Follistatin-344 Research Guide: Protein Identity, Mechanism, and Evidence

Readers searching for “follistatin 344” or “follistatin peptide” often encounter descriptions of a small peptide and a single-purpose myostatin blocker. Neither description is accurate. Follistatin-344 is a large, cysteine-rich glycoprotein precursor whose biological behavior depends on processing, disulfide formation, glycosylation, expression host, isoform, and delivery system. Those variables are central to both evidence review and lot documentation.

Key takeaways

• Follistatin-344 is the canonical 344-amino-acid precursor; residues 1-29 are a signal peptide, leaving a 315-residue mature chain.
• Follistatin is a secreted glycoprotein with 18 experimentally supported disulfide bonds, not a short synthetic peptide.
• It sequesters several TGF-beta-family ligands, including activins, myostatin, and GDF11; it is not a selective myostatin inhibitor.
• Large phenotype findings mostly come from transgenic expression, viral vectors, or engineered proteins, not delivery of a generic FST344 vial.
• Expression host is a material identity variable. If the host, mature-chain boundary, and glycan state are not declared, the record is incomplete.

What Is Follistatin-344? Identity and Protein Class

The human follistatin precursor sequence and genomic organization were reported in 1988, in a paper that described follistatin as an inhibitor of follicle-stimulating hormone release.^[1] Its activin-binding function was established in 1990, when an activin-binding protein purified from rat ovary was shown to be follistatin.^[2] The canonical FST precursor in UniProt P19883 contains 344 amino acids. It is secreted, cysteine-rich, N-glycosylated, and folded into an N-terminal domain followed by three follistatin domains.

Two N-glycosylation sequons are predicted in the canonical record, but a sequence motif is not proof of site occupancy. Some literature describes three putative sites under a different prediction convention. The correct wording is therefore that follistatin is glycosylated and its produced glycoform pattern must be measured; an undeclared expression host cannot be assumed to reproduce human processing.

What Do 344, 315, 317, and 288 Mean?

The most important correction is that 344 describes the precursor, not the mature secreted chain. UniProt annotates residues 1-29 as a signal peptide and residues 30-344 as the 315-residue mature protein, follistatin 315 (FST315, also written FS-315). The FST transcript can also produce a 317-residue precursor whose processing yields follistatin 288 (FST288, FS-288). Literature may use precursor-length and mature-length names side by side, so the naming convention should always be explicit.

FS-315 is generally described as more circulating-oriented, whereas FS-288 shows stronger cell-surface association through heparan-sulfate interactions. Isoform-specific work found differences in activin neutralization and surface behavior.^[3][4] The clinical gene-transfer program selected the FS344 coding sequence in part because its mature FS-315 product has roughly tenfold lower activin affinity than FS-288, a difference intended to limit the pituitary activin-inhibin/FSH consequences of activin antagonism.^[5] That is a construct-selection rationale for vector studies, not an endocrine-effect claim for a supplied reagent. FS303 is another characterized circulating form. These products are related, but they are not interchangeable labels for every experiment.

How Does Follistatin Bind Activin and Myostatin?

Follistatin acts outside the cell by binding selected TGF-beta-family ligands and preventing them from engaging type II receptors. Structural work on the minimal follistatin Fs12 fragment showed two Fs12 molecules wrapping an activin dimer and blocking its type II receptor-binding site.^[6] This is ligand sequestration, not direct stimulation of a muscle-growth receptor.

The three follistatin domains do not contribute identically. Follistatin domain 2 is a major contributor to activin binding, while domain 1 contributes more strongly to myostatin binding. Wild-type follistatin can antagonize activins, myostatin, and GDF11 with different potencies depending on construct and assay.^[7] That breadth is why “selective myostatin inhibitor” is an inaccurate shorthand.

Mechanistic claims should name the exact isoform or construct, ligand, matrix, and endpoint. An engineered follistatin fragment or Fc fusion can deliberately alter ligand preference, proteolysis, distribution, and residence time. Its result does not automatically describe wild-type FS315 produced in an unspecified host.

What Do Preclinical Follistatin Models Show?

Genetic models established that follistatin can regulate muscle phenotype, but they also illustrate how strongly delivery determines interpretation. Transgenic expression increased muscle growth in mice.^[8] In a later experiment, the combination of myostatin knockout and a follistatin transgene produced muscle mass roughly four times that of wild-type mice.^[9] The comparator is crucial: it was not a fourfold increase over the myostatin-null animals.

Transgenic expression of human Follistatin-344 also increased skeletal muscle mass in pigs,^[10] while AAV6 delivery of an FS-288 construct produced hypertrophy findings in mice.^[11] Those systems create sustained expression inside an organism. They do not test a lyophilized recombinant protein with an undeclared expression host.

Protein-delivery studies are not uniformly positive. A mouse study of systemic FS-288 reported increases in muscle mass and reduced fat accumulation.^[12] A reinnervated-muscle rat study using FS-288 reported smaller muscles after three months of denervation, reduced force in the sham-denervated arm at three months, a trend toward increased force after six months of denervation, and satellite-cell counts that rose after denervation but fell after sham denervation; its own conclusion was that follistatin had mixed effects on muscle weight and force.^[13] The exact isoform, delivery method, model, comparison arm, time point, and endpoint belong in the same sentence as the result.

