The myostatin brake
Skeletal muscle mass is held in check by a signalling brake. In 1997, McPherron, Lawler and Lee identified a new member of the TGF-β superfamily — myostatin (GDF-8) — and reported that mice lacking the gene carried roughly twice the muscle mass of wild-type animals (McPherron et al., 1997). Myostatin signals through the activin type IIB receptor (ActRIIB) and the Smad2/3 cascade, restraining myoblast proliferation and differentiation.
The brake is not merely permissive. Systemic administration of myostatin in mice induced a cachectic phenotype, evidence that the ligand is sufficient to drive muscle loss when present in excess (Zimmers et al., 2002).
Two research strategies have been used to remove that brake. The first is an engineered decoy receptor; the second is the body's own antagonist protein.
Two ways to neutralise myostatin
ACE-031 (ActRIIB-Fc). A recombinant fusion protein: the extracellular ligand-binding domain of human ActRIIB joined to a human IgG1 Fc domain. It circulates as a soluble ligand trap, binding myostatin, activin A, activin B, BMP9 and BMP10, and preventing them from engaging cell-surface receptors.
Follistatin-344. A 344-amino-acid secreted glycoprotein that binds myostatin and activin A in a near-irreversible stoichiometric complex. Follistatin's ligand coverage is broader than myostatin alone — it also neutralises activin A and several bone morphogenetic proteins — and comparative animal work indicates that blocking myostatin plus activin produces greater muscle hypertrophy than myostatin inhibition alone.
That broader coverage is also where the two compounds diverged in the clinic.
ACE-031 in humans: lean mass up, then discontinued
A single ascending-dose Phase 1 study in healthy postmenopausal women showed an increase in lean body mass with a corresponding reduction in fat mass, sustained for four to eight weeks, with no serious adverse events at the doses tested (Rooks et al., 2017).
A randomised, double-blind, placebo-controlled Phase 2 study followed in boys with Duchenne muscular dystrophy. The pooled dose group gained lean body mass relative to placebo at 24 weeks, but the trial was terminated early (Campbell et al., 2012).
The discontinuation, in July 2011, followed reports of epistaxis — some severe — telangiectasia and menstrual irregularities. The vascular effects are attributed to the compound's binding of BMP9 and BMP10, both of which are involved in maintaining vascular endothelial integrity. The lesson recorded in the literature is that a ligand trap affects every ligand in its binding profile, not only the intended one.
Follistatin-344: deep animal data, no completed human protein trial
Follistatin's evidence base is almost entirely preclinical, but it is a substantial one. AAV-delivered follistatin improved muscle mass and function in non-human primates (Haidet et al., 2008). In the dystrophin-deficient mdx mouse, follistatin gene delivery increased muscle mass, reduced fibrosis and prolonged survival (Kota et al., 2009). Transgenic expression of human follistatin-344 in pigs significantly increased skeletal muscle mass (Yang et al., 2016).
The human record is thinner. A Phase 1/2a trial of AAV-delivered follistatin in Becker muscular dystrophy was initiated (NCT01519349), and published results are limited (Mendell et al., 2015). No completed human clinical trial of the recombinant follistatin-344 protein has been published. On this site's four-tier scale that places follistatin-344 at Limited and ACE-031 at Moderate — the latter modestly stronger because human data exist for it, even though its development was stopped.
The UK position
Neither compound is a licensed medicine in the United Kingdom. Neither is a controlled drug. Both sit in the grey area that applies to a substance that is neither authorised nor scheduled: legal to possess, but not authorised to be supplied as a medicine.
Under the Medicines Act 1968 and the Human Medicines Regulations 2012, supplying a substance alongside a medicinal claim without a marketing authorisation is an offence, and the MHRA can act against sellers who market unlicensed compounds with therapeutic claims. Both compounds are also prohibited by the World Anti-Doping Agency, and the sports-medicine literature continues to classify myostatin inhibitors as unapproved for athletic use (Becker et al., 2025).
For researchers, the relevant point is not that these compounds are illegal — they are not controlled substances — but that the safety signals attached to ACE-031 are unusually well documented for a compound in this class. Any non-clinical use carries the vascular risk profile the trial reported. This article is a research summary, not consumption guidance, and nothing in it should be read as advice to use either compound.
What the class shows
Myostatin inhibition was one of the first tests of whether a peptide or protein approach could shift body composition. ACE-031 showed that it can, and showed the cost: a ligand trap is only as selective as its ligand-binding profile allows. Later programmes have pursued narrower selectivity. The route from receptor biology to a licensed medicine in this class remains open, and, on the present evidence, unfinished.
This article is AI-researched and editorially reviewed. It is provided for research and educational purposes only and is not medical advice. Research peptides are not licensed for human consumption in the UK.