TB-500 (Thymosin Beta-4 17–23)

A seven-residue fragment, not the protein

TB-500 is seven amino acids long. Its sequence is Ac-Leu-Lys-Lys-Thr-Glu-Thr-Gln-OH, written Ac-LKKTETQ, corresponding to residues 17 through 23 of thymosin beta-4 with the N-terminal leucine acetylated. PubChem records it as CID 62707662 with molecular formula C38H68N10O14 and a molecular weight near 889 g/mol, CAS registry number 885340-08-9, UNII QHK6Z47GTG. Thymosin beta-4 is a 43-residue protein of roughly 4963 g/mol. The two are related the way a phrase is related to the paragraph it was cut from, and they are not interchangeable.

The identity was established analytically rather than by disclosure. Esposito and colleagues, working at the Ghent doping control laboratory, obtained a commercial TB-500 formulation and characterised its contents by high-performance liquid chromatography coupled to high-resolution Orbitrap mass spectrometry, reporting in 2012 that the peptide present was the N-terminally acetylated 17-23 fragment of human thymosin beta-4. They then synthesised Ac-LKKTETQ independently by solid-phase peptide synthesis and proposed a detection strategy for plasma and urine. The name TB-500 describes a product; the molecule inside it had to be worked out from the outside.

That gap has regulatory consequences. In its May 2026 evaluation of TB-500 for the section 503A bulk drug substances list, the FDA concluded that neither TB-500 free base nor TB-500 acetate is physically and chemically well characterised, citing naming conventions that follow no established standard and the absence of publicly available data on peptide impurities, aggregates, microbiological quality and bacterial endotoxin. The agency noted it had encountered multiple salts and derivatives, including different active moieties, sold commercially under the same common name.

The numbers that travel with the name

Product listings for TB-500 across the research-chemical and aggregator category carry a consistent set of figures: molecular weight 4963 g/mol, formula C212H350N56O78S, CAS 77591-33-4, forty-three amino acids. Every one of those belongs to a different substance. CAS 77591-33-4 is timbetasin, the international nonproprietary name for synthetic full-length thymosin beta-4, PubChem CID 16132341. The native protein carries CAS 77642-24-1 and UniProt accession P62328. A listing quoting 4963 g/mol for a product named TB-500 is describing the parent protein while selling the fragment, or the reverse, and the document does not say which.

Registry data are not uniformly clean either. The FDA Global Substance Registration System record for UNII QHK6Z47GTG names the substance correctly as N-acetyl-L-leucyl-L-lysyl-L-lysyl-L-threonyl-L-alpha-glutamyl-L-threonyl-L-glutamine and gives CAS 885340-08-9, but its calculated molecular formula field reads C36H66N10O13 at 846.98 Da. That is the unacetylated heptapeptide LKKTETQ, which PubChem lists separately as CID 10169788, fequesetide, CAS 476014-70-7. The registry's own calculation omits the acetyl group its own systematic name specifies. PubChem and the FDA briefing document agree on C38H68N10O14 and 889.01; the GSRS property field does not.

A second error runs through secondary summaries. TB-500 is repeatedly described as a seventeen-amino-acid fragment, which appears to be a misreading of the phrase 17-23 fragment. The withdrawn compounding nomination that prompted the FDA review contained four separate identity inconsistencies of its own: the substance named did not match the certificate of analysis supplied, the certificate's registry data described the acetate salt rather than the free base, the molecular formula given matched neither, and an alternative CAS number offered in the package, 476014-70-7, was the unacetylated peptide.

What has been given to an animal

Ho and colleagues at the Hong Kong Jockey Club racing laboratory published the first post-administration identification of TB-500 in 2012. Thoroughbred geldings received a single subcutaneous dose of a preparation containing 10 mg of N-acetylated LKKTETQ. Plasma concentrations peaked at 0.05 and 0.08 ng/mL between 60 and 120 minutes and fell below quantification between 6 and 10 hours. No intravenous arm was run, so absolute bioavailability by the subcutaneous route was not established. The number of animals is not stated in the abstract, and the FDA summary of the study refers only to geldings in the plural.

Rahaman and colleagues repeated the exercise in rodents in 2024, giving TB-500 by intraperitoneal injection at 50 mg/kg to six-week-old male Sprague-Dawley rats and quantifying urinary metabolites by orbitrap mass spectrometry. N-acetylated LK was the most abundant metabolite over the first six hours; N-acetylated LKK was still detectable at 72 hours. Traces of N-acetylated lysine appeared in rat urine but not in the horse samples. The paper opens by stating plainly that the biological effects of TB-500 have not been documented.

One published in vivo experiment has measured an outcome after giving TB-500 itself. Biçer and colleagues transected and repaired the Achilles tendon in 32 male Sprague-Dawley rats of about 330 g, randomised to four groups of eight, and gave BPC-157 at 10 micrograms per kilogram per day, TB-500 at 60 micrograms per kilogram per day, both, or nothing, intraperitoneally for four weeks. Maximum load to failure was higher in the TB-500 group than in controls at four weeks and reached statistical significance; total Bonar score was lower at p equals 0.016 and total Movin score at p equals 0.017. The combination arm conferred no additional benefit.

Acetylation, and the metabolite that outperformed the parent

The same 2024 paper contains the least convenient finding in this record. Confluent fibroblast cultures were scratched to create uniform wounds and incubated for eight hours with TB-500 at 50 micrograms per millilitre or vehicle. Wound closure did not differ significantly between them. Under identical conditions the C-terminally truncated metabolite N-acetylated LKKTE at the same concentration produced a small but significant closure. The authors' own reading is that the wound-healing activity previously reported in the literature may belong to a metabolite rather than to the administered peptide.

Acetylation is not a cosmetic modification. It alters charge, hydrophobicity and size irreversibly, and with them a peptide's lifespan, folding and binding behaviour, which is why the FDA reviewers stated that the pharmacological profile of the non-acetylated heptapeptide LKKTETQ cannot be directly extrapolated to the acetylated form. Most of the functional literature invoked in support of TB-500 was generated with the unacetylated peptide. It is the same category of error that arises when data generated with an analogue are reported under the name of the parent molecule.

Metabolism itself is well described. Sequential loss of residues from the C-terminus generates Ac-LKKTET, Ac-LKKTE, Ac-LKKT, Ac-LKK and Ac-LK, first shown in homogenised equine liver and then confirmed in horse plasma and urine. Zvereva and colleagues incubated TB-500 with human serum, human kidney microsomes, human liver microsomes and pooled liver S9 fraction for two hours and recovered the same truncation series, with proportions varying by system: Ac-LKKTE and Ac-LKK predominated in serum, Ac-LK in liver microsomes. Deamidation was not observed.

Where the claims were borrowed from

Philp and colleagues reported in 2003 that thymosin beta-4 accelerated dermal repair in db/db diabetic mice and in aged mice, and that a synthetic seven-amino-acid peptide reproducing the actin-binding domain matched the parent molecule in the aged animals. That peptide was LKKTETQ without the acetyl group, applied topically at 0.01 percent in phosphate-buffered saline to 3 mm punch wounds in 26-month-old female BALB/cBy mice on the day of wounding and 48 hours later, with histology at day 7. The FDA reviewers noted the study includes no dose-response assessment.

Shah and colleagues tested the same 17-23 sequence against the amino-terminal 1-15 sequence in early-passage human hepatic stellate cells stimulated with PDGF-BB, and reported that 17-23 but not 1-15 blocked upregulation of the PDGF-beta receptor, alpha-smooth muscle actin and collagen 1, and blunted Akt phosphorylation at both Thr308 and Ser473. Sosne and colleagues had earlier mapped thymosin beta-4's activities onto separate domains in a 2010 review, assigning anti-inflammatory and antifibrotic effects to the N-terminal Ac-SDKP tetrapeptide and cell migration to the central actin-binding region.

Cardiac claims map somewhere else entirely. Hinkel and colleagues synthesised seventeen domain combinations of thymosin beta-4 and tested them in embryonic cardiac cells, in infarcted mice and in pigs, concluding that the C-terminal tetrapeptide AGES, not the actin-binding region, is the domain responsible for the molecule's benefit in the ischaemic heart. AGES sits at residues 40 to 43. TB-500 stops at residue 23 and does not contain it. Any cardiac repair claim attached to TB-500 is inherited from experiments on a sequence the compound lacks.

