Follistatin 344

Origin and structure

Follistatin was isolated in 1987 at the Salk Institute in California, from the fluid of pig ovarian follicles. The researchers named it after its effect in the lab: it suppressed the release of follicle-stimulating hormone (FSH), but not of luteinising hormone, from rat pituitary cells. It later proved to bind activin and several related growth factors, and in 2001 a group at Johns Hopkins showed that it also blocks myostatin, the protein that limits muscle growth.

The human gene gives two precursors through alternative splicing, of 344 and 317 amino acids. Each begins with a 29-amino-acid signal sequence that the cell cuts off before release, so FS344 becomes FS315, the form that predominates in blood, and FS317 becomes FS288, which works mainly in the reproductive system. The mature FS315 is a glycoprotein of 315 amino acids held together by 18 disulfide bridges, far larger than the short peptides that make up most of this wiki. The name follistatin-344 therefore describes the precursor, the gene product used in the gene-therapy trials, rather than the protein that circulates in the body.

How it works

Follistatin is a binding protein, not a hormone with a receptor of its own. Outside the cell it wraps around members of the TGF-beta family, including activin, myostatin, GDF11 and some bone morphogenetic proteins, and covers the site they use to reach their receptors, so they can no longer signal. Myostatin normally acts as a brake on muscle growth. Mice engineered to overproduce follistatin in their muscles gained muscle mass on a scale comparable to mice that lack myostatin altogether.

Because follistatin was discovered as a substance that lowers FSH, its effect on the reproductive hormones has always been the main theoretical concern. The gene-therapy researchers chose FS344 for that reason: its product, FS315, circulates in the blood, while FS288 acts mainly in reproductive tissue. In their Becker and inclusion body myositis trials, FSH, LH, testosterone and oestrogen were measured repeatedly and did not change.

How long follistatin stays active depends on how it is delivered. A gene carried into muscle by an AAV vector keeps making the protein for months: in the monkey study, follistatin levels in the injected thigh muscle were still clearly raised 15 months later. The native protein given by injection behaves differently. Researchers at Lilly found that native FS315 was poorly suited to work as an injectable medicine in mice; only after they fused it to an antibody fragment and removed its heparin-binding site, which lengthened its half-life about 100-fold, did weekly injections produce a robust muscle effect.

Is follistatin-344 approved?

No. No medicine containing follistatin is authorised in the EU, and the FDA has not approved one in the United States. Follistatin gene therapy has only been given in investigational trials. In the Netherlands it is not a registered medicine, so it may not be sold, advertised or supplied for use in people; it exists only as research material.

In sport it is prohibited at all times. The 2026 WADA Prohibited List names follistatin as an example of a myostatin-binding protein in class S4.3, agents preventing activin receptor IIB activation. Introducing a follistatin gene would also count as gene doping under class M3. The Dopingautoriteit's Dutch translation of the 2026 list names follistatine in the same class, and the Dutch anti-doping rules apply that list.

What the research shows

The animal evidence is strong and consistent. Besides the transgenic mice, a single injection of an AAV1 vector carrying FS344 into the thigh of cynomolgus monkeys (six treated animals, followed for up to 15 months) produced lasting increases in muscle size and strength. In the three monkeys given the stronger construct, thigh circumference rose 15% within 8 weeks, and no abnormal changes were found in the key organs.

The human data all concern gene therapy, not injected protein. The published trials come from one group at Nationwide Children's Hospital in Columbus, Ohio, which injected the AAV1.CMV.FS344 vector into the thigh muscles of people with muscle diseases. None had a placebo group, all were small, and one of the investigators had an equity interest in work licensed to Milo Biotechnology and consulted for the company. Commercial plasmid versions have since been given to healthy volunteers, mostly in Honduras, without published results.

