TB-500 and Brain Research: The One Study That Used the Fragment

Published August 23, 2026 · Artemis Labs

TB-500 and brain research — answer capsule: TB-500 is a seven-amino-acid piece of a larger protein called thymosin β4. Only one published study, Ou and colleagues (2026), gave TB-500 itself to animals and measured brain outcomes. It reported better memory scores in Alzheimer’s-model mice, and no change in hippocampal amyloid-beta burden. Every other brain paper here used the full-length parent protein. No human trial of TB-500 exists.

Key findings

• One study, one fragment. Ou 2026 (PMID 41443105) is the only paper in this record that gave TB-500 itself and measured brain and nerve-cell outcomes.
• The same paper carries a null. Ou 2026 reported that “hippocampal Aβ burden remained unchanged” — the plaque measurement did not move, even though the behavioural scores did.
• The blood-brain-barrier work is cells, not people. Stewart 2025 (PMID 41326489) used full-length thymosin β4 on human brain endothelial cells in culture, and the protective effect disappeared when the S1PR1 receptor was blocked.
• It may not even be TB-500 doing the work. Rahaman 2024 (PMID 38382158) opens with the line “the biological effects of TB-500, however, have not been documented.”

What is TB-500, and how is it different from thymosin β4?

TB-500 is a lab-made peptide — a short chain of amino acids, the building blocks that link together to form proteins. Its chain is seven links long, written Ac-LKKTETQ. The “Ac” is an acetyl cap on one end, a small chemical group that adds about 42 daltons. Its molecular weight is 889.0 g/mol (PubChem CID 62707662, CAS 885340-08-9).

Thymosin β4 is the parent protein it came from. That chain is 43 amino acids long and weighs about 4,963 daltons, roughly five and a half times heavier. TB-500 matches positions 17 through 23 of it. The two are related, and they are not the same molecule.

This distinction shapes the whole page. Almost every brain study on record used the 43-residue parent, not the seven-residue fragment. A 2026 sports-medicine review lists them as two separate entries among unapproved compounds: “Tβ4 (thymosin beta-4), and TB-500 (thymosin beta-4 fragment)” (Mendias 2026, PMID 41966639). The full comparison is on TB-500 vs thymosin β4.

Which brain study actually used TB-500 itself?

Ou and colleagues (2026) is the single paper in this record that administered TB-500 and measured brain outcomes. It is a mouse-and-cell study. No people were involved at any stage.

The work had three parts. First, TB500 and a second thymosin β4 fragment called Ac-SDKP were applied to HT22 cells (a mouse nerve-cell line) and to primary cortical neurons, both damaged with Aβ25-35, a piece of the amyloid-beta protein that builds up in Alzheimer’s disease. The paper reported less neurite atrophy — neurites are the branching arms nerve cells use to reach each other — and restored cell viability.

Second, BV2 microglia, the brain’s own immune cells, were stimulated with LPS to provoke inflammation; the paper reported suppressed nitric oxide output and lower pro-inflammatory cytokine levels. Third, 5×FAD transgenic mice, bred to develop Alzheimer’s-like changes early, performed better on the Morris water maze and on novel object recognition, two standard rodent memory tests. The authors also reported reduced glial activation and restored axonal density in the perirhinal cortex, a region involved in recognising objects.

What did that same study fail to find?

Ou 2026 reported a result that ran against its own headline: hippocampal Aβ burden remained unchanged.

The hippocampus is the brain’s main memory structure, and amyloid-beta plaque buildup there is the defining marker of Alzheimer’s disease in these mouse models. Here the memory scores moved and the plaque measurement did not, so the study’s own data does not attribute the behavioural change to plaque clearance. We publish that line because leaving it out would misrepresent the paper.

What does the human-cell work on the blood-brain barrier show?

Stewart and colleagues (2025) worked with human brain microvascular endothelial cells, which line the tiny blood vessels of the brain and form most of the blood-brain barrier — the filter that decides what passes from the bloodstream into brain tissue. The researchers deprived the cells of oxygen, a condition called hypoxia. Treating them with thymosin β4 beforehand reversed the resulting damage to barrier components. When the team blocked a receptor called S1PR1, that protection went away, which points to S1PR1 as the pathway the effect travelled through.

Two limits travel with this citation. The compound tested was full-length thymosin β4, not the TB-500 fragment. And “human” here means human cells in a dish, not human patients — nobody was treated. The authors’ own language is careful: thymosin β4 “may be tested as a potential treatment modality and warrant further investigation.”

What do the other nerve and brain studies cover?

Several more papers sit in this record. Every one used the full-length parent protein or a different fragment, and none involved people. Rats with traumatic brain injury received thymosin β4 in Xiong 2011 (PMID 20486893). Mice carrying the APP/PS1 Alzheimer’s model, then challenged with LPS, received it in Othman 2023 (PMID 36878045). SJL/J mice in an EAE model, the standard rodent stand-in for multiple sclerosis, received it in Zhang 2009 (PMID 19782721).

