Follistatin 344 Research, Specifications & Scientific Information

Introduction

Follistatin 344 is named after a number that does not describe the protein anyone ends up with, and the confusion that causes is worth clearing up immediately.

The human FST gene is transcribed into two alternatively spliced messenger RNAs, encoding precursors of 344 and 317 amino acids. Both were sequenced in 1988 from a human testis library ^[1]. The first 29 residues of each are a signal peptide, removed as the protein is secreted. A cell that expresses the 344 construct therefore releases a 315-residue protein; a cell that expresses the 317 construct releases a 288-residue one. The number 344 identifies a splice form and a coding sequence. It does not identify a species that circulates in blood.

That is not a pedantic point. It is the reason the human trials in this literature are gene therapy trials: investigators chose the FS344 complementary DNA, packaged it in an adeno-associated virus and injected the vector, so that tissue would make the protein locally ^[7]. No clinical trial has administered follistatin 344 as a protein.

This page is a reference record. It describes research, it keeps the gene-transfer literature and the protein literature apart, and it contains no guidance of any kind on handling the material.

What Is Follistatin 344?

Follistatin is a secreted glycoprotein, discovered as a gonadal factor that suppresses follicle-stimulating hormone release, and later understood as a general-purpose trap for a set of transforming growth factor beta superfamily ligands: activin, myostatin, growth and differentiation factor 11, and some bone morphogenetic proteins.

Structurally it is modular. The mature chain is built from contiguous domains encoded by precisely separated exons, three of them closely similar to each other and to human epidermal growth factor and human pancreatic secretory trypsin inhibitor — an arrangement the 1988 authors read as evidence of exon shuffling during evolution ^[1].

The isoform naming runs as follows, and all four numbers refer to the same gene:

• FST344 — the longer precursor, 344 residues including the signal peptide.
• FST-315 — the mature protein secreted from that precursor. The FDA/NCATS register carries follistatin under UNII 506IY26H2I as this 315-residue chain.
• FST317 — the shorter precursor, produced by alternative splicing.
• FST-288 — the mature protein secreted from that one, which lacks the acidic C-terminal extension present in the 315 form.

It is not an approved medicine anywhere, as a protein or as a gene therapy.

Follistatin 344 Specifications

The 344-residue chain above is the sequence of the human follistatin precursor recorded in UniProtKB under accession P19883, isoform 1. Residues 1 to 29 are a signal peptide; residues 30 to 344 are the mature secreted chain, which is 315 residues long and is the substance covered by the separate entry for follistatin 315. That relationship is the single most important fact about this compound's naming: 344 counts the precursor including the signal peptide, so a cell that expresses and secretes the 344 construct releases a 315-residue protein, not a 344-residue one. The number 344 therefore identifies a coding sequence and a splice form rather than a species that circulates. The FDA/NCATS Global Substance Registration System carries follistatin under UNII 506IY26H2I as the 315-residue mature chain with a calculated average mass of 31600 g/mol; it carries no separate record for the precursor, no CAS registry number and no molecular formula, and no PubChem compound identifier resolves for the name, so those three fields are shown as unknown. The computed mass of 38007 Da above is UniProt's value for the unmodified 344-residue chain and excludes glycosylation; the protein carries N-glycosylation in vivo. The certificate of analysis supplied with a laboratory order is the record of what a given lot contains, including whether it is the precursor or the mature chain.

Values that a public register does not carry are shown as not publicly characterised rather than estimated. Identifiers are reference values; the certificate of analysis supplied with a laboratory order is the record for a given lot.

How Does Follistatin 344 Work?

Not by activating a receptor. Follistatin has no receptor of its own in the conventional sense; it works by binding signalling ligands and removing them from circulation before they can reach theirs.

The consequence in skeletal tissue is myostatin antagonism. Myostatin, a transforming growth factor beta family member, is a negative regulator of growth in skeletal musculature, and animals lacking it — engineered mice, and naturally occurring cattle, sheep, dogs and one described human — carry markedly more of it ^[3]. Sequestering myostatin lifts that restraint.

But follistatin is not a myostatin-specific reagent, and the literature has spent considerable effort establishing how non-specific it is. Its affinity profile spans activin, myostatin and growth and differentiation factor 11, with bone morphogenetic proteins 6 and 7 bound more weakly and 2 and 4 not at all ^{[2, 5]}. Activin antagonism is undesirable in a compound aimed at the myostatin pathway, because activin does many other things, and attempts to engineer that selectivity are described in the mechanism section below.

