Myostatin Propeptide

The recombinant N-terminal prodomain of GDF-8 that binds mature myostatin and holds it in a latent, inactive complex; it enlarges muscle in mice but has never been given to humans in a registered trial.

Also referenced as: GDF-8 propeptide, GDF8 propeptide, Myostatin prodomain, MSTN propeptide, Myostatin N-terminal propeptide

Also appears in: Longevity

What is Myostatin Propeptide?

Myostatin propeptide is not a peptide in the sense that most research-market compounds are. It is a recombinant protein domain — the N-terminal prodomain of growth/differentiation factor 8 (GDF-8, better known as myostatin), spanning residues 24–266 of the 375-residue human myostatin precursor (UniProt O14793). That domain is 243 amino acids and roughly 27.7 kDa unglycosylated; bacterial constructs typically carry an added N-terminal methionine and are catalogued at 244 residues and about 27.8 kDa. Either way it is roughly an order of magnitude heavier than the short synthetic chains it is compared against in the research market, and it behaves like a protein biologic rather than a peptide.

Its biology is straightforward: myostatin is made as a precursor, a furin-type protease separates the prodomain from the C-terminal growth factor, and the two stay noncovalently associated in a latent complex that cannot signal. The propeptide is therefore the body’s own myostatin off-switch, and supplying more of it is one of the oldest proposed ways to release the brake myostatin puts on skeletal muscle.

It reaches buyers as a life-science reagent rather than as a research-peptide vial — protein suppliers list it in microgram fills for laboratory use, marketed as the “natural” myostatin inhibitor. That framing puts it in the same conversation as follistatin 344, which also restrains myostatin but does so as a different protein with a broader target range that includes the activins. It should not be confused with myostatin itself, which is the growth factor this domain suppresses; the research-market vials labeled simply “Myostatin” are a separate listing, not this molecule.

How it works

• Direct binding and neutralization of the mature growth factor. Purified recombinant GDF-8 propeptide forms a noncovalent complex with purified GDF-8, suppresses GDF-8 activity in a (CAGA)12 reporter assay in A204 rhabdomyosarcoma cells, and blocks GDF-8 binding to L6 myoblasts (Thies et al., Growth Factors, 2001;18(4):251–259). Two propeptide monomers inhibit one myostatin homodimer, N-linked glycosylation is not required for that inhibition, and the inhibitory activity maps to residues 42–115 while the 99–266 region mainly confers stability (Jiang et al., Biochemical and Biophysical Research Communications, 2004;315(3):525–531).
• Competition at the activin type II receptor. The mature myostatin dimer binds ActRIIB, and less strongly ActRIIA; propeptide inhibited that binding, though only at higher concentrations than follistatin required. Transgenic mice expressing the propeptide from a skeletal-muscle promoter showed muscle mass increases comparable to myostatin-knockout mice (Lee & McPherron, PNAS, 2001;98(16):9306–9311).
• It is the dominant endogenous myostatin carrier. Immunoprecipitation of the native complex from normal mouse and human serum showed that more than 70% of circulating myostatin is bound to its own propeptide, with follistatin-related gene (FLRG) as the other major binder (Hill et al., Journal of Biological Chemistry, 2002;277(43):40735–40741). The mechanism being sold is a real physiological one, not an invented target.
• Proteolysis sets how long the inhibition lasts. Members of the BMP-1/tolloid metalloproteinase family cleave the propeptide inside the latent complex and thereby release active myostatin; a cleavage-resistant mutant propeptide, by contrast, produced significant increases in muscle mass when injected into adult mice (Wolfman et al., PNAS, 2003;100(26):15842–15846). Unmodified propeptide is itself a substrate for the enzyme that would destroy its effect.

Research status

No human trials. A ClinicalTrials.gov search returns no registered study using myostatin propeptide as an intervention. There is no published human pharmacokinetic, safety, or efficacy data for the protein in any form, and no regulatory authority has approved it for anything.