At the cell level, Li 2022 (PMID 35979771) treated rat spinal-cord neural stem and progenitor cells — the cells capable of becoming new nerve cells — with thymosin β4 while injuring them with hydrogen peroxide. Rat cells in a dish, parent protein, no people.

One study is easy to mix up with TB-500. Zhang 2017 (PMID 28245754) tested rats after traumatic brain injury using AcSDKP, the 1–4 fragment of thymosin β4. It is a third molecule, neither TB-500 nor the parent, and the same one Ou 2026 tested alongside TB500.

What the research does not show

No published human clinical trial of TB-500 exists. Not for memory, not for brain injury, not for anything else.

Ou 2026 is a cell and rodent study, and its own amyloid measurement did not change. Stewart 2025 is cultured cells, not patients. The rest is rats and mice given the parent protein, a different molecule from the one Artemis sells.

There is a deeper open question. Rahaman 2024 followed TB-500 through human serum, enzyme systems, and the urine of treated rats, then concluded that “the previously reported wound-healing activity of TB-500 in literature may be due to its metabolite Ac-LKKTE rather than the parent form.” It is not settled that the seven-residue molecule sold as TB-500 is the active species at all.

Counter-evidence exists too. Sun and colleagues (2025) reported that thymosin β4 released by mast cells reduced tight junction proteins and impaired intestinal epithelial barrier integrity in rats and mice, with elevated Tβ4 measured in the colonic mucus of IBS patients. That study examined endogenous full-length Tβ4, not the TB-500 fragment, and had no human dosing arm. We cover it on TB-500 safety research. Mendias 2026 puts the position for this compound class bluntly: “rigorous human safety data are scarce, and there is potential for serious harm to patients.”

TB-500 is not sold on its own at Artemis Labs. The canonical listing is the BPC-157 and TB-500 research pair, supplied as two separate vials with a lot-specific certificate of analysis.

Frequently asked questions

Has TB-500 been tested in people for any brain condition?

No. There is no published human clinical trial of TB-500 for brain or nervous-system research, and none for anything else either. The one paper that gave TB-500 to animals and measured brain outcomes is Ou 2026.

Is thymosin β4 research the same as TB-500 research?

No, and treating it that way is the most common error on this compound. Thymosin β4 is a 43-amino-acid protein; TB-500 is a seven-amino-acid piece of it with an acetyl cap. A finding in one is not automatically a finding in the other, which is why every study here is labelled with the form it used.

Did the mouse memory study show TB-500 cleared amyloid plaque?

It did not. Ou 2026 reported improved scores on two rodent memory tests and, in the same paper, that hippocampal Aβ burden remained unchanged.

Why does Ac-SDKP keep showing up next to TB-500?

Ac-SDKP is a different fragment of the same parent protein, from positions 1 through 4 rather than 17 through 23. Ou 2026 tested both. It is a third molecule, and no page should merge it with TB-500.

References

1. Ou et al. (2026). Thymosin β4-derived peptides TB500 and Ac-SDKP in Alzheimer’s models. Int Immunopharmacol. PMID 41443105 · DOI 10.1016/j.intimp.2025.116097
2. Stewart et al. (2025). Thymosin β4 and S1PR1-dependent protection of human brain microvascular endothelial cells. Sci Rep. PMID 41326489 · DOI 10.1038/s41598-025-28435-2
3. Rahaman et al. (2024). Metabolism of TB-500 in human serum, enzyme systems, and rat urine. PMID 38382158
4. Xiong et al. (2011). Thymosin β4 in a rat traumatic brain injury model. PMID 20486893
5. Othman et al. (2023). Thymosin β4 in APP/PS1 mice with LPS challenge. PMID 36878045
6. Zhang et al. (2009). Thymosin β4 in SJL/J mice, EAE model. PMID 19782721
7. Zhang et al. (2017). AcSDKP in a rat traumatic brain injury model. PMID 28245754
8. Li et al. (2022). Thymosin β4 and rat spinal-cord neural stem/progenitor cells under H2O2 injury. PMID 35979771
9. Mendias et al. (2026). Unapproved peptides in sports medicine. Sports Med. PMID 41966639
10. Sun et al. (2025). Thymosin β4 released by mast cells impairs the intestinal epithelial barrier in IBS. World J Gastroenterol. PMID 41278163 · DOI 10.3748/wjg.v31.i42.111706
11. Sosne et al. (2010). Review describing LKKTETQ as the central actin-binding domain of thymosin β4. PMID 20179146
12. Philp et al. (2003). LKKTETQ, a seven-amino-acid synthetic peptide, in aged mice. PMID 12581423

Methodology: this page draws on the Artemis Labs TB-500 verified facts sheet, built from PubMed abstracts retrieved through NCBI E-utilities and compound identity confirmed against PubChem. Every citation was checked against the abstract it is attached to. Last verified August 23, 2026.

All compounds sold by Artemis Labs are for laboratory research use only. Nothing on this page is medical advice, and no statement has been evaluated by the FDA.