The isoforms differ from one another in a way that has nothing to do with ligand affinity, which is the second thing worth understanding about this protein.

Follistatin 344 Mechanism of Action

In vitro research

Why the isoforms behave differently

Recombinant follistatin isoforms and the related protein follistatin-like 3 were produced and compared directly. Activin-binding affinities and kinetics were comparable across all of them. Cell-surface binding was not: it ran FST288 > FST303 > FST315 > follistatin-like 3, a wide spread ^[2].

That spread, rather than ligand affinity, predicted biological effect. Inhibition of endogenous activin activity correlated closely with surface binding. In one bioassay the 288 isoform suppressed activin-dependent proliferation while the 315 isoform enhanced it — opposite signs, from proteins with equivalent activin-binding affinity. And when follistatin-like 3, which does not associate with cell membranes, was expressed as a membrane-anchored protein, its inhibitory activity increased dramatically ^[2].

The mechanism is therefore partly about where the protein sits. A trap bound to the cell surface intercepts ligands before they reach receptors on that surface; a trap circulating freely does not, and can instead act as a reservoir. Since the 344 precursor yields the 315 isoform, it yields the weakly surface-binding member of the pair.

Which domains bind which ligand

A panel of point and domain-swapping mutants established that activin binding and neutralisation are mediated primarily by the second follistatin domain, while myostatin binding depends more on the first. Deleting the second domain, or replacing it with an extra copy of the first, produced myostatin antagonists with greatly reduced activin antagonism — the selectivity that had been sought. Those same mutants, however, still bound growth and differentiation factor 11 ^[5].

What the structure shows

The crystal structure of myostatin in complex with the 288-residue isoform resolved two things. The N-terminal domain of the isoform undergoes a conformational rearrangement to bind myostatin, and probably supplies the antagonist's specificity. And the complex creates a continuous electropositive surface that substantially increases affinity for heparin, which translates into stronger interaction with the cell surface and enhanced myostatin degradation in the presence of either the 288 or the 315 isoform ^[6].

What Is Follistatin 344 Being Researched For?

Three lines, of which only one has reached humans.

• Gene transfer for muscular dystrophies and inflammatory myopathy. The FST344 coding sequence delivered by adeno-associated virus, in three completed early-phase trials ^{[7, 9, 10, 11]}.
• Preclinical myostatin-pathway biology. Transgenic and vector-delivered overexpression in mice and pigs ^{[3, 4, 8]}.
• Structural and biochemical characterisation. Isoform comparison, domain mapping and crystallography, all in vitro ^{[2, 5, 6]}.

None of that is research into follistatin 344 administered as a protein, and none of it is research into research-grade material supplied for laboratory use.

Human Research on Follistatin 344

Human clinical research

Results from pharmaceutical clinical trials describe the investigational material and populations used in those studies and should not be interpreted as establishing the effects of research-grade materials offered for laboratory use.

What was administered in all three trials. rAAV1.CMV.huFollistatin344 — an adeno-associated virus serotype 1 vector carrying the human FST344 complementary DNA under a cytomegalovirus promoter — delivered by direct bilateral intramuscular injection into the quadriceps. Not a protein, and not repeated administration: a single vector exposure intended to produce sustained local expression.

Phase 1/2a in Becker muscular dystrophy

Population. Six participants with Becker muscular dystrophy, a dystrophin-deficiency variant arising from DMD gene mutations, for which no treatment exists ^{[7, 10]}.

Design. Two cohorts. Three participants received 3 × 10¹¹ vector genomes per kilogram per leg; three received 6 × 10¹¹. The primary outcome was distance walked on the six-minute walk test ^[7].

Result. In the first cohort, two participants improved by 58 and 125 metres and one showed no change. In the second, two improved by 108 and 29 metres and one showed no improvement. Biopsy showed reduced endomysial fibrosis, reduced central nucleation and a more normal fibre size distribution with hypertrophy, most marked at the higher exposure. No adverse effects were encountered ^[7].

Limitations. Six participants, open-label, no control group, no randomisation, and a functional primary outcome that varies substantially within individuals. Two of six showed no change at all. The authors described it as a proof-of-principle trial, which is the right description.

