Myostatin (GDF-8)

Myostatin is synthesised as a precursor, cleaved into a prodomain and a mature dimer, and kept latent in circulation until further proteolysis frees it. Active myostatin binds ActRIIB (and ActRIIA), which recruits ALK4/ALK5 and phosphorylates Smad2/3, inhibiting the Akt/mTOR protein synthesis pathway and upregulating atrophy-associated ubiquitin ligases. Natural loss-of-function mutations in cattle, dogs, sheep and a small number of humans produce dramatic muscle hypertrophy, which is the entire proof of concept for this drug class. Its physiological role goes beyond muscle - it influences adipose tissue, bone and metabolic rate - which is why blocking it has broader consequences than pure hypertrophy.

Targets: Activin receptor type IIB (ActRIIB), Activin receptor type IIA (ActRIIA), ALK4/ALK5, Smad2/3 pathway

• · Recombinant myostatin is used as a laboratory reagent and administered to animals in cachexia research. There is no scenario in which giving yourself myostatin is desirable - it causes muscle loss.

Work out your exact syringe units →

• very commonMuscle atrophy— This is what myostatin does. Administering it is the opposite of the goal.
• commonCachexia in animal models— Systemic myostatin overexpression produces profound wasting in rodents.

• There is no therapeutic use case. Do not inject this.

• conflictfollistatin-344 — Follistatin exists to neutralise myostatin; they cancel each other out.
• conflictapitegromab — Apitegromab blocks myostatin activation.
• conflicttrevogrumab — Trevogrumab neutralises mature myostatin directly.

• · Serum myostatin assays exist but are poorly standardised and cross-react with GDF-11, so a number from one is not clinically meaningful.

Not a medicine. Recombinant GDF-8 is sold as a research reagent for laboratory use only.

• McPherron, Lawler and Lee 1997, Nature - myostatin knockout mice show dramatic skeletal muscle hypertrophy (preclinical)
• Schuelke et al. 2004, NEJM - myostatin mutation associated with gross muscle hypertrophy in a child (other)
• Reviews of myostatin signalling and therapeutic inhibition strategies (review)

Myostatin is in this class as the target rather than the therapy, and understanding its activation cascade is what makes sense of why the drugs against it differ so much. The gene product is a 375-residue precursor. The signal peptide goes, a furin-type protease cleaves between the propeptide and the growth-factor domain, and the two stay non-covalently associated as a latent complex - the growth factor is made and then immediately holstered. A tolloid-family metalloprotease cutting the propeptide is what draws it. Once free, the mature dimer binds ActRIIB with high affinity and ActRIIA more weakly, the type II receptor recruits and transphosphorylates ALK4 or ALK5, and those phosphorylate Smad2 and Smad3. The Smad complex with Smad4 enters the nucleus and does two things: it suppresses the transcription of genes driving myoblast differentiation, notably by downregulating MyoD and Pax7, and it acts in concert with FoxO to transcribe the atrophy ubiquitin ligases MAFbx/atrogin-1 and MuRF1, which tag myofibrillar protein for proteasomal destruction. Smad3 also inhibits Akt, so myostatin signalling suppresses mTORC1 as a second, independent mechanism. The whole architecture - latent reservoir, protease-gated activation, multiple extracellular inhibitors including follistatin, FSTL3, GASP-1 and the propeptide itself - is a system built for tight regulation, and it gives drug designers four separate points of attack. Apitegromab blocks activation, trevogrumab neutralises the mature ligand, follistatin traps it, bimagrumab and ACE-031 block or occupy the receptor. What the human genetics show is worth stating precisely: McPherron's myostatin-null mice were dramatically more muscular, the Belgian Blue and Piedmontese double-muscled cattle carry loss-of-function myostatin mutations, and Schuelke described a child with a myostatin mutation and gross muscle hypertrophy in the New England Journal of Medicine in 2004. Lifetime absence of myostatin during development produces an extraordinary phenotype. Blocking it pharmacologically in an adult does not, and Salzler's proteomics showed directly that genetic deficiency and antibody blockade produce different muscle. That gap between the genetics and the drugs is the single most important thing to carry away from this record.

