TB-500: Research Overview

The honest summary

TB-500 is a lab-made piece of a natural protein. The full protein is thymosin beta-4, which is 43 amino acids long; TB-500 is a seven-amino-acid snippet cut from its middle, the part that grabs onto actin — the material cells use to build their internal scaffolding and to crawl toward an injury.

That difference in size is the most important fact on this page. Nearly all the encouraging research, including the one nerve-related experiment in this whole index, used the big protein and not the small piece ^[14][16]. TB-500 marketing borrows those results anyway.

The nerve-related experiment is a rat stroke study. Rats were given a stroke, then treated with the full protein starting a day later, and their neurological function improved at two of the three amounts tested ^[14]. That is genuine and it is also four steps away from a person with a damaged nerve: different species, different molecule, brain rather than peripheral nerve, and a hospital-style injury rather than an everyday one.

No controlled human trial of TB-500 itself has been completed for anything.

What it is — and what it is not

TB-500 is a synthetic, N-terminally acetylated heptapeptide, Ac-Leu-Lys-Lys-Thr-Glu-Thr-Gln-OH, corresponding to residues 17 to 23 of thymosin beta-4, the endogenous 43-amino-acid protein encoded by TMSB4X. That LKKTETQ stretch is the conserved actin-binding region shared across the beta-thymosins.

In commerce and in the anti-doping literature, "TB-500" means the heptapeptide, roughly 889 daltons. In the efficacy literature, the molecule under test is usually full-length thymosin beta-4, roughly 4,963 daltons — more than five times the mass, with everything outside the actin-binding motif still attached.

Whether the isolated seven-mer reproduces the parent protein's effects is not established in any controlled human trial. This site treats the two as different compounds, marks which one each study used, and does not let a result from one silently become a claim about the other. It is the single most common error in writing about this peptide.

One further practical point: material sold as TB-500 for research is not manufactured to medicine-grade standards, and identity, purity and even correct sequence — full-length versus fragment — vary between suppliers. Analytical characterisation of these preparations exists in part because anti-doping laboratories needed it.

How the parent protein works

The mechanism is structurally well established, and it belongs to thymosin beta-4 rather than to the fragment.

X-ray crystallography of a gelsolin-domain-1–thymosin beta-4 hybrid bound to actin, solved at 2 ångström, showed that thymosin beta-4 forms a one-to-one complex with monomeric G-actin and sequesters it by capping both ends of the monomer, preventing it from polymerising; the WH2 actin-interacting motif is what does this ^[17]. In cells, that makes the protein the major intracellular buffer holding a reserve pool of unpolymerised actin, which in turn regulates cytoskeletal dynamics, cell migration and motility.

A 2012 review consolidates what follows from that in injury models: thymosin beta-4 promotes cell mobilisation and migration and stem-cell activity, decreases myofibroblast numbers and so reduces scar formation, is released by platelets and macrophages after injury where it limits apoptosis, inflammation and microbial growth, and promotes angiogenesis. The review presents this as the rationale for clinical trials in dermal wounds, corneal injury, and heart and central-nervous-system repair ^[15].

The listed cellular targets are monomeric actin, hair follicle bulge stem cells, vascular endothelial cells, cardiomyocytes and epicardial progenitor cells, and keratinocytes and corneal or dermal epithelium — with the stem-cell and cardiac entries established for the full-length protein.

Whether buffering the actin monomer pool matters to a regenerating nerve is a reasonable question. No study in this index asks it.

What the research shows

Neurological function after stroke (rat, full-length protein). In male Wistar rats given an embolic middle cerebral artery occlusion, intraperitoneal thymosin beta-4 — 2, 12 or 18 mg/kg beginning 24 hours after the stroke, then every three days for four further doses — improved neurological function at 2 and 12 mg/kg, significant from day 14 through day 56 (p<0.05). The highest amount, 18 mg/kg, produced no significant benefit, and the authors modelled an optimal dose of about 3.75 mg/kg ^[14].

That non-monotonic result is worth pausing on, because it cuts against a common community assumption. More was not better; the highest amount tested did nothing measurable.

Safety in humans (full-length protein). A randomised, placebo-controlled Phase 1 study gave synthetic thymosin beta-4 intravenously to 40 healthy volunteers in four cohorts of ten — a single dose, then daily for 14 days, at 42, 140, 420 or 1,260 mg. It was well tolerated, with only infrequent mild-to-moderate adverse events, no dose-limiting toxicities and no serious adverse events; pharmacokinetics were dose-proportional, with half-life increasing at higher doses ^[16]. This is a safety study of the parent protein in healthy people. It is not an efficacy result, and it is not a study of TB-500.

Structural basis. The actin-sequestration crystallography described above ^[17].

Mechanistic synthesis. The 2012 multi-model review ^[15].

Current standing. A 2026 Sports Medicine narrative review of approved and unapproved peptide therapies for musculoskeletal injury and athletic performance lists TB-500 and thymosin beta-4 among the unapproved peptides, and concludes that many such compounds show favourable tissue-repair outcomes in animal models while rigorous human safety data are scarce, the potential for serious harm is real, and the whole category operates largely outside regulatory oversight ^[13].

Two preclinical results temper the picture further, and both are recorded here because they cut against the compound. In dystrophin-deficient mice, long-term thymosin beta-4 increased the number of regenerating muscle fibres but did not improve muscle strength, cardiac function or fibrosis — more regeneration on paper without better function. And systemic thymosin beta-4 failed to reduce myocardial ischaemia-reperfusion injury in a porcine study. Clinical momentum has also stalled commercially, including a withdrawn injectable acute-stroke trial, so any picture of a busy clinical pipeline overstates the present state.