Sporadic inclusion body myositis

Population. Six participants with sporadic inclusion body myositis, an inflammatory myopathy whose distribution of weakness resembles that of Becker muscular dystrophy, compared against eight untreated participants matched for age, sex and baseline measures ^[9].

Design. rAAV1.CMV.huFS344 at 6 × 10¹¹ vector genomes per kilogram to the quadriceps of both legs, with an exercise regimen included in the protocol for each participant. Primary outcome the six-minute walk distance ^[9].

Result. Annualised to a median one-year change, treated participants improved by 56.0 metres per year against a decline of 25.8 metres per year in the untreated comparison group (p = 0.01). Four of the six treated participants improved by 58 to 153 metres; two improved minimally, by 5 to 23 metres. Treatment effects included decreased fibrosis and improved regeneration. The authors noted that more advanced disease, with discernible tissue loss, poses challenges ^[9].

Limitations. Six treated participants against a non-randomised matched comparison group, with an exercise programme embedded in the intervention arm, which makes the contribution of the vector itself difficult to isolate.

Duchenne muscular dystrophy

A further trial of the same construct enrolled three participants with Duchenne muscular dystrophy as a phase 1/2 study and has completed ^[11].

What none of this shows. These are small, early, mostly uncontrolled gene-transfer studies. They do not establish efficacy, they do not concern an administered protein, and they say nothing whatever about research-grade material supplied for laboratory use.

Preclinical Research on Follistatin 344

Animal research

The animal work is where the myostatin-pathway argument was built, and it is worth noting that almost all of it uses genetic overexpression rather than protein administration.

The ceiling experiment came from a follistatin transgene crossed onto a myostatin-null background. Myostatin-null mice already carry about twice the mass of skeletal musculature of wild-type animals. Adding the follistatin transgene to those animals produced roughly four times wild-type — which demonstrates that follistatin is inhibiting something in addition to myostatin, since there was no myostatin left to inhibit ^[3]. That single result is the cleanest available evidence that this protein's effects are not a pure myostatin readout.

The durability experiment used vector delivery rather than a transgene: a one-time gene administration of myostatin-inhibitor proteins enhanced mass and strength in normal and dystrophic mouse models for more than two years, and did so even when delivered to aged animals ^[4]. That study is the direct preclinical antecedent of the human trials described above, from the same institution.

Livestock work extended the finding to a large species. Transgenic Duroc pigs expressing human follistatin 344 specifically in muscle tissue showed an increased proportion of skeletal muscle and a reduced proportion of body fat, with lean meat percentage 72.95 ± 1.0% against 69.18 ± 0.97% in wild-type animals. Myofibre hypertrophy was observed in the longissimus dorsi, Smad2 phosphorylation was reduced and Akt Ser473 phosphorylation increased, and no cardiac hypertrophy or reproductive abnormality was reported ^[8].

Findings described in this section were observed in animals, and nothing in them establishes anything about humans. Note also that in every study above the protein was produced endogenously by the animal's own transduced or transgenic cells.

Current Research Status

Status as of September 22, 2026. This block is rendered from maintained fields, not from prose, so it cannot go stale in one place and stay current in another.

Follistatin 344 vs Related Compounds

The comparisons below are structural and evidential. No comparison of effect is made on this page.

Follistatin 315 :: The mature protein produced from this precursor, 315 residues after the 29-residue signal peptide is cleaved. Not a different molecule so much as a later stage of the same one, and the species that is actually secreted.

FST-288

The mature protein from the alternative 317 precursor. It binds cell-surface proteoglycans far more strongly than the 315 form, and that difference — not ligand affinity — is what separates the isoforms biologically ^[2].

ACE-031 :: The other myostatin-pathway compound in this library, and a completely different construction: a soluble fusion of the activin receptor type IIB extracellular domain with an antibody Fc region, which traps ligands at the receptor rather than at the ligand-binding protein. It reached phase 2 and was halted; that record belongs to its own entry.

Evidence base

Follistatin 344's human record is three small early-phase gene therapy trials of a viral vector. ACE-031's is a halted phase 2 programme of an administered protein. Neither is a record of a protein given as a research reagent, and the two are not interchangeable.