Animal work is real, but almost all of it is genetic or engineered. Transgenic overexpression of the pro domain produced a dramatic muscling phenotype in mice — a 17–30% increase in body weight and a 22–44% increase in carcass weight by 9 weeks, driven by myofiber hypertrophy rather than hyperplasia (Yang et al., Molecular Reproduction and Development, 2001;60(3):351–361) — matching the transgenic result in Lee & McPherron (2001) above. Only a small number of studies have injected propeptide protein into animals at all, and each used a stabilized form rather than the plain domain: the BMP-1/tolloid-resistant D76A mutant (Wolfman et al., 2003), an IgG-Fc fusion in dystrophic mice (Bogdanovich et al., 2005), and an injectable GDF-8 propeptide-Fc dosed weekly at 20 mg/kg for four weeks in 24-month-old mice, which raised tibialis anterior mass by about 7% and extensor digitorum longus fiber diameter by about 16% while leaving femoral bone volume, stiffness, and ultimate force unchanged (Arounleut et al., Experimental Gerontology, 2013;48(9):898–904).

Dystrophy models. A myostatin propeptide stabilized by fusion to IgG-Fc improved the pathophysiology of mdx mice, the standard model of Duchenne muscular dystrophy, with a specific-force improvement the authors reported as exceeding what antibody-mediated myostatin blockade had achieved (Bogdanovich et al., FASEB J, 2005;19(6):543–549). AAV8 delivery of the propeptide gene raised skeletal muscle mass in normal mice by two routes — intraperitoneal AAV-MPRO in neonates and intravenous AAV-MPRO76AFc, an Fc-fused version carrying the same D76A substitution that resists BMP-1/tolloid cleavage, in adults — through myofiber hypertrophy rather than hyperplasia. In mdx mice, where only the MPRO76AFc construct was tested, muscle mass also rose, with larger and more uniform myofibers, less fibrosis, lower serum creatine kinase, improved grip and tetanic contractile force, and no cardiac hypertrophy. The same study found reduced treadmill endurance in treated mdx mice compared with untreated controls (Qiao et al., Human Gene Therapy, 2008;19(3):241–254) — a negative finding that rarely travels with the muscle-mass claim.

Healthy and aged mice. A single intravenous dose of AAV8 expressing a mutated propeptide raised the mass of several muscles in healthy adult mice at 8 and 17 weeks, driven by type IIB fiber hypertrophy without hyperplasia; notably, the hypertrophied EDL did not show the specific-force deficit seen in germline myostatin-null mice (Matsakas et al., Neuromuscular Disorders, 2009;19(7):489–499). In aged mice, AAV8 propeptide increased body and muscle weight within 7 weeks, downregulated ubiquitin-mediated proteolysis genes, shifted fibers toward a glycolytic phenotype, and preserved normal contractile properties (Collins-Hooper et al., Journals of Gerontology Series A, 2014;69(9):1049–1059).

Structural context. The crystal structure of unprocessed pro-myostatin shows an open, domain-swapped V shape, and the pro-mature complex remaining after furin cleavage is both substantially less active than mature myostatin and resistant to follistatin — latency is conferred by several distinct features acting together (Cotton et al., EMBO J, 2018;37(3):367–383).

