TB-500 Explained: Mechanism, Research Evidence, and Use Context

TB-500 in brief

Three numbers tell most of the story. How much of the protein TB-500 copies, how many people have taken TB-500 in a study, and what the one animal tendon study used.

Every figure on this page about TB-500 and thymosin beta 4 comes from a PubMed abstract listed at the end. Doses appear only where a study states them. What gym users do is reported as vendor guides and forum posts, never as a recommendation.

What this TB-500 guide covers

What is TB-500, in plain words

Start with the protein it comes from. Thymosin beta 4 is a small protein, 43 building blocks long, found in almost every cell in your body and packed into platelets. Its day job is to hold on to single actin molecules, the raw material of the cell's internal scaffolding. It releases them where the cell needs to move or rebuild (PMID 22074294). When tissue is injured, platelets and immune cells release it at the wound. That is why researchers have studied it for healing since the 1990s.

TB-500 is not that protein. It is a copy of one short piece of it. The piece is the seven building blocks at positions 17 to 23, written LKKTETQ. A small chemical cap, an acetyl group, is added to the front end. That stretch is the part of thymosin beta 4 that grabs actin (PMID 23084823). The product first appeared as a veterinary preparation for racehorses. The first scientific paper about it was a doping test from a Hong Kong racing laboratory in 2012. You will see the name written TB-500, TB500 and TB 500. It is the same product.

Why does the difference matter? Because every human study in this field used the whole 43 building block protein, made by a company called RegeneRx. The whole protein has other active regions besides the actin site. A body also handles a 43 building block protein differently from a seven building block fragment. Nobody has shown that the fragment does in a living person what the whole protein does. It might. It has not been tested.

There is a second twist. Some research vendors sell the full 43 building block protein and call it TB-500. Others sell the seven building block fragment. A 2023 analysis of TB500 and TB1000 products bought online found that the contents were not consistently what the labels and descriptions said (PMID 36482504). So when someone says TB-500 worked for them, you do not know which molecule they took, or whether they took either.

What has been measured on TB-500, and in whom

The record splits cleanly in two. The whole protein, thymosin beta 4, has a real if small human programme: a safety study in healthy volunteers, wound trials, eye drop trials. The fragment, TB-500, has one rat tendon study, a horse doping paper, and a product analysis. The table keeps the two apart.

Read the molecule column first. Every human row says full thymosin beta 4. Every TB-500 row says rats, horses or seized products. There is no row where a person received TB-500. That is the single most important fact on this page.

The thymosin beta 4 human trials, in plain words

These trials matter because they are the only human data anyone can point to. They also show what the molecule was being developed for. None of it was muscle or tendon.

Safety in healthy volunteers. In 2010, forty healthy people were split into four groups of ten. Each group got a single intravenous dose of thymosin beta 4 or placebo, at 42, 140, 420 or 1260 milligrams. After a safety review, they got the same dose every day for 14 days. Side effects were infrequent and mild or moderate. There were no serious adverse events and no dose limiting toxicity. The half-life got longer as the dose went up (PMID 20536472). The company was aiming at heart attack patients, and this was the first step.

Leg ulcers. Seventy three patients with venous ulcers at eight European sites took part. They got a thymosin beta 4 gel or placebo on the ulcer for 84 days, at rising doses. Safety was the same as placebo. The 0.03 percent dose looked like it might speed healing. About a quarter of patients fully healed in three months, mostly those with smaller ulcers (PMID 20536470). The authors called it a suggestion, not a proof.

Dry eye. Two phase 2 trials of 0.1 percent eye drops. The tiny one had nine patients and 18 eyes. It found discomfort down 35.1 percent and corneal staining down 59.1 percent against vehicle at day 56 (PMID 25826322). The larger one, 72 patients for 28 days, missed both of its primary endpoints. Some secondary measures improved, discomfort by 27 percent in a dry chamber test (PMID 26056426). A later phase 3 enrolled 18 patients with a nerve related corneal defect. Complete healing came in 6 of 10 on the drops against 1 of 8 on placebo, p = 0.0656, just short of significance (PMID 36613994).

Skin wounds. A 2016 review from the developers covers phase 2 trials in pressure ulcers, stasis ulcers and epidermolysis bullosa wounds. It reports faster repair and says the protein was safe and well tolerated (PMID 27450738). The protein was first shown to speed skin healing in rats in 1999. Wounds re-covered 42 percent faster at day 4 and up to 61 percent faster at day 7 (PMID 10469335).

