Thymosin beta-4

Overview

Thymosin beta-4 (Tβ4) is a small, naturally occurring 43-amino-acid peptide that is among the most abundant peptides inside mammalian cells, where it is found at high intracellular concentrations in many tissues. Its primary, best-characterized job is to act as the major G-actin-sequestering molecule of the cell: it binds monomeric actin and helps regulate the rapid assembly and disassembly of the actin cytoskeleton that underlies cell shape and movement (Goldstein et al., 2012).

Beyond this housekeeping role, Tβ4 has been studied for a range of regenerative and reparative activities, including effects on cell migration, angiogenesis (blood-vessel formation), wound healing, corneal and ocular surface repair, cardiac repair, and anti-inflammation (Goldstein et al., 2012; Xing et al., 2021). Because of these properties it was advanced as a drug candidate by RegeneRx Biopharmaceuticals and partners under the development names RGN-259 (a topical ophthalmic formulation) and RGN-352 (a systemic injectable formulation). These programs reached small, early-stage human trials, but none has produced an approved medicine.

A point of frequent confusion: full-length Tβ4 is not the same as the widely marketed research peptide TB-500, which analytical work identifies as a short acetylated fragment (Ac-LKKTETQ, residues 17-23) of the protein rather than the whole protein. This entry concerns the full-length 43-amino-acid molecule.

Chemistry and structure

Human thymosin beta-4 is a 43-amino-acid, N-terminally acetylated peptide with a molecular weight of approximately 4963 g/mol. It contains no cysteine residues and therefore forms no disulfide bonds. Its sequence is:

Ac-SDKPDMAEIEKFDKSKLKKTETQEKNPLPSKETIEQEKQAGES

In solution the peptide is largely unstructured and flexible, lacking a hydrophobic core; when it binds actin it adopts an extended conformation with an N-terminal helix and a C-terminal capping helix that together wrap across the actin monomer (Irobi et al., 2004). Within the sequence, residues 17-23 (LKKTETQ) form the principal actin-binding motif.

It is precisely this short motif that is the basis of the marketed peptide TB-500. Analytical chemistry of material sold as "TB-500" identifies it as the acetylated 17-23 fragment alone (Ac-LKKTETQ), a 7-residue peptide, not the full 43-residue protein. The table contrasts the two:

Mechanism of action

Tβ4's proposed activities come from a large body of in-vitro and animal work, plus structural studies of how it interacts with actin.

• G-actin sequestration. Tβ4 binds monomeric G-actin at roughly a 1:1 ratio, maintaining a reservoir of actin and regulating the balance between monomeric and filamentous (F-actin) forms. Structural work shows the peptide sterically blocks the bound actin monomer from joining either the barbed or pointed end of a filament, while still permitting exchange with profilin during cell movement (Irobi et al., 2004). This is its best-established function.
• Cell migration and angiogenesis. Through its effects on the cytoskeleton and on endothelial cells, Tβ4 has been linked to enhanced cell migration and to angiogenesis, including associations with vascular endothelial growth factor (VEGF) signaling (Goldstein et al., 2012; Xing et al., 2021).
• Integrin-linked kinase (ILK) / Akt signaling. In a cardiac-injury model, Tβ4 was reported to form a complex with PINCH and ILK and to activate the survival kinase Akt, promoting cardiac-cell migration and survival (Bock-Marquette et al., 2004).
• Wound healing, anti-inflammation, and tissue remodeling. Tβ4 is associated with accelerated re-epithelialization, reduced inflammation, modulation of collagen deposition, and reduced scarring in various wound and corneal models (Goldstein et al., 2012; Xing et al., 2021).

Research and evidence

It is useful to read the evidence in two layers: a deep preclinical layer (cell culture and animal models), and a thinner, early-stage human layer consisting of small trials, some of which have not reproduced their early signals.

Animal and in-vitro evidence

Most of the encouraging biology is preclinical. Tβ4 accelerates re-epithelialization and wound closure in rodent dermal and corneal models, reduces inflammation, and promotes angiogenesis in endothelial assays (Goldstein et al., 2012; Xing et al., 2021). In a mouse cardiac-injury study it improved cardiac-cell survival and function via ILK/Akt signaling (Bock-Marquette et al., 2004). These are animal- and cell-based findings and do not, on their own, establish clinical benefit.

Human evidence — RGN-259 (ophthalmic)

The most developed human program is RGN-259, a topical 0.1% thymosin beta-4 ophthalmic solution studied for dry eye disease and neurotrophic keratopathy (a corneal disease in which the surface fails to heal). In a small randomized, placebo-controlled, double-masked Phase 3 neurotrophic keratopathy trial (SEER-1; 18 subjects total), 60% (6/10) of RGN-259-treated patients achieved complete corneal healing at 4 weeks versus 12.5% (1/8) on placebo. The primary endpoint showed a strong trend but narrowly missed conventional statistical significance (p = 0.0656), while a post-treatment durability measure reached significance; the treatment was well tolerated (Sosne et al., 2023). A subsequent European Phase 3 neurotrophic keratitis trial (reported as SEER-3) did not meet its primary endpoint, an outcome attributed in part to a stronger-than-expected placebo response. The dry eye program has likewise produced mixed results across trials. Net: promising but inconsistent, with no approval.