• Becker muscular dystrophy (Mendell et al., 2015): six patients at two dose levels. Four walked further on the 6-minute walk test, by 58, 125, 108 and 29 metres; two did not improve. Biopsies showed less fibrosis and larger muscle fibres, especially at the higher dose. No adverse events were attributed to the treatment.
• Sporadic inclusion body myositis (Mendell et al., 2017): six treated patients, all on an exercise programme, compared with eight untreated clinic patients matched for age, sex and walking distance but not randomised. Walking distance rose by a median of 56.0 metres a year with treatment and fell by 25.8 metres a year without (p = 0.01). Four improved by 58 to 153 metres and two by only 5 to 23 metres; the four who kept to the exercise programme improved most. A critic noted that the patients also took high-dose prednisone, that the registered primary outcome was safety rather than walking distance, and that the clinic controls were not matched for exercise or for the effect of knowingly receiving a new treatment (Greenberg, 2017).
• Duchenne muscular dystrophy (NCT02354781, 2015 to 2017): three of the six planned boys received a higher dose spread over more muscles. According to the results posted on ClinicalTrials.gov, none walked further on the 6-minute walk test, one improved on the North Star motor scale and there were no dose-limiting toxicities. The study was never published as a paper.
• Plasmid gene therapy from the company Minicircle: an open-label phase 1 study in 43 healthy adults on Roatán, Honduras, ran from August 2022 to August 2023 and was registered only afterwards, in May 2024 (NCT06411366). A follow-up combining follistatin with klotho in 14 adults aged 50 to 80, on Roatán and in Austin, Texas, finished in April 2026 (NCT07285629). Neither has posted results on ClinicalTrials.gov or appeared as a peer-reviewed trial report.
• Under way: a phase 1/2a study of an AAV9 follistatin vector, alone or after a VEGF plasmid, in 12 planned adults aged 35 to 75 in Honduras, recruiting since June 2026 (NCT07443826).

Gene therapy, the vial and ACE-083

The human results above say little about vials sold as follistatin-344. Gene therapy turns muscle cells into a local source that keeps releasing FS315 for months. A vial holds protein that has to be injected and is then cleared, and in the Lilly mouse studies native follistatin only worked as an injectable after it had been re-engineered for a much longer half-life. No study has tested an injection of follistatin protein in people.

The closest thing that has been tested is ACE-083, made by Acceleron Pharma: a modified human follistatin fused to the Fc part of an antibody (IgG2), designed to act only in the muscle it is injected into. In two randomised, placebo-controlled phase 2 trials it enlarged the injected muscles. In 55 evaluable people with facioscapulohumeral muscular dystrophy, muscle volume rose 16.4% more than placebo in the biceps and 9.5% more in the lower-leg muscle; in 44 treated people with Charcot-Marie-Tooth disease it rose 13.5% more, and ankle strength improved. But walking, other functional tests and quality of life did not improve consistently, and the sponsor ended both studies early. Larger muscles did not translate into better function.

Frequently asked questions

Is follistatin-344 legal in the Netherlands?

It is not a registered medicine in the Netherlands or anywhere else in the EU, so it may not be sold, advertised or supplied for use in people. Buying and selling it as research material is a different status, which our guide on research-use-only products in the Netherlands explains. For athletes under anti-doping rules it is prohibited at all times, however it was obtained.

Does follistatin-344 build muscle in humans?

For the protein itself nobody knows, because it has never been tested in a human trial. Gene therapy that makes muscle produce follistatin was followed by better walking in 8 of 12 patients with Becker muscular dystrophy or inclusion body myositis, in small trials without a placebo group, but not in 3 boys with Duchenne. ACE-083, a follistatin-based drug, enlarged injected muscles against placebo without improving function. Large gains in muscle have only been shown in mice and monkeys.

What is the difference between follistatin-344 and follistatin gene therapy?

Gene therapy delivers the FS344 gene, usually inside a modified virus (AAV), into muscle, which then keeps producing FS315 for months. A vial sold as follistatin-344 contains protein that is injected and cleared again, and in mice the native protein worked poorly as an injectable until it was re-engineered. All the human results in this article come from gene therapy, so they cannot be carried over to vials.

Is follistatin banned in sport?

Yes, at all times, in and out of competition. The WADA Prohibited List names follistatin in class S4.3 as a myostatin-binding protein, and using a follistatin gene would be gene doping under class M3. The Dopingautoriteit's Dutch version of the 2026 list names follistatine in the same class.

Does follistatin affect fertility or hormones?

It was discovered as a substance that suppresses FSH, which is why researchers worry about the reproductive hormones. In the Becker and inclusion body myositis gene-therapy trials, FSH, LH, testosterone and oestrogen were monitored and did not change, and the treated monkeys showed no abnormal changes in key organs. Those data cover 12 adults and six animals given the FS315 form by gene therapy; for injected protein there are no human data.

Is follistatin-344 safe?

Nobody can say for the protein, because it has never been given to people in a study. In the gene-therapy trials no serious adverse event was attributed to the treatment, but results exist for only 15 patients. With the related drug ACE-083, injection-site reactions were the most common side effect and many participants formed antibodies against it. There are no long-term data.

Storage and handling

Store according to product-label instructions and batch handling notes. Lyophilized research peptides are commonly kept dry, protected from light, and handled with clean laboratory technique.

References

Checked against these sources on 28 October 2026.

Related peptides