What the pathway has done in humans, using other molecules. The propeptide has no clinical record, but the target does, and it is mostly one of disappointment. The anti-myostatin antibody MYO-029 (stamulumab) in 116 adults with muscular dystrophy was tolerable apart from cutaneous hypersensitivity at 10 and 30 mg/kg, and showed no improvement in exploratory strength or function endpoints — though the trial was a safety study and was not powered to detect efficacy, and it did report a trend toward increased muscle size in a limited number of subjects (Wagner et al., Annals of Neurology, 2008;63(5):561–571). Domagrozumab, tested in a randomized Phase 2 of 120 ambulatory boys with Duchenne muscular dystrophy, met its safety objective but missed its primary functional endpoint outright — a between-group difference in four-stair-climb time at week 49 of 0.27 seconds (95% CI −7.4 to 7.9; p = 0.94) — with no significant separation from placebo on any secondary clinical endpoint and only non-significant gains in muscle volume (Wagner et al., Neuromuscular Disorders, 2020;30(6):492–502). Apitegromab, an antibody that binds the pro- and latent forms of myostatin — that is, myostatin still held by this propeptide — is the one myostatin-directed agent to meet a pivotal endpoint, in spinal muscular atrophy, where it was added on top of SMN-targeted therapy in the Phase 3 SAPPHIRE trial. Per sponsor announcements it then received a Complete Response Letter in September 2025 attributed solely to observations at a third-party fill-finish facility rather than to its efficacy or safety data; the application was resubmitted in March 2026 and accepted with a September 30, 2026 action date, leaving it unapproved in the United States as of August 2026. None of these results are evidence about the propeptide itself.

Common dosage forms

• Lyophilized recombinant protein in biochemical-reagent vials, most often in microgram quantities — 5 µg and 25 µg fills are typical, with milligram fills sold at a steep premium — supplied carrier-free or with a carrier protein such as BSA, and labeled for laboratory research use only.
• Bacterially expressed (non-glycosylated) and mammalian-cell expressed (glycosylated) versions are both offered; Jiang et al. (2004) showed glycosylation is not required for myostatin inhibition, but the two are not interchangeable for every use. Catalog items also differ in ways the label may not foreground: many carry a polyhistidine purification tag, and some widely stocked ones are the mouse rather than the human sequence.
• Fc-fusion constructs are catalogued by some protein suppliers, mostly as mouse propeptide-Fc chimeras; the cleavage-resistant D76A mutants that produced much of the in vivo data are not routine catalog items and are made to order or expressed from a vector.
• No approved, compounded, or oral presentation exists. A protein of this size would not survive oral administration, and nothing in the clinical or pharmacy supply chain corresponds to it.

This section describes formats only and is not dosing guidance.

Key considerations

• It is a protein, not a peptide, and that changes everything downstream. At 243 residues and roughly 28 kDa, myostatin propeptide has protein-scale requirements for expression system, folding, storage, and reconstitution, and its identity and purity cannot be confirmed by the simple mass-spectrometry checks that work for a 5-to-40-residue synthetic chain. Certificates of analysis written for short peptides do not establish that a protein of this size is correctly folded or active.
• There is no human data of any kind, and the pathway’s human record is poor. No registered trial, no pharmacokinetics, no safety database. Where the myostatin pathway has been tested in people with other molecules, the recurring result has been increases in muscle size that were small or statistically non-significant, with no improvement in strength or function.
• The sold format does not match the studied format. Nearly all of the in vivo muscle growth attributed to “myostatin propeptide” came from AAV gene delivery, IgG-Fc fusion, or the cleavage-resistant D76A mutant — often more than one at once. Every study that injected the protein systemically used a stabilized version, because wild-type propeptide is a substrate for the BMP-1/tolloid proteases that release myostatin. Claims built on the animal literature are describing engineered constructs, not the plain recombinant domain.
• Prohibited in sport, by name. The WADA Prohibited List names “myostatin propeptide” explicitly, alongside follistatin, as an example of a prohibited myostatin-binding protein under S4.3, Agents Preventing Activin Receptor IIB Activation — prohibited at all times, in and out of competition, and classified as a non-Specified Substance. The same section names apitegromab, domagrozumab, landogrozumab, and stamulumab in its adjacent antibody bullet.
• Several names collide here. Myostatin propeptide is the inhibitor; myostatin is the growth factor it inhibits, so the two entries describe opposite biology. Follistatin 344 is a different myostatin-binding protein with a different binding profile that also captures activins. Searches on “myostatin propeptide” additionally return anti-propeptide antibodies and unrelated short “myostatin inhibitory peptides,” none of which are this molecule.