Human evidence — RGN-352 (systemic)

RGN-352 is an intravenous Tβ4 formulation developed for systemic indications such as acute myocardial infarction and other organ injury. A published randomized, placebo-controlled Phase 1 study gave single and multiple intravenous doses of synthetic Tβ4 (42-1260 mg) to healthy volunteers and found it well tolerated with no dose-limiting toxicity and no serious adverse events; the half-life was short and increased with dose (Ruff et al., 2010). A planned Phase 2 trial in acute myocardial infarction (NCT01311518) was placed on an FDA clinical hold in 2011 because of a contract manufacturer's non-compliance with current Good Manufacturing Practice (cGMP) rules (a manufacturing issue, not a safety signal), and the trial was ultimately never initiated (it is now recorded as withdrawn). As a result the systemic efficacy of Tβ4 in humans remains undemonstrated.

A separate first-in-human Phase 1 study of recombinant human Tβ4 in healthy Chinese volunteers similarly reported no serious adverse events or dose-limiting toxicities and a short, dose-proportional half-life of roughly 0.5-2 hours (Wang et al., 2021). An earlier European randomized, placebo-controlled study of topical Tβ4 in venous stasis ulcers reported good tolerability with healing effects described as preliminary (Guarnera et al., 2007).

Safety and risks

In the controlled trials conducted to date, thymosin beta-4 has generally been well tolerated: phase 1 studies of both synthetic and recombinant Tβ4 reported no dose-limiting toxicities and no serious adverse events, with mostly mild, self-resolving events (Ruff et al., 2010; Wang et al., 2021). That is reassuring as far as it goes, but the human safety database is small and short-term, and several important risks are unresolved.

• Limited long-term human data. Trials have been small and of limited duration. Chronic effects, immunogenicity over time, and rare adverse events are not well characterized. "Well tolerated in small trials" is not the same as "proven safe for long-term use."
• Pro-angiogenic activity and cancer caution. Tβ4's pro-angiogenic and pro-migratory biology is a double-edged sword. Overexpression of Tβ4 has been reported to increase tumor metastasis and angiogenesis in melanoma models, and Tβ4 has been linked to epithelial-mesenchymal transition in some cancers (Cha et al., 2003). A pro-angiogenic, pro-migratory agent could, in principle, support tumor growth or spread. This is a recognized theoretical reason for caution, demonstrated in animal cancer models but not characterized as a clinical outcome in people.
• Product-quality risk for non-pharmaceutical material. Thymosin beta-4 sold as a "research" peptide is not made to pharmaceutical standards and may be mislabeled, contaminated, or of variable potency, a safety concern independent of the molecule itself.

Regulatory and legal status

• Approval. Thymosin beta-4 is not approved as a medicine by the FDA, EMA, or any other major regulator as of 2026. The RGN-259 and RGN-352 programs remain investigational; some trials have been completed, some missed endpoints, and some were interrupted by manufacturing issues. None has produced a marketed product.
• United States. It is neither an approved drug nor a lawfully marketed dietary supplement; peptides of this type do not meet the dietary-ingredient definition. Material circulating online is typically labeled "for research use only, not for human consumption."
• Sport. Thymosin beta-4 is prohibited in sport at all times (in- and out-of-competition). The World Anti-Doping Agency (WADA) lists thymosin-β4 and its derivatives under Section S2 (Peptide Hormones, Growth Factors, Related Substances and Mimetics), among growth factors affecting muscle, tendon, or ligament. As a non-approved substance it may also fall under Section S0.
• Other jurisdictions. It is generally not approved as a medicine elsewhere; national rules on possession, import, and sale of unapproved peptides vary.

This entry is educational and does not constitute legal or medical advice.

How it compares

The table sets full-length thymosin beta-4 against its marketed fragment TB-500 and against BPC-157, the peptide it is most often discussed alongside in injury-recovery contexts. See also the Recovery & injury hub. The central point: human efficacy data are limited or absent for all three, and only full-length Tβ4 has reached formal human trials.

Versus TB-500: full-length Tβ4 is the complete protein, and essentially all of the favorable biology and all of the human trials describe the full molecule, not the short fragment that is sold as TB-500. The fragment lacks the regions of the protein that govern its broader signaling and pharmacokinetics.

Versus BPC-157: the two are chemically unrelated (different sequences, different proposed mechanisms) and are paired by convention in recovery discussions rather than by shared evidence. BPC-157 has no human efficacy trials, whereas Tβ4 has reached small human trials; neither is approved for any indication.