The one nerve study, and the four things it is not

The rat stroke experiment ^[14] is the closest thing to neural evidence anywhere on this site, so it is worth stating precisely what it does and does not establish.

What it establishes: in rats, an embolic stroke followed by intraperitoneal thymosin beta-4 starting a day later produced measurable neurological improvement lasting to day 56 at two of three amounts tested, with a modelled optimum below the amounts that worked and no benefit at the highest one ^[14].

What it is not:

Not human. The species substitution, unresolved. No human efficacy trial of thymosin beta-4 for stroke has reported in this index, and one injectable acute-stroke programme was withdrawn.

Not TB-500. The molecule substitution, and the sharpest one in the hub. The study used the full-length 43-amino-acid protein. TB-500 is seven of those amino acids. Applying the parent protein's results to the fragment is an extrapolation that has not been confirmed, and the fragment's effects could be weaker, different, or absent.

Not peripheral nerve. The tissue substitution. This is a central-nervous-system injury caused by a blocked artery — a stroke model — and its outcome measure is neurological function scoring after that stroke. It is not a crushed or severed nerve in a limb, and it does not report axonal regrowth.

Not an everyday injury or route. The compartment substitution. Dosing was intraperitoneal, beginning 24 hours after a controlled experimental stroke, on a fixed schedule.

Stack those four and what remains is a real, well-conducted animal experiment that supports further investigation of the parent protein in central nervous system injury. It does not support the claim that the seven-amino-acid peptide sold as TB-500 repairs nerves in people. That claim currently has no study behind it at all.

Reported effects, cautions & safety

What people report. Everything in this subsection is anecdotal, not clinical evidence — self-reported experience from peptide forums, athletic and biohacker community blogs and supplier review pages, with no controlled trial behind it and no dose attached here.

Faster recovery from tendon, ligament and muscle injuries is the main reason people in research-use communities reach for TB-500, and reported timelines vary widely from person to person. Less joint pain and stiffness with easier movement is frequently described, usually after a few weeks, as is a general sense of improved flexibility and physical resilience around three to four weeks in. A smaller number describe reduced swelling or post-workout soreness, or cuts and surgical sites appearing to heal faster — the latter lining up with what animal wound studies show for the parent protein, though the reports themselves remain anecdotes. Hair regrowth is rarely mentioned and is usually confounded by other things being used at the same time.

On the adverse side, a small sore or swollen injection site is by far the most common complaint and is typical of injected peptides generally. Temporary tiredness or lethargy for a day or two is the most consistent systemic report, most often early on. Occasionally people describe a brief head rush, lightheadedness or mild headache shortly after injecting, or a short flu-like feeling in the first day or two. Nausea is rarely reported and tends to come up with larger amounts. Two vaguer reports appear rarely: an old injury feeling more "active" for a week or two, and short-lived mood changes. Neither has any clinical evidence behind it.

Documented cautions. Human safety is essentially unstudied. There are no completed controlled human trials of the TB-500 heptapeptide for any use, and the 2026 sports-medicine review concludes that unapproved peptides of this class show animal-model promise while human safety data are scarce, the potential for serious harm is real, and oversight is largely absent ^[13][16].

The cancer concern is the one most consistently flagged. The parent protein is overexpressed in several cancers and has been linked to tumour spread and to the growth of new blood vessels that feed tumours. The same pro-migration, pro-angiogenic actions that may assist repair could in principle assist tumour progression — mechanism-based reasoning that has never been measured for TB-500 in people, and a reason people with current, past or strong family cancer risk are the group most often told to avoid it.

TB-500 is prohibited by WADA under its peptide and growth-factor categories, and anti-doping laboratories have published methods to detect it and its breakdown products; it is also a prescription medicine in some jurisdictions and has been encountered as a designer drug in racehorses. Reported benefits may also overstate what the compound does: the dystrophin-deficient mouse result above is the standing counter-example, where more regenerating fibres did not translate into better function ^[15].

Because the parent protein influences blood-vessel formation and cell migration and is released by platelets at injury sites, clotting disorders and imminent surgery are precautionary concerns, unstudied for the fragment. The same reasoning — that this molecule acts on cell movement and new blood-vessel growth, both central to development — puts pregnancy, breastfeeding and the still-growing in the avoid-entirely category, with no human data to argue otherwise. Community "loading then maintenance" schedules have no basis in any controlled human trial and are not reproduced on this site.

Where TB-500 stands on the nerve question

TB-500 carries the hub's only neural efficacy result and the hub's weakest claim to it. All four substitutions are active at once: species, molecule, tissue and route ^[14].

It is also the member whose file is most likely to be misread in the optimistic direction, because the parent protein has a legitimate research story — a solved structure ^[17], a coherent mechanism ^[15], and a clean Phase 1 safety record at intravenous doses up to 1,260 mg in healthy volunteers ^[16]. None of that is evidence about the seven-amino-acid fragment, and the 2026 review is explicit that the category as a whole lacks the human safety data it would need ^[13].

The defensible summary is narrow. Full-length thymosin beta-4 improved neurological outcomes in rats after an experimental stroke and was tolerated in a small human safety study. TB-500 is a fragment of that protein with no completed controlled human trial, sold through unregulated channels, banned in tested sport, and carrying a theoretical tumour concern that no one has measured. Anything more confident than that is not coming from the literature.