The human record

There is none. The FDA search of PubMed, Embase, ClinicalTrials.gov, DailyMed and Drugs@FDA for the July 2026 advisory committee found no article in which TB-500 was administered to a human by any route, and no clinical pharmacokinetic or pharmacodynamic data. A search of the FDA Adverse Event Reporting System through 26 March 2025 returned no reports naming TB-500. The Human Foods complaint system returned two reports concerning a blended TB-500 and BPC-157 product, neither carrying a safety assessment.

ClinicalTrials.gov holds exactly one interventional record naming TB-500, NCT07487363, a phase 1/2 dose-escalation study in 80 adults with stable atherosclerotic cardiovascular disease, sponsored by Hudson Biotech, listed as recruiting at Peking University Shenzhen Hospital with a start date of 5 February 2026. Its own brief summary opens: this fictional study is an example of a ClinicalTrials.gov-style record. The oversight module records the study as not FDA-regulated. A registry entry that describes itself as an example is not evidence of a trial, and it is the only registry entry TB-500 has.

The parent protein is in a different position. Ruff and colleagues gave synthetic thymosin beta-4 intravenously to four cohorts of ten healthy volunteers at 42, 140, 420 and 1260 mg, single dose and then daily for 14 days, and recorded no dose-limiting toxicity. Guarnera and colleagues randomised 73 patients with venous stasis ulcers to topical thymosin beta-4 or placebo across eight European sites. Phase 3 ophthalmic trials of the full-length peptide, run as RGN-259, have enrolled 601 and 700 participants. None of that programme used the fragment.

Regulatory and anti-doping position

Reviewers at the FDA proposed on 15 May 2026 that neither TB-500 free base nor TB-500 acetate be added to the section 503A bulks list, on grounds of inadequate physicochemical characterisation, no demonstrable history of compounding use, no evidence of effectiveness, and no human safety data at all. The evaluation also recorded no acute toxicity, repeat-dose toxicity, genotoxicity, reproductive toxicity or carcinogenicity study of either substance, and flagged injectable administration of a poorly characterised peptide as an immunogenicity risk. The Pharmacy Compounding Advisory Committee heard the evaluation on 23 July 2026. Its recommendation is advisory and does not change what may lawfully be compounded.

Anti-doping status is unambiguous and easy to verify. The 2026 WADA Prohibited List, section S2.3, growth factors and growth factor modulators, names thymosin-beta4 and its derivatives, giving TB-500 as the worked example. The class is prohibited at all times, in and out of competition, and all substances in it are non-Specified. WADA funded a project in 2013 specifically to determine detection limits for TB-500 metabolites and build them into peptide screening methods.

What is sold under the name is a separate question again. Delcourt and colleagues at the French racing laboratory analysed products marketed online as TB500 and TB1000 and reported that their contents are not systematically consistent with the descriptions attached to them. TB-500 appears in no United States, European or Japanese pharmacopoeia, is not a component of any approved drug in the United States, Canada, Australia, the United Kingdom or the European Union, and has no marketing authorisation anywhere. A veterinary preparation marketed for equine and greyhound racing appeared around 2011.

What is not known

Almost everything a person would want to know. There is no human pharmacokinetic, pharmacodynamic or safety data for TB-500 by any route, and no case series or registry cohort; the single ClinicalTrials.gov record describes itself as a fictional example. No acute toxicity, repeat-dose toxicity, genotoxicity, developmental or reproductive toxicity, or carcinogenicity study of either the free base or the acetate salt was identified by FDA in May 2026, and none was located here. Immunogenicity has not been assessed, which matters because the compound is proposed for injection and because the certificates of analysis in public circulation test appearance, identity and purity but not aggregates or bacterial endotoxin. The one in vivo efficacy result, in 32 rats over four weeks, is a single exploratory study whose own authors describe it as preliminary and pending dose-optimisation; it has not been replicated. Whether TB-500 has any activity of its own, as distinct from activity of its truncation metabolites, is unresolved — the only direct comparison found the parent inactive and one metabolite active. TB-500 is not an approved medicine anywhere and material sold under the name has been shown not to correspond reliably to its stated contents.

Questions