The failure mode here is conceptual rather than clinical, and it is the most consequential error in this whole class. People see the Belgian Blue photograph and the Schuelke case report and conclude that blocking myostatin in an adult will approximate the phenotype. It will not. Those phenotypes are the product of complete absence of the ligand throughout embryonic and postnatal development, when muscle fibre number is being determined - and adults cannot add fibres. Salzler showed directly that antibody blockade does not reproduce the proteomic signature of genetic deficiency. Mariot went further and showed that in the neuromuscular diseases these drugs were built for, the myostatin pathway is already downregulated, which is a large part of why the trials underperformed. And RESILIENT showed that even when a drug in this class does add muscle mass, function need not follow. The second error is treating serum myostatin measurements as meaningful: commercial ELISAs detect different things and are not comparable.

• Activin receptor type IIB (ActRIIB) — The high-affinity receptor for mature myostatin

  Recruits and transphosphorylates ALK4/ALK5, initiating Smad2/3 signalling. This is the node bimagrumab and ACE-031 block.

• Activin receptor type IIA (ActRIIA) — Lower affinity than ActRIIB but functionally relevant - Morvan showed dual blockade is needed for maximal hypertrophy

  Same downstream cascade. Its existence is why ActRIIB-only strategies leave growth available.

• ALK4 (ACVR1B) and ALK5 (TGFBR1) — Type I receptors recruited by the ligand-bound type II receptor rather than binding myostatin directly

  Phosphorylate Smad2 and Smad3, which complex with Smad4 and translocate to the nucleus.

• Follistatin, FSTL3, GASP-1, GASP-2 and the myostatin propeptide — High-affinity endogenous inhibitors that bind the ligand rather than the receptor

  The body's own myostatin-blocking system. Follistatin therapy is an attempt to run it harder from outside.

Not applicable as an administered compound. For orientation on the pathway: resistance training reduces skeletal muscle myostatin mRNA expression within hours to days of a session, disuse and bed rest raise it within days, and pharmacological blockade produces measurable lean mass change over four to twelve weeks in the trials that show any.

There is nothing to stack, because myostatin is not administered. The practically useful version of this section is which pathway-targeting agents make sense together, and the answer is essentially none of them: follistatin, ACE-031, bimagrumab, apitegromab and trevogrumab all converge on the same signalling node, so combining them adds risk without adding mechanism. The one exception is the deliberate modular design in COURAGE, where an anti-myostatin antibody and an anti-activin A antibody are combined specifically because they block different ligands feeding the same receptor. What does combine sensibly with myostatin blockade is anything working through a different pathway - resistance training first and foremost, which downregulates myostatin expression on its own, then GLP-1 agonists for the composition question, and androgens or IGF-1 signalling for independent anabolic input.

The useful comparison is not between myostatin and other compounds but between the points on its own pathway where drugs act. Blocking activation of the latent complex, as apitegromab does, is the most selective and gives the smallest, cleanest effect. Neutralising the mature ligand, as trevogrumab does, is next. Trapping the ligand with follistatin catches activins and BMPs as well and brings endocrine consequences. Occupying the receptor with a soluble ActRIIB decoy, as ACE-031 does, blocks everything and produced telangiectasia and epistaxis. Blocking the receptors with an antibody, as bimagrumab does, gives the largest documented body-composition effect with muscle spasms as the characteristic cost. Effect size and risk rise together as you move down that list, and there is currently no agent that escapes the trade.

Not a purchasable therapeutic. Recombinant myostatin is sold as a laboratory reagent for assay work and has no consumer application.

• ?Circulating myostatin half-life is not well characterised in humans, and figures differ depending on whether the latent complex or the free mature dimer is being measured.
• ?Commercial serum myostatin immunoassays are not standardised and detect different molecular species, so absolute values are not comparable across assays or studies.
• ?The extent to which resistance training alters circulating rather than intramuscular myostatin in humans is inconsistent across studies.
• ?Whether pharmacological myostatin blockade in a healthy trained adult produces any hypertrophy at all beyond that achieved by training has not been tested.
• ?The relative contribution of myostatin versus activin A to restraining muscle mass in humans is unresolved - it is precisely what the COURAGE arms are designed to separate.

• Regulation of skeletal muscle mass in mice by a new TGF-beta superfamily member McPherron AC, Lawler AM, Lee SJ, Nature, 1997 · PMID 9139826

  The paper that identified myostatin and showed that deleting it produces dramatically more muscular mice. Everything in this class descends from it.

• Double muscling in cattle due to mutations in the myostatin gene McPherron AC, Lee SJ, Proceedings of the National Academy of Sciences USA, 1997 · PMID 9356471

  Belgian Blue carries an 11-nucleotide frameshift deletion, Piedmontese a missense mutation. The double-muscled cattle phenotype is myostatin loss of function, confirmed genetically.

• Myostatