TB-500 and Alzheimer's Disease: What the Thymosin Beta-4 Neurodegeneration Research Shows

Q: Why is thymosin beta-4 linked to microglia?

Thymosin beta-4 is expressed in microglia, the brain's immune cells, and is upregulated when they are activated. Studies summarized in a 2018 review by Pardon reported that it can suppress pro-inflammatory signaling in microglia, and Wang 2021 linked its effects in APP/PS1 mice to reduced TLR4/NF-kB signaling. Because chronic microglial inflammation is implicated in Alzheimer's, that link is the main reason researchers study it in this disease.

> Research disclaimer: This article summarizes published laboratory and animal research for informational purposes only. TB-500 is not an FDA-approved drug, it has not been tested for Alzheimer's disease or any other neurodegenerative disease in people, and nothing here is medical advice, a dosing protocol, or a claim that it treats or prevents any condition. Anyone concerned about memory loss should see a physician.

The short answer

There is no human evidence that TB-500 or thymosin beta-4 does anything for Alzheimer's disease. What exists is a small, recent cluster of preclinical papers. In mouse models engineered to build up amyloid, thymosin beta-4 (Tβ4) or TB-500 was associated with less brain inflammation, healthier-looking neurites and better scores on memory tasks. A 2025 study in human brain organoids found that neurons from familial Alzheimer's models made less of the gene that encodes Tβ4, and that adding Tβ4 rescued some of the defects. Those are interesting leads. They are also exactly the kind of results that have a long history of not carrying over to patients. A search of ClinicalTrials.gov on October 30, 2026 found no registered trial of thymosin beta-4 or TB-500 in Alzheimer's, Parkinson's or ALS.

---

Why researchers looked at thymosin beta-4 in the brain at all

Tβ4 is a 43-residue protein found in most cells. Its best-known job is binding monomeric actin, which helps control how cells change shape and move. It is also one of the more abundant small proteins in the central nervous system, where it is expressed in neurons and in microglia, the brain's resident immune cells.

That second location is what draws Alzheimer's researchers. A 2018 review by Marie-Christine Pardon at the University of Nottingham (Expert Opinion on Biological Therapy) laid out the case: Tβ4 is upregulated in activated microglia, it appears to suppress pro-inflammatory signaling in those cells, and chronic microglial inflammation is now treated as a driver of neurodegenerative disease rather than a bystander. Pardon's review framed Tβ4 as a candidate for controlling brain inflammation, while stating plainly that its role in the healthy and diseased brain was "poorly understood."

The other thread comes from injury research. Rat studies of traumatic brain injury and embolic stroke reported better functional recovery and more new blood vessels and oligodendrocytes after Tβ4 treatment; our TB-500 traumatic brain injury and stroke guide covers those papers. Alzheimer's is a different problem, a slow, decades-long loss of synapses and neurons, but the inflammation and repair questions overlap.

A naming problem you need to solve first

Most of the papers below used full-length thymosin beta-4, either as a protein or by overexpressing its gene. TB-500, as identified by doping-control chemists, is a seven-residue acetylated fragment, Ac-LKKTETQ, matching residues 17-23 of Tβ4. Only one of the Alzheimer's studies we found tested a product its authors called TB500 directly. When you read "TB-500 Alzheimer's" claims online, check which molecule the underlying study used. Our TB-500 vs thymosin beta-4 guide explains the difference in detail.

---

The Alzheimer's studies, one by one

Zhang et al., PLoS One (2023): neuron-like cells exposed to amyloid

This was a cell-culture study. The authors exposed SH-SY5Y neuroblastoma cells, a common stand-in for human neurons, to amyloid-beta (Aβ) to mimic Alzheimer's-type stress. They then raised Tβ4 levels with a lentivirus or lowered them with siRNA. Cells with more Tβ4 survived better and showed less apoptosis: lower caspase-3, caspase-8 and Bax, higher Bcl-2. Oxidative-stress markers moved in a protective direction too, with lower MDA, LDH and reactive oxygen species and higher SOD and GSH-Px. The authors tied the effect to inhibition of the ERK/p38 MAPK pathway and increased expression of the serotonin receptor 5-HTR1A.

What it shows: in a dish, more Tβ4 inside the cell made amyloid-stressed neuron-like cells more resilient. What it doesn't: anything about an added peptide crossing the blood-brain barrier, or about a living brain.