What was never tested in people. Tendons, ligaments, muscle, joints, recovery from training. The heart programme produced the safety study and then stalled. A 2013 review called the heart story a possible missed beat. Two mouse studies disagreed on whether the protein makes new heart muscle at all (PMID 22628110). In mice with muscular dystrophy, six months of thymosin beta 4 produced more regenerating fibres. It gave no gain in grip strength, heart function or scarring (PMID 20126456).

TB-500 dosage: what was tested, and what people use

Here is the honest answer to the dosage question. There is no human dose of TB-500, because no human has taken TB-500 in a study. There is one animal dose with a healing result, from 2026. Everything else is either the whole protein in a trial that had nothing to do with tendons, or a vendor guide.

The one TB-500 dose that did something. In the 2026 rat study, 32 rats had their Achilles tendon cut and repaired. Eight got BPC-157 at 10 micrograms per kilogram a day. Eight got TB-500 at 60 micrograms per kilogram a day. Eight got both, and eight got nothing. All injections went into the belly cavity for four weeks. At the end, the TB-500 tendons took more force to break than controls, and that difference reached significance. Their tissue scores were better too (PMID 42542926).

What that converts to on paper. The FDA method for turning a rat dose into a human starting dose divides the milligram per kilogram figure by 6.2. So 60 micrograms per kilogram in a rat becomes about 9.7 micrograms per kilogram in a person. For someone who weighs 80 kg that is roughly 775 micrograms a day. Over a week that is about 5.4 mg. It is arithmetic on one rat study with eight animals per group, not a dose anyone has tested in a person.

Where the online figures come from. Vendor guides and forum posts describe a loading phase of 2 to 2.5 mg under the skin twice a week for four to six weeks. Then 2 mg once or twice a month to maintain (vendor guides, forum posts). That is 4 to 5 mg a week in the loading phase, which happens to sit near the paper conversion above. The resemblance is a coincidence. The forum pattern is years older than the rat study. It came from the veterinary product, where a single horse dose was 10 mg (PMID 23084823).

What the whole protein doses tell you. Not much that transfers. Healthy volunteers tolerated 42 to 1260 mg a day into a vein for two weeks (PMID 20536472). That is a different molecule by a different route, and the point of that study was safety, not healing. The wound and eye trials used gels and drops on the surface of the body, where the dose means something else entirely.

Whatever is in your vial, the syringe arithmetic has a right answer. A 5 mg vial mixed with 1 ml of bacteriostatic water gives 5 mg per ml. Each unit on a U-100 insulin syringe then holds 50 micrograms, so 2 mg is 40 units. The peptide calculator does this for any vial and any volume. It does not tell you how much to take, because that has not been measured.

TB-500 side effects, by the evidence

TB-500 side effects come from three places: what the molecule does, what is in the vial, and the rules around it. For the molecule, the only human safety data belongs to the whole protein, and it is reassuring. For the fragment there is nothing. The vial and the rules carry the concrete risks.

The safety story people quote is the 2010 volunteer study, and it is a good study. Forty healthy people took up to 1260 mg a day of thymosin beta 4 into a vein for two weeks. Side effects were mild (PMID 20536472). But that was pharmaceutical grade whole protein, made under supervision, in a trial. The fragment in a research vial is a different molecule from an unregulated source. The 2023 analysis of TB500 products found the contents did not consistently match the descriptions (PMID 36482504). You cannot borrow a safety result from a molecule you may not even be taking.

The theoretical worry is the one the protein's own biology raises. Thymosin beta 4 heals by growing new blood vessels, recruiting progenitor cells and moving cells into a wound (PMID 22074294). A 2026 review of unregulated injectable peptides lists pathological angiogenesis and carcinogenesis among the mechanisms of concern for compounds including TB-500 (PMID 42757290). Nobody has shown that happens. Nobody has run the study that would rule it out either. The general page on peptide side effects covers the sterility and injection problems shared by every mixed vial.

BPC-157 and TB-500: what the stack has behind it

The BPC-157 and TB-500 pairing, sold as a blend and nicknamed the Wolverine stack online, is the form most people actually buy. The logic is that the two work differently: BPC-157 acts on growth factor and nitric oxide signalling, TB-500 on actin and cell movement. Two paths, one wound. The evidence for the pair is one study, and it cuts the other way from what you might expect.

In the 2026 rat Achilles study, the rats that got both peptides did not do better than the rats that got TB-500 alone. TB-500 on its own gave the significant gain in tendon strength. The combination group beat controls on one of two tissue scores. It did not beat TB-500 alone on strength (PMID 42542926). Eight rats per group, so read it as a hint, not a verdict. The hint is that stacking added nothing in that model.