Chemical & Molecular Characteristics

The sequence is a UniProt record. UniProtKB accession P19883, isoform 1, carries the 344-residue precursor with a computed mass of 38007 Da. Residues 1 to 29 are annotated as a signal peptide and residues 30 to 344 as the mature chain. The internal architecture is annotated as alternating follistatin-like and Kazal-like domains across the length of the mature protein.

The register carries the mature chain, not the precursor. The FDA/NCATS Global Substance Registration System record for follistatin, UNII 506IY26H2I, is the 315-residue chain, with a calculated average mass of 31600 g/mol and a cross-reference to the same UniProt accession. There is no separate register record for the 344-residue precursor. This page therefore publishes the precursor sequence from UniProt and the mature-chain identifier from the register, and says which is which rather than merging them.

Three fields are blank on purpose. No CAS registry number, no molecular formula and no PubChem compound identifier are published for this substance by any register consulted. All three are shown as unknown rather than estimated.

The computed mass excludes glycosylation. Follistatin is N-glycosylated, and the UniProt figure is calculated from the unmodified amino acid chain. A mass measured on real material will not match it exactly, and the amount of the difference depends on the expression system.

What a supplied material actually contains is an open question. A protein expressed and secreted from a mammalian system under this construct is the 315-residue mature chain. A protein expressed intracellularly in a bacterial system may retain the signal sequence. Those are different substances with different masses sold under the same name, which is why the certificate of analysis for a given lot is the record for that lot.

Frequently Asked Questions

Scientific References

1. Shimasaki S, Koga M, Esch F, et al.. Primary structure of the human follistatin precursor and its genomic organization Proceedings of the National Academy of Sciences of the United States of America; 1988. PMID 3380788 doi:10.1073/pnas.85.12.4218
2. Sidis Y, Mukherjee A, Keutmann H, 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. PMID 16627583 doi:10.1210/en.2006-0089
3. Lee SJ. Quadrupling muscle mass in mice by targeting TGF-beta signaling pathways PloS one; 2007. PMID 17726519 doi:10.1371/journal.pone.0000789
4. Haidet AM, Rizo L, Handy C, et al.. Long-term enhancement of skeletal muscle mass and strength by single gene administration of myostatin inhibitors Proceedings of the National Academy of Sciences of the United States of America; 2008. PMID 18334646 doi:10.1073/pnas.0709144105
5. Schneyer AL, Sidis Y, Gulati A, et al.. Differential antagonism of activin, myostatin and growth and differentiation factor 11 by wild-type and mutant follistatin Endocrinology; 2008. PMID 18535106 doi:10.1210/en.2008-0259
6. Cash JN, Rejon CA, McPherron AC, et al.. The structure of myostatin:follistatin 288: insights into receptor utilization and heparin binding The EMBO journal; 2009. PMID 19644449 doi:10.1038/emboj.2009.205
7. Mendell JR, Sahenk Z, Malik V, et al.. A phase 1/2a follistatin gene therapy trial for becker muscular dystrophy Molecular therapy : the journal of the American Society of Gene Therapy; 2015. PMID 25322757 doi:10.1038/mt.2014.200
8. Chang F, Fang R, Wang M, et al.. The transgenic expression of human follistatin-344 increases skeletal muscle mass in pigs Transgenic research; 2017. PMID 27787698 doi:10.1007/s11248-016-9985-x
9. Mendell JR, Sahenk Z, Al-Zaidy S, et al.. Follistatin Gene Therapy for Sporadic Inclusion Body Myositis Improves Functional Outcomes Molecular therapy : the journal of the American Society of Gene Therapy; 2017. PMID 28279643 doi:10.1016/j.ymthe.2017.02.015
10. Follistatin Gene Transfer to Patients With Becker Muscular Dystrophy and Sporadic Inclusion Body Myositis 2012. NCT01519349
11. Clinical Intramuscular Gene Transfer of rAAV1.CMV.huFollistatin344 Trial to Patients With Duchenne Muscular Dystrophy 2015. NCT02354781

Every identifier above is resolved against PubMed, Crossref or ClinicalTrials.gov at build time, and the title returned by the register is compared with the title stored here. A page does not publish if a reference fails to resolve.

Research-Use Information

For in vitro research use only. This material is a laboratory reagent. It is not a drug, food, dietary supplement, or cosmetic and is not for human or veterinary use, including ingestion, injection, or any other administration. No information on this page describes or implies any effect in humans or animals. Sold only to researchers under our Terms of Sale.