your control over concentration, and it introduces a pH-adjusted, preserved vehicle which must be matched in the control condition.
Related research families
A few compounds appear regularly in repair protocols without belonging to any of the mechanistic classes above. PDRN is a polydeoxyribonucleotide preparation obtained from salmon DNA and acts at the adenosine A2A receptor, not through peptide signalling. LL-37, an antimicrobial cathelicidin studied chiefly in host defence, has reported angiogenic effects. VIP, 28 residues long, is a neuropeptide with broad anti-inflammatory signalling. They sit alongside the repair peptides because they get used for the same purposes, not because the mechanisms are related — and reviews that blur that distinction make the field appear more unified than it really is.
Regulatory position
No peptide discussed here is approved as a tissue repair therapy in either the European Union or the United States. BPC-157 appears on the FDA's list of bulk substances posing significant safety risks in compounding, and WADA prohibits it in competition, as it does TB-500 and other growth-factor modulators. ARA-290 has orphan designations but no marketing authorisation. GHK-Cu is permitted as a cosmetic ingredient for topical use, though that permission says nothing about injectable research material. Every product mentioned is research grade and supplied exclusively for in-vitro and laboratory work.
Matching catalogue section: Peptides for Healing & Recovery
Questions
Are TB-500 and thymosin beta-4 the same compound?
They are not. Thymosin beta-4 is a 43-residue protein of roughly 4,963 Da, while TB-500 is a synthetic heptapeptide reproducing only its actin-binding segment, with a mass near 889 Da; the parent protein has activities the fragment lacks. If your study is meant to model thymosin beta-4 itself, order the full-length protein rather than the fragment.
Among these compounds, which is best supported by evidence?
It depends what kind of evidence. For replicated cell-culture and topical clinical work accumulated across five decades by many separate groups, GHK-Cu leads. For rigorous controlled human trials, ARA-290 does. BPC-157 has by far the broadest animal literature, but it comes from a very small circle of authors and no human trial has been completed and published — a distinction in evidence quality that matters.
Is there a known receptor for BPC-157?
No receptor has been found. What the mechanistic literature describes are downstream events — modulation of nitric oxide, VEGFR2 signalling, altered growth-factor expression — rather than a binding site. The practical consequence is a shortage of mechanistic controls: receptor-knockout and antagonist designs, standard elsewhere in this field, are simply unavailable.
Why are repair peptides so often sold as blends?
Products such as the KLOW and Wolverine blends combine an actin-binding peptide with an angiogenic one, sometimes adding an anti-inflammatory tripeptide and a copper matrikine, all freeze-dried into a single vial. That cuts down handling in fixed-combination screening, but because the ratio is set at manufacture, a blend cannot support concentration–response analysis of any individual component.
What makes GHK-Cu handling different?
Its activity depends on the copper(II) it carries, so the complex and free GHK are not interchangeable and the COA ought to state copper content. Solutions respond to pH and must never meet a chelating buffer such as EDTA, which strips the metal out. Once reconstituted, keep it dark and cold, and divide it into aliquots rather than thawing the same vial repeatedly.
What should a tissue repair study actually measure?
It varies with the tissue. For ligament and tendon: stiffness and load to failure. For skin: rate of closure and extent of re-epithelialisation. Histology gives the type I to type III collagen ratio; CD31 staining gives capillary density; and a panel of inflammatory cytokines completes the picture. Quicker closure reported without any mechanical measurement is an incomplete result, since the speed of closure and the strength of the resulting scar do not necessarily move together.