Common misconceptions

• "Thymosin beta-4 and TB-500 are the same thing." They are different molecules. Full-length Tβ4 is the complete 43-amino-acid protein; TB-500 is the short Ac-LKKTETQ fragment (residues 17-23). They differ in sequence, size, and pharmacokinetics.
• "Thymosin beta-4 is a proven healing drug." It is investigational. Human trials are small and mixed: a Phase 3 neurotrophic keratopathy trial showed a strong but non-significant healing trend, a later European Phase 3 missed its primary endpoint, and the systemic cardiac program was interrupted by manufacturing problems. No product is approved.
• "It's natural, so it's safe." Tβ4 is naturally occurring, but natural origin does not establish safety — and its pro-angiogenic biology underlies a theoretical cancer caution. Human safety data are limited and short-term.
• "It's FDA-approved or a legal supplement." It is neither. It is an unapproved investigational peptide, also sold as an unregulated research chemical, and it is prohibited in sport.
• "The animal results prove it works in people." Strong preclinical data are not the same as demonstrated human efficacy. For Tβ4 the human results so far are preliminary and, in places, negative.

Community claims & recent evidence

These points address claims circulating in the peptide community (including popular video "masterclasses"), checked against the primary sources we could reach. Some presentations pin dramatic human "cardiac reversal" results on TB-500/Tβ4; where we could trace them, most led back to animal work rather than to human trials, and a few cite papers we could not locate in any database we searched. An important caveat throughout: not finding a citation here does not make a claim false. A great deal of relevant research — paywalled journals, regional (including Russian and Chinese) literature, conference abstracts, clinical practitioner experience, and unpublished or proprietary data — is not indexed in the sources available to us. Absence of evidence in reachable databases is not evidence of absence.

Verified additions

• Tβ4 mobilises adult epicardial progenitor cells and drives new-vessel formation in the injured heart. [Animal] In mice, Tβ4 reactivated quiescent adult epicardium, restored progenitor-cell potential, and promoted neovascularisation after cardiac injury (Smart et al., Nature 2007).
• Tβ4 limits infarct size in a large-animal ischaemia model. [Animal] Acting as the key paracrine factor of embryonic endothelial progenitor cells, regional Tβ4 reduced infarct size from ~54% to ~37% of the area at risk in a pig coronary-occlusion model (Hinkel et al., Circulation 2008).
• A first controlled human heart trial now exists — and it was neutral overall. [Human] A randomised, double-blind, placebo-controlled trial of recombinant human Tβ4 in 96 patients with acute ST-elevation myocardial infarction after reperfusion found no significant reduction in infarcted area at 90 days for the full cohort; only a subgroup dosed within 8 hours of PCI (n = 43) showed a significant reduction (Zhang et al., Cardiovascular Research 2025). This is the first controlled human efficacy signal for systemic Tβ4 in the heart that we could locate — and it is preliminary and subgroup-dependent, not a demonstrated benefit.

Claims we could not verify

• "A 2005 human trial (Renoult, Circulation) of 30 post-MI patients showed ejection fraction rose and scar reversed, maintained for 3 years." We could not trace this specific citation to any identifiable published study in Circulation or the databases we searched — which is a limit of what we could reach, not proof it does not exist. What we can document positively is that the one systemic-Tβ4 cardiac program in humans that we could find, RGN-352 for acute MI (NCT01311518), was placed on an FDA clinical hold over a manufacturing (cGMP) issue and never ran, and that the first human cardiac RCT we could locate (Zhang et al., 2025, above) appeared two decades later and was neutral overall. On the accessible record, a "proven 2005 reversal" is not something we can confirm. [Unverified citation]
• "A 2003 case series (Osman, European Heart Journal) converted 5 of 7 chronic-AF patients to sinus rhythm, maintained for life." We could not locate this specific paper, and we found no human trial of Tβ4 or TB-500 for atrial fibrillation in the literature accessible to us — again, a limit of reach rather than a disproof. The AF "fibrosis-reversal" material we were able to find rests on animal models. [Unverified citation]
• "Cardiac regeneration with TB-500 is documented and irrefutable — reversing scar, AF and dilated cardiomyopathy in people." Fibrosis reduction, angiogenesis, and functional recovery with Tβ4 are shown in mice and pigs; the same outcomes are not established in humans in the evidence we could reach, and the single controlled human heart trial we located did not meet its overall endpoint. Preclinical promise is not, on its own, human proof — though it also is not a refutation. [Overstated relative to the human evidence we found — animal data]
• The compound is likely mis-named throughout. Essentially all of the cardiac biology we could trace used full-length, 43-amino-acid thymosin β4, whereas the video attributes it to "TB-500," the marketed 7-residue Ac-LKKTETQ fragment. We found no human trial data specific to the fragment; carrying the full-length protein's studies over to it treats the two as interchangeable when the accessible evidence does not support that. [Probable misattribution]