In people, the pair has one data point and it is weaker still. A chart review followed 16 patients who got BPC-157 injected into a painful knee. Four of them also got TB-500, and those four reported slightly less relief than the twelve on BPC-157 alone. No controls, no dose reported, phone follow up. The full BPC-157 story, including that study, is on BPC-157 dosage explained, and its stack section is at BPC-157 and TB-500.

TB-500 in the gym: what users report, and what is measured

The trials cannot write this section, so it reports two things separately. First, how users describe TB-500. Second, the measured data that any of it could rest on. Nothing in the first part is evidence. It is a description of a habit.

What users describe. Vendor guides and forum logs describe TB-500 as the peptide for a nagging tendon, a shoulder that will not settle, or a pulled muscle. The pattern is a loading phase of 2 to 2.5 mg twice a week for four to six weeks, then 2 mg once or twice a month. Users inject it under the skin anywhere rather than near the injury, because they believe it acts through the whole body. Most pair it with BPC-157. Runs are described as short, tied to the injury, and repeated when something else hurts (vendor guides, forum posts). The reasons given are faster recovery, less inflammation and more range of motion. The complaints are headache and fatigue after a shot and a wave of tiredness on injection day. A common verdict is that it is hard to tell what the peptide did and what time did (forum posts).

What the reasons rest on. The whole body idea comes from the protein's biology. Thymosin beta 4 is released by platelets at a wound and travels. In rats it sped skin healing whether applied on the wound or injected into the belly (PMID 10469335). The tendon idea rests on the rat Achilles study (PMID 42542926) and the rat knee ligament study. There, 1 microgram of the whole protein in fibrin glue improved the ligament at four weeks (PMID 23523891). The recovery and inflammation ideas rest on the protein's general properties in reviews (PMID 22074294). None of this was measured in a person with a sore shoulder.

What is measured that a lifter could use. One thing, and it is indirect. TB-500 at 60 micrograms per kilogram a day for four weeks made a surgically cut rat tendon stronger (PMID 42542926). A cut and sutured tendon in a rat is not an overuse tendon in a person. Four weeks of daily belly injections is not two shots a week. But it is the first controlled result for the fragment in any animal, and it points the right way. How to read a result like that without overreaching is on peptide research vs human use.

Five mistakes people make about TB-500

1. Calling it thymosin beta 4

TB-500 is a seven building block fragment of a 43 building block protein (PMID 23084823). Every human trial used the whole protein. The fragment has no human study.

2. Quoting the volunteer safety study for the vial

That study gave pharmaceutical grade thymosin beta 4 into a vein under supervision (PMID 20536472). The vial is a different molecule from an unregulated source, and a lab found the contents did not match the labels (PMID 36482504).

3. Treating the eye and ulcer trials as healing proof

The largest dry eye trial missed its primary endpoints (PMID 26056426). The ulcer trial called its result a suggestion (PMID 20536470). The phase 3 eye trial had 18 patients and p = 0.0656 (PMID 36613994). Real, small, and nothing to do with tendons.

4. Calling the Wolverine stack proven synergy

In the one controlled test, rats on both peptides did not beat rats on TB-500 alone (PMID 42542926). The four humans who got both in a knee chart review reported slightly less relief than those on BPC-157 alone.

5. Assuming it is legal to compete on

Thymosin beta 4 and TB-500 are banned by WADA and by horse racing authorities. Both the fragment and a synthesis impurity can be detected (PMID 39314109, PMID 23084823).

Verdict: a real protein, a borrowed story, and one rat study

Judged by the papers rather than the vendor pages, thymosin beta 4 is a serious molecule with a thin but real human record. It is safe at very high intravenous doses for two weeks. It may speed the healing of skin ulcers and eye defects. It has not yet earned an approval anywhere (PMID 20536472, PMID 20536470, PMID 36613994). TB-500 is a piece of that protein. It was first sold for horses, first studied by a doping lab, and tested for healing exactly once, in 32 rats, in 2026 (PMID 23084823, PMID 42542926). The one study went its way.

For readers here, the practical conclusion is simple. The dosage online is a horse protocol scaled by habit. The safety record belongs to a different molecule. The vial may not contain what it says. The stack with BPC-157 showed no advantage in the only test. The other half of that stack is on BPC-157. The whole group is on healing peptides. The syringe arithmetic for any vial is on the peptide calculator.