Why Do AAV Findings Not Transfer to a Protein Vial?

Adeno-associated viral delivery installs genetic instructions that can produce tissue expression over time. A recombinant protein preparation is an external material with its own folding, degradation, distribution, and clearance. Those are different interventions even when a sequence name overlaps. Gene-delivery outcomes cannot establish the expected exposure or effect of an FST344 vial.

Engineered proteins add another layer. An FS-288-Fc program reported that systemic exposure was limited by proteolysis even though local delivery could produce a muscle signal.^[14] Glycosylation studies of follistatin-Fc constructs showed that heterogeneous glycans can alter pharmacokinetic behavior.^[15] Recombinant chicken follistatin-315 was produced in soluble, ligand-binding form in E. coli only as a maltose-binding-protein fusion in a strain engineered for cytoplasmic disulfide formation; periplasmic expression using the pMAL-p5x vector yielded no soluble protein.^[16] Bacterial systems also do not perform mammalian N-linked glycosylation, so a bacterially expressed follistatin is not glycoform-equivalent to a mammalian-expressed one.

What Does Human Follistatin Research Show?

Human research remains small and construct-specific. A six-participant open-label Becker muscular dystrophy publication from NCT01519349 evaluated the official intervention rAAV1.CMV.huFollistatin344; the registry records 15 actual enrollees for the completed phase 1 program.^[17] Another study in sporadic inclusion body myositis compared six treated participants with eight matched untreated participants, used gene transfer, and included an exercise cointervention; that publication carries no registry accession.^[18] Both trials delivered the FS344 coding sequence by AAV vector. The isoform name matches the vial label, but nothing else about the intervention does.

ACE-083 was a locally acting engineered ligand trap, not wild-type FST344. The completed phase 1 study NCT02257489 enrolled 58 participants and reported increased muscle volume in healthy volunteers without demonstrating increased strength.^[19] Later randomized phase 2 studies in facioscapulohumeral muscular dystrophy (NCT02927080) and Charcot-Marie-Tooth disease (NCT03124459) found muscle-volume changes without consistent functional benefit, although the CMT study reported an ankle-dorsiflexion strength signal.^[20][21] Those studies and the open-label extension NCT03943290 were terminated, with the registry stating that functional secondary endpoints were not achieved.

No follistatin-based product appears in the FDA’s approved-drug records. NCT02354781 registered a three-participant Duchenne muscular dystrophy study, but no results publication was identified for that record. A registration establishes that a study was listed; it does not establish an outcome or describe delivery of recombinant Follistatin-344 protein.

Both published follistatin gene-transfer reports come from the same investigator group at Nationwide Children’s Hospital and The Ohio State University, and every ACE-083 and FST288-Fc result discussed here comes from one sponsor, Acceleron Pharma. These are not independent replications of one another.

The 2020 Dağ case series requires particularly careful language. Investigators retrospectively described 11 male bodybuilding athletes who presented to one clinic with reduced visual acuity and optical-coherence-tomography findings of central serous chorioretinopathy, each reporting subcutaneous injection of complete 1 mg “follistatin-344” vials.^[22] The material was not independently verified, there was no control group, and retrospective association cannot establish causation. It is appropriately treated as a human safety signal, not as instructions, a frequency estimate, proof of a dose-response relationship, or confirmation that the material was FST344.

What Analytical Documentation Can Verify

A meaningful Follistatin-344 analytical record must define the construct before it reports a percentage. Does “344” mean an encoded precursor, a recombinant construct retaining a signal segment, or the mature FS315 chain? Which host expressed it? Was the signal peptide processed? What glycan state, disulfide pattern, and oligomeric state were expected? Without those fields, a chromatographic purity number cannot resolve identity.

Peptide mapping or equivalent orthogonal testing can support sequence coverage and mature-chain boundaries. Intact mass should state whether the target is glycosylated, deglycosylated, precursor, or mature material. Chromatography describes the detected species under one method; it does not prove sequence, correct disulfide pairing, glycan occupancy, low host-cell impurities, endotoxin status, sterility, or ligand-neutralizing activity. A declared functional assay can support activity only under that assay’s conditions.

The HPLC purity guide, mass-spectrometry guide, and Certificate of Analysis guide explain those method boundaries. The lab-verification library provides the broader traceability framework. This article does not cache a current lot result, price, availability, or undeclared host.

Where Does the Follistatin-344 Evidence Stop?

The evidence is strongest for biological principles: follistatin binds multiple ligands; isoforms differ; genetic overexpression can alter muscle phenotypes; and delivery engineering changes exposure. The evidence becomes weaker when a claim crosses from a transgene to a vector, from a vector to a protein, from FS-288 to FS-315, or from an engineered ligand trap to wild-type follistatin. No controlled human trial identified in this review tested an independently verified recombinant FST344 vial.