Wang et al., Journal of Neuroinflammation (2021): APP/PS1 mice

APP/PS1 mice carry human mutations that make them deposit amyloid plaques as they age. Wang and colleagues reported that Tβ4 protein levels were already elevated in these mice, which fits the idea that the brain raises Tβ4 as part of its response to amyloid. Overexpressing Tβ4 further was associated with less Aβ accumulation, more insulin-degrading enzyme (one of the enzymes that breaks down Aβ), a shift of microglia and astrocytes back toward a resting state, better neuronal function and better performance on learning and memory tests. The authors linked these changes to downregulation of TLR4/MyD88/NF-κB signaling, and adding a TLR4 antagonist or an NF-κB inhibitor produced no further benefit, which they read as Tβ4 working through that pathway.

This is gene overexpression, not a peptide given from outside the body, and it was done in one transgenic strain. Still, it's the paper most often cited for a Tβ4-microglia link in Alzheimer's models.

Othman et al., International Immunopharmacology (2023): infection on top of amyloid

Pardon's group asked a narrower question. Systemic infection and gut-derived lipopolysaccharide (LPS) are thought to speed up Alzheimer's progression. Since Tβ4 has been studied for reducing LPS-driven inflammation in sepsis models, they gave middle-aged APP/PS1 mice and normal littermates an LPS challenge, then treated some with intravenous Tβ4 over the following week.

Tβ4 limited LPS-induced weight loss and loss of a natural food-burrowing behavior, more so in the APP/PS1 mice than in normal mice. It prevented the LPS-induced increase in amyloid plaque load in APP/PS1 mice. But in normal mice given LPS, Tβ4 increased astrocyte and microglia proliferation in the hippocampus. That mixed result is worth remembering: the same protein that calmed one inflammatory picture amplified glial growth in another.

Ou et al., International Immunopharmacology (2026): TB500 and Ac-SDKP in 5xFAD mice

This is the study that tested something its authors called TB500, alongside Ac-SDKP, another Tβ4-derived peptide (the N-terminal tetrapeptide; see our TB-500 vs Ac-SDKP comparison). The work had three layers:

Neuron cultures. In HT22 hippocampal cells and primary cortical neurons exposed to an amyloid fragment (Aβ25-35), both peptides reduced neurite atrophy, restored cell viability and shifted apoptosis-related gene expression.

Microglia cultures. In BV2 microglia stimulated with LPS, both peptides reduced nitric oxide production and pro-inflammatory cytokines and inhibited "M1" inflammatory polarization.

5xFAD mice. 5xFAD mice carry five familial Alzheimer's mutations and develop amyloid early. Treated mice did better on the Morris water maze and novel object recognition tests, and their brains showed less glial activation, less neuronal apoptosis, restored axonal density in the perirhinal cortex and fewer plaque-associated dystrophic neurites.

The detail that matters most is buried in the abstract: hippocampal Aβ burden did not change. Whatever the peptides did in these mice, they did not clear amyloid from the hippocampus. The authors frame the effects as neuroprotection, anti-inflammation and axonal regeneration rather than amyloid removal. Transcriptomic profiling flagged genes such as Foxb2 and Or2k2, which are leads for mechanism work, not established targets.

Zeng et al., Stem Cell Reports (2025): human brain organoids

The most novel study came from ShanghaiTech University. The team grew cerebral organoids, small three-dimensional brain-like tissues, from induced pluripotent stem cells carrying amyloid precursor protein (APP) mutations linked to familial Alzheimer's. Compared with healthy-control organoids, the familial Alzheimer's organoids had markedly fewer mature neurons, more cellular senescence and more Aβ production.

Two findings stand out. First, expression of TMSB4X, the gene that encodes Tβ4, was significantly lower in the organoids' neurons and in excitatory neurons from people with Alzheimer's disease. Second, treating the organoids with Tβ4 rescued the neurodevelopmental deficits and reduced Aβ formation, and the authors reported benefits in 5xFAD mice as well.

Two caveats. Organoids model early brain development, and whether Alzheimer's has a developmental origin is, in the authors' own words, "arguably debated." And the paper discloses that a patent application on using TMSB4X for Alzheimer's treatment has been filed by the university, with two of the authors as inventors. That's normal in academic research and doesn't invalidate the work, but it's a disclosure readers should see.

---

Putting the Alzheimer's evidence in perspective

Reading these five papers together, a few patterns are clear:

The signal is consistent but early. Every study reported a protective direction, across cells, two mouse strains and organoids. That consistency is the strongest thing the literature has going for it.

Almost all of it is full-length Tβ4. Only Ou et al. tested a TB500-labeled peptide, so most of the evidence belongs to the parent protein, not the fragment.

Amyloid effects are inconsistent. Wang and Zeng reported less Aβ; Othman reported prevention of an LPS-driven plaque increase; Ou found no change in hippocampal Aβ. The common thread is inflammation and neuron survival, not plaque clearance.

Brain access is unanswered. None of these studies settles whether a peptide given outside the brain reaches it in meaningful amounts in people.

