BPC-157 vs TB-500: differences, evidence and regulatory status

Last reviewed 26 August 2026

Two compounds are named together so consistently that the pairing reads as a chemical relationship. It is not one. BPC-157 and TB-500 share no parent protein, no sequence, no length, no N-terminal chemistry and no registry number. One is a fifteen-residue sequence attributed to a protein described in gastric juice; the other is seven residues cut out of a fully sequenced human protein and capped with a synthetic acetyl group.

The two molecules side by side

BPC-157: fifteen residues, four of them proline

GEPPPGKPADDAGLV is four prolines, three glycines, two aspartates, two alanines, and one each of glutamate, lysine, leucine and valine. Three prolines are consecutive, at positions 3 to 5, the fourth at position 8 — and proline is the only proteinogenic residue whose backbone nitrogen sits in a ring, so carries no amide hydrogen, while glycine is the only one with no side chain. There is no cysteine, methionine, aromatic residue or side-chain hydroxyl anywhere in it. Both termini are unprotected, and at neutral pH two positive charges (N-terminal amine, lysine) sit against four negative: the glutamate, both aspartates, the C-terminal carboxylate. Net charge −2.

Provenance is where the two diverge most sharply. Chang and colleagues open their 2011 paper by describing the pentadecapeptide as “a partial sequence of body protection compound (BPC) that is discovered in and isolated from human gastric juice”. What is registered, though, is the pentadecapeptide itself: no curated database record for the parent was located for this article, so the derivation is reported as the literature states it rather than as verified fact.

TB-500: seven residues taken out of a 43-residue human protein

TB-500 is a trade designation, not a chemical name, and what material sold under it contains was settled analytically: Esposito and colleagues examined it by HPLC with high-resolution mass spectrometry, synthesised the candidate peptide independently, and identified the constituent as the acetylated 17–23 fragment of human thymosin β4, Ac-LKKTETQ.

UniProt entry P62328 carries 44 residues including the initiator methionine, which is removed, leaving serine as residue 1 and N-acetylated. Numbering from that mature serine, residues 17 to 23 are Leu-Lys-Lys-Thr-Glu-Thr-Gln. The acetyl on the commercial fragment sits on that leucine’s α-amino group and is a synthetic cap — not the parent’s own acetyl, which is on serine 1 and no part of this fragment. The charge inventory is balanced where BPC-157’s is not: the capped N-terminus contributes nothing, two lysines two positive, the glutamate and C-terminal carboxylate two negative. Net charge zero.

Why that window and not another has a published answer, in the mapping work at reference 4. Note also that the peptide Philp and colleagues used in 2003 was unacetylated LKKTETQ; Rahaman and colleagues describe commercial material as the acetylated form.

TB-500 is not thymosin β4, and the arithmetic says so

Mature thymosin β4 is 43 residues, C₂₁₂H₃₅₀N₅₆O₇₈S, average mass 4963 g/mol. The fragment is 889.0 g/mol — roughly 18 per cent of the parent by mass, short by 36 residues. It carries neither the methionine at position 6 that accounts for the parent’s single sulfur atom, nor the phenylalanine at position 12, which Van Troys and colleagues place in the N-terminal helix — a region their mapping assigns to a structural entity separate from the 17–22 motif. A study performed on the 43-residue protein is a study of the 43-residue protein, and conflating the two is the commonest error here. A further collision of names: thymosin α1 is no structural relative of thymosin β4 at all, being residues 2 to 29 of prothymosin alpha (UniProt P06454).

The identification data that do not reconcile

A second formula and mass for TB-500, C₃₉H₆₇N₉O₁₃ and 877.99 g/mol, circulates widely under the same CAS number. It cannot describe Ac-LKKTETQ. Count the atoms and the peptide gives 38 carbons, including the acetyl group’s two, and ten nitrogens — seven backbone amides, two lysine side chains, the glutamine amide. A formula asserting 39 and 9 describes something else. Queried by CAS 885340-08-9, PubChem returns CID 62707662: C₃₈H₆₈N₁₀O₁₄, 889.0 g/mol, with Ac-Leu-Lys-Lys-Thr-Glu-Thr-Gln-OH among its synonyms.

The consequence is practical. A confirmation run against 877.99 will not match material that is in fact the published heptapeptide, and the failure looks like a supply problem rather than a bookkeeping error. Expected masses should come from the sequence, not from a catalogue — see peptide sequence notation and how to read a certificate of analysis.

What the sequences determine

• Neither peptide can be quantified at 280 nm, because that measurement depends on tryptophan and tyrosine and neither sequence has either; content runs on peptide-bond absorbance near 214 nm, on amino acid analysis, or gravimetrically. The parent protein is the same: one phenylalanine, no tryptophan, no tyrosine.
• Neither has a disulfide to scramble or a thioether to oxidise, containing neither cysteine nor methionine — whereas thymosin β4 carries one methionine, so an assumption moved from parent to fragment transfers a liability the fragment lacks.
• The hydrolytic routes differ. BPC-157 has an Asp-Asp-Ala motif. Li and colleagues (2009) investigated the aqueous stability of four model decapeptides in acetate buffer at pH 4.0 to 5.5 and reported cleavage at the bond C-terminal to the aspartate to be the major degradation pathway, with a strong pH dependence below pH 5. [13] With no asparagine or glutamine, deamidation is closed to BPC-157. TB-500 is the reverse: no aspartate, one C-terminal glutamine, two threonine hydroxyls. See lyophilised and in-solution stability.
• Equal mass is not equal molarity. The ratio 1419.5 ÷ 889.0 = 1.60 makes equal masses roughly 1 : 1.6 molar, favouring the shorter peptide. See concentration after reconstitution and what is in a blend.

What the published literature investigated

Each entry states who investigated what, in which model, and what the authors reported. No conclusion is drawn across studies.

Analytical chemistry

• Esposito and colleagues (2012) investigated the identity of the peptide in TB-500, by HPLC with high-resolution mass spectrometry alongside independent synthesis, and reported it to be Ac-LKKTETQ. [1]
• Rahaman and colleagues (2024) investigated its metabolite profile, in in-vitro enzyme systems and in rats, and reported Ac-LK at highest concentration over 0–6 h and Ac-LKK detectable to 72 h. [2]
• Mazzarino and colleagues (2026) investigated a workflow for 54 prohibited compounds in dried blood spots, serum and plasma, and reported that at 4 and 22 °C BPC-157 and TB-500 degraded completely in serum within a week, while in dried matrices all analytes stayed detectable. [3]

In vitro

• Van Troys and colleagues (1996) investigated the actin-binding site of thymosin β4, in chemically synthesised full-length variants, reporting that “the N-terminal part (residues 1-16) and a hexapeptide motif (residues 17-22) form separate structural entities”, with electrostatic contacts at lysine 18 and lysine 14 important to binding. [4]
• Chang and colleagues (2011) investigated cultured rat Achilles tendon fibroblasts and explants exposed to BPC-157, and reported increased outgrowth, increased survival under hydrogen peroxide stress, increased migration in a transwell assay, and increased phosphorylation of FAK and paxillin. [5]

Rodent models

• Staresinic and colleagues (2003) investigated BPC-157 in a rat Achilles tendon transection model with a cultured-tendocyte arm, reporting biomechanical, functional, microscopical and macroscopical measures to day 14. [6]
• Philp and colleagues (2003) investigated full-length thymosin β4 and, separately, a seven-residue peptide carrying its actin-binding domain, in dermal models in db/db diabetic and 26-month-old mice, and reported that in the aged-mouse arm LKKTETQ gave an outcome comparable to the parent. [7]
• Biçer and colleagues (2026) investigated BPC-157, TB-500 and the two together in a rat Achilles tendon transection model, in 32 male Sprague-Dawley rats in four groups of eight at four weeks, and reported significantly higher maximum load to failure in the TB-500 group than controls, lower total Movin scores in the TB-500 and combined groups, and no added effect from combining them. [8]

Human studies

No published, peer-reviewed randomised controlled trial of BPC-157, and none of Ac-LKKTETQ, was identified for this article. Randomised controlled trials of full-length thymosin β4 have been published: Sosne and Ousler (2015) investigated synthetic full-length thymosin β4 against placebo, in 72 randomised human subjects, and reported no significant difference between groups on either primary endpoint. [9] The section above sets out why a trial of the 43-residue protein is not a trial of the seven-residue fragment. See the evidence grading method.

Review literature

• McGuire and colleagues (2025) reviewed BPC-157 in a musculoskeletal context and reported only three pilot studies in humans. [10]
• Mateescu and colleagues (2026) reviewed its biopharmaceutical position and reported development to remain rudimentary, with no approved formulation and no completed phase II trial. [11]
• Mendias and Awan (2026) reviewed twelve approved and unapproved peptides, treating BPC-157, thymosin β4 and TB-500 separately, and reported rigorous human safety data for the unapproved ones to be scarce. [12]

Why the two fall under different sections of the anti-doping code

The World Anti-Doping Agency classifies by structural relationship where one exists, so the divergence follows from the facts above. Both are prohibited at all times under the 2026 Prohibited List, in force from 1 January 2026. TB-500 is caught by its parent: section S2.3, Growth factors and growth factor modulators, lists “Thymosin-ß4 and its derivatives e.g. TB-500” — named there because it is a derivative of a listed protein, the relationship the Esposito identification establishes. BPC-157 has no such parent and falls to the residual section: S0, Non-approved substances, covers any pharmacological substance “not addressed by any of the subsequent sections of the List” and lacking regulatory approval for human therapeutic use, and the 2026 text names BPC-157 expressly.

The two are mutually exclusive by construction, since S0 applies only where no later section addresses the substance, and they carry opposite designations under Article 4.2.2 of the Code: of S0 the List states that “All prohibited substances in this class are Specified Substances”, and of S2 that “All prohibited substances in this class are non-Specified Substances”. One vial engages both. The UK legal position is separate: see is BPC-157 legal in the UK.

References and provenance

Sequences, formulae, masses and residue numbering come from reference 15. Titles are omitted where the published title states an effect or names an indication; a PMID is given in every case.

1. Esposito S, Deventer K, Goeman J, Van der Eycken J, Van Eenoo P. Synthesis and characterization of the N-terminal acetylated 17-23 fragment of thymosin beta 4 identified in TB-500, a product suspected to possess doping potential. Drug Test Anal. 2012;4(9):733–8. Analytical chemistry. PMID 22962027.
2. Rahaman KA, Muresan AR, Min H, et al. J Chromatogr B. 2024;1235:124033. In vitro and rodent. PMID 38382158.
3. Mazzarino M, Colpaert T, Deventer K, Van Eenoo P. Rapid and harmonized analytical workflow for the determination of peptidic and non-peptidic doping agents in dried and liquid blood matrices. Analyst. 2026;151(15):4398–4413. Analytical chemistry. PMID 42328738.
4. Van Troys M, Dewitte D, Goethals M, Carlier MF, Vandekerckhove J, Ampe C. The actin binding site of thymosin beta 4 mapped by mutational analysis. EMBO J. 1996;15(2):201–10. In vitro. PMID 8617195.
5. Chang CH, Tsai WC, Lin MS, Hsu YH, Pang JH. J Appl Physiol (1985). 2011;110(3):774–80. In vitro. PMID 21030672.
6. Staresinic M, Sebecic B, Patrlj L, et al. J Orthop Res. 2003;21(6):976–83. Rodent, with an in vitro arm. PMID 14554208.
7. Philp D, Badamchian M, Scheremeta B, Nguyen M, Goldstein AL, Kleinman HK. Wound Repair Regen. 2003;11(1):19–24. Rodent. PMID 12581423.
8. Biçer O, Adanir O, Güleryüz Y, et al. Jt Dis Relat Surg. 2026;37(3):822–837. Rodent. PMID 42542926.
9. Sosne G, Ousler GW. Clin Ophthalmol. 2015;9:877–84. Randomised, placebo-controlled phase II trial; full-length thymosin β4, not TB-500. PMID 26056426.
10. McGuire FP, Martinez R, Lenz A, Skinner L, Cushman DM. Curr Rev Musculoskelet Med. 2025;18(12):611–619. Review. PMID 40789979.
11. Mateescu DM, Gavrilescu DM, Constantinescu FE, et al. BPC-157 as an Investigational Peptide Therapeutic: Biopharmaceutical Challenges, Formulation Strategies, and Translational Development Barriers. Pharmaceutics. 2026;18(5):625. Review. PMID 42198317.
12. Mendias CL, Awan TM. Sports Med. 2026; online ahead of print. Review. PMID 41966639.
13. Li N, Fort F, Kessler K, Wang W. Factors affecting cleavage at aspartic residues in model decapeptides. J Pharm Biomed Anal. 2009;50(1):73–8. Peptide degradation chemistry. PMID 19395214.
14. World Anti-Doping Agency. Prohibited List 2026, valid 1 January 2026. Regulatory instrument; S0 p. 4, S2 pp. 7–8.
15. PubChem CID 9941957, CID 62707662 (by CAS 885340-08-9), CID 45382195; UniProtKB P62328, P06454. Curated databases.

Related reference pages

Library entries for BPC-157, TB-500 and BPC-157 + TB-500; the healing and repair class index; the peptide glossary; the full library and research sets.

Last reviewed 9 August 2026 — first publication.