One nonhuman-primate gene-delivery paper carrying an erratum whose content could not be read, and one engineered follistatin-variant pharmacokinetics paper carrying an unread addendum, were both excluded rather than cited without correction context. The identity, mechanism, model, delivery-gap, and human-evidence conclusions above do not depend on either record.

The conclusions here stay narrower than the claims commonly attached to this compound: no general “muscle-building” promise, no assumption that all follistatin constructs are equivalent, and no inference that a research material has pharmaceutical status.

References

1. Shimasaki S, et al. Primary structure of the human follistatin precursor and its genomic organization. Proc Natl Acad Sci U S A. 1988;85:4218-22. PMID 3380788.
2. Nakamura T, et al. Activin-binding protein from rat ovary is follistatin. Science. 1990;247:836-8. PMID 2106159.
3. Hashimoto O, et al. Difference between follistatin isoforms in the inhibition of activin signalling: activin neutralizing activity of follistatin isoforms is dependent on their affinity for activin. Cell Signal. 2000;12:565-71. PMID 11027950.
4. Sidis Y, et al. Biological activity of follistatin isoforms and follistatin-like-3 is dependent on differential cell surface binding and specificity for activin, myostatin, and bone morphogenetic proteins. Endocrinology. 2006;147:3586-97. PMID 16627583.
5. Al-Zaidy SA, et al. Follistatin Gene Therapy Improves Ambulation in Becker Muscular Dystrophy. J Neuromuscul Dis. 2015;2:185-192. PMID 27858738.
6. Harrington AE, et al. Structural basis for the inhibition of activin signalling by follistatin. EMBO J. 2006;25:1035-45. PMID 16482217.
7. Schneyer AL, et al. Differential antagonism of activin, myostatin and growth and differentiation factor 11 by wild-type and mutant follistatin. Endocrinology. 2008;149:4589-95. PMID 18535106.
8. Lee SJ, et al. Regulation of myostatin activity and muscle growth. Proc Natl Acad Sci U S A. 2001;98:9306-11. PMID 11459935.
9. Lee SJ. Quadrupling muscle mass in mice by targeting TGF-beta signaling pathways. PLoS One. 2007;2:e789. PMID 17726519.
10. Chang F, et al. The transgenic expression of human follistatin-344 increases skeletal muscle mass in pigs. Transgenic Res. 2017;26:25-36. PMID 27787698.
11. Winbanks CE, et al. Follistatin-mediated skeletal muscle hypertrophy is regulated by Smad3 and mTOR independently of myostatin. J Cell Biol. 2012;197:997-1008. PMID 22711699.
12. Gangopadhyay SS, et al. Systemic administration of follistatin288 increases muscle mass and reduces fat accumulation in mice. Sci Rep. 2013;3:2441. PMID 23942549.
13. Feger MA, et al. Follistatin Protein Enhances Satellite Cell Counts in Reinnervated Muscle. J Brachial Plex Peripher Nerve Inj. 2022;17:e12-e21. PMID 35747585.
14. Castonguay R, et al. Follistatin-288-Fc Fusion Protein Promotes Localized Growth of Skeletal Muscle. J Pharmacol Exp Ther. 2019;368:435-445. PMID 30563942.
15. Datta-Mannan A, et al. Insights into the Impact of Heterogeneous Glycosylation on the Pharmacokinetic Behavior of Follistatin-Fc-Based Biotherapeutics. Drug Metab Dispos. 2015;43:1882-90. PMID 26354950.
16. Lee SB, et al. Production of bioactive chicken follistatin315 in Escherichia coli. Appl Microbiol Biotechnol. 2014;98:10041-51. PMID 25411099.
17. Mendell JR, et al. A phase 1/2a follistatin gene therapy trial for becker muscular dystrophy. Mol Ther. 2015;23:192-201. PMID 25322757.
18. Mendell JR, et al. Follistatin Gene Therapy for Sporadic Inclusion Body Myositis Improves Functional Outcomes. Mol Ther. 2017;25:870-879. PMID 28279643.
19. Glasser CE, et al. Locally acting ACE-083 increases muscle volume in healthy volunteers. Muscle Nerve. 2018;57:921-926. PMID 29486514.
20. Statland JM, et al. Randomized phase 2 study of ACE-083, a muscle-promoting agent, in facioscapulohumeral muscular dystrophy. Muscle Nerve. 2022;66:50-62. PMID 35428982.
21. Thomas FP, et al. Randomized Phase 2 Study of ACE-083 in Patients With Charcot-Marie-Tooth Disease. Neurology. 2022;98:e2356-e2367. PMID 35545446.
22. Dağ U, et al. Central serous chorioretinopathy associated with high-dose follistatin-344: a retrospective case series. Int Ophthalmol. 2020;40:3155-3161. PMID 32671599.

Frequently Asked Questions

Continue Your Research

Use related records to separate protein identity from the wider growth-axis literature and from analytical-method questions. The complete Apex Research Library provides the wider index.

Research Use Disclaimer