Mouse models have a poor track record. Transgenic amyloid mice have produced many positive results for compounds that later failed in human trials. Better memory-test scores in 5xFAD or APP/PS1 mice are a reason to keep researching, not a prediction of clinical benefit.

For readers comparing this to TB-500's other research areas, keep in mind that most of the repair literature rests on Tβ4's actin-binding and cell-migration effects, which are a different question from neuroinflammation.

---

Sourcing research material

If you're sourcing TB-500 for laboratory work, identity comes before purity. Apollo Peptide Sciences publishes lot-level third-party certificates of analysis; check that the listed sequence reads Ac-LKKTETQ and that the mass-spectrometry result matches it before relying on the HPLC purity figure. Studies like Wang 2021 and Zeng 2025 used full-length Tβ4, which is a different molecule. Affiliate disclosure: we may earn a commission if you buy through this link, at no extra cost to you.

---

Parkinson's, ALS and dementia: biomarkers, not treatments

Outside Alzheimer's, thymosin beta-4 shows up in neurodegeneration research mostly as a measurement, not an intervention:

Parkinson's disease. A 2021 plasma proteomics study (Zhao et al., Proteomics Clinical Applications) quantified 555 proteins in people with Parkinson's and healthy controls and listed TMSB4X among proteins implicated in Parkinson's pathogenesis. The authors called it a preliminary biomarker study. A 2025 zebrafish study (Fiametti et al., International Journal of Molecular Sciences) examined synthetic intracellular peptides from several precursor proteins, including thymosin beta-4, in a 6-OHDA Parkinson's model; it found the peptides were not cytotoxic in cell lines, which is a starting point rather than a result.

ALS. A 2013 imaging mass-spectrometry study of post-mortem spinal cord (Hanrieder et al., Journal of Neurochemistry) mapped Tβ4 mainly to gray matter in normal tissue. The disease-linked change it highlighted was a truncated ubiquitin form, not Tβ4.

Creutzfeldt-Jakob disease. A 2012 study (Le Pera et al., Archives of Neurology) found cerebrospinal-fluid Tβ4 markedly higher in confirmed CJD than in frontotemporal dementia, Alzheimer's disease or rapid-onset progressive dementia, and proposed it as a diagnostic marker for CJD.

Rising Tβ4 levels in diseased tissue can mean the body is mounting a protective response, or that Tβ4 is part of the damage, or simply that inflamed microglia make more of it. Biomarker studies can't tell those apart. Our multiple sclerosis and remyelination article covers the one neurological area with more treatment-style animal data.

What a human trial would need to answer

No registered trial exists, so it's worth being specific about what is missing:

1. Which molecule. Full-length Tβ4, the TB-500 fragment and Ac-SDKP behave differently. A trial has to pick one, and the published Alzheimer's data mostly supports the full protein.
2. Whether it reaches the brain. Plasma levels say little about brain exposure. Cerebrospinal-fluid or imaging measures would be needed.
3. Safety over years. Alzheimer's is chronic. The human safety data for Tβ4 comes from short trials in dry eye, wound healing and heart attack, not from long-term exposure in older adults. Tβ4's role in cell migration and angiogenesis also raises questions in tumor biology that a long trial would need to monitor.
4. The right endpoint. Animal studies measured maze performance and glial activation. Human trials need validated cognitive and functional outcomes, ideally alongside amyloid and tau biomarkers.

Regulatory and sport status in 2026

FDA. TB-500 is not an FDA-approved drug for any use. On April 15, 2026, FDA said it would remove the thymosin beta-4 fragment from its Category 2 list of bulk substances that raise significant safety concerns. On July 23, 2026, FDA's Pharmacy Compounding Advisory Committee voted 8-6, with one abstention, to recommend adding TB-500 to the 503A bulks list, against the recommendation of FDA's review staff. The vote is advisory; listing would still require rulemaking, and none of it is an approval for Alzheimer's or anything else.

Sport. WADA's 2026 Prohibited List bans thymosin beta-4 and its derivatives under S2, naming TB-500, at all times.

Research-chemical sales. "For research use only" is a sales category, not a quality or safety signal.

---

Frequently asked questions

Does TB-500 help Alzheimer's disease?

There is no human evidence that it does. TB-500 and thymosin beta-4 have never been tested in a clinical trial for Alzheimer's disease. A handful of cell, organoid and mouse studies published between 2021 and 2026 reported less neuroinflammation, better neuron survival and better memory-test performance in amyloid mouse models, but mouse-model results in Alzheimer's have a long record of not translating to patients.

Is there a clinical trial of thymosin beta-4 for Alzheimer's or Parkinson's?

Not that we could find. A ClinicalTrials.gov search for thymosin beta-4 on October 30, 2026 returned trials in dry eye, neurotrophic keratopathy, pressure and venous ulcers, epidermolysis bullosa, heart attack, healthy volunteers and one cardiovascular biomarker study of the TB-500 fragment, but none in Alzheimer's disease, Parkinson's disease or ALS.

Did the Alzheimer's studies use TB-500 or full-length thymosin beta-4?

Mostly full-length thymosin beta-4. Wang 2021 and Zhang 2023 raised Tβ4 levels by overexpression, Othman 2023 gave Tβ4 protein, and Zeng 2025 treated organoids with Tβ4. Ou et al. (2026) is the one study that tested a peptide its authors called TB500, alongside Ac-SDKP, in 5xFAD mice.

Does thymosin beta-4 remove amyloid plaques?

The results are mixed. Wang 2021 and Zeng 2025 reported less amyloid-beta, and Othman 2023 found Tβ4 prevented an infection-driven increase in plaque load. But Ou 2026 found that TB500 and Ac-SDKP did not change hippocampal amyloid burden in 5xFAD mice even though memory scores and neurite health improved. The more consistent findings are about inflammation and neuron survival.

Why is thymosin beta-4 linked to microglia?

Tβ4 is expressed in microglia, the brain's immune cells, and is upregulated when they are activated. Studies summarized in a 2018 review by Pardon reported that Tβ4 can suppress pro-inflammatory signaling in microglia, and Wang 2021 linked its effects in APP/PS1 mice to reduced TLR4/NF-κB signaling. Because chronic microglial inflammation is implicated in Alzheimer's, that link is the main reason researchers study Tβ4 in this disease.

Is thymosin beta-4 a biomarker for dementia?

It has been studied as one. A 2012 study found cerebrospinal-fluid Tβ4 markedly higher in Creutzfeldt-Jakob disease than in Alzheimer's or frontotemporal dementia, and a 2021 plasma proteomics study listed it among proteins implicated in Parkinson's disease. Neither is an approved diagnostic test.

Before you buy anything for a research project

Start with the molecule your protocol actually calls for. If it's the TB-500 fragment, Apollo Peptide Sciences posts third-party COAs for its TB-500 lots; confirm the sequence and measured mass match Ac-LKKTETQ. If it's full-length thymosin beta-4, a TB-500 vial is the wrong material. Affiliate disclosure: we may earn a commission if you buy through this link, at no extra cost to you.

---

Sources: Zhang GH, Chin KL, Yan SY, Pare R, "Antioxidant and antiapoptotic effects of Thymosin β4 in Aβ-induced SH-SY5Y cells via the 5-HTR1A/ERK axis," PLoS One 2023;18(10):e0287817, doi:10.1371/journal.pone.0287817; Wang M, Feng LR, Li ZL, et al., "Thymosin β4 reverses phenotypic polarization of glial cells and cognitive impairment via negative regulation of NF-κB signaling axis in APP/PS1 mice," J Neuroinflammation 2021;18(1):146, doi:10.1186/s12974-021-02166-3; Othman O, Marshall H, Masterson M, et al., "Thymosin beta 4 prevents systemic lipopolysaccharide-induced plaque load in middle-age APP/PS1 mice," Int Immunopharmacol 2023;117:109951, doi:10.1016/j.intimp.2023.109951; Ou H, Chen R, Zhou L, et al., "Thymosin β4-derived peptides alleviate neuroinflammation and neurite atrophy in both in vitro models and in vivo 5 × FAD mice," Int Immunopharmacol 2026;170:116097, doi:10.1016/j.intimp.2025.116097; Zeng PM, Sun XY, Li Y, et al., "Thymosin beta 4 as an Alzheimer disease intervention target identified using human brain organoids," Stem Cell Reports 2025;20(9):102601, doi:10.1016/j.stemcr.2025.102601; Pardon MC, "Anti-inflammatory potential of thymosin β4 in the central nervous system," Expert Opin Biol Ther 2018;18(sup1):165-169, doi:10.1080/14712598.2018.1486817; Zhao Y, et al., Proteomics Clin Appl 2022;16(2):e2100010, doi:10.1002/prca.202100010; Fiametti LO, et al., Int J Mol Sci 2025;26(5):2017, doi:10.3390/ijms26052017; Hanrieder J, et al., J Neurochem 2013;124(5):695-707, doi:10.1111/jnc.12019; Le Pera M, et al., Arch Neurol 2012;69(7):868-872, doi:10.1001/archneurol.2011.3558; Esposito S, et al., Drug Test Anal 2012, doi:10.1002/dta.1402; ClinicalTrials.gov search "thymosin beta 4" (accessed October 30, 2026); FDA, "Certain Bulk Drug Substances for Use in Compounding that May Present Significant Safety Risks"; WADA, 2026 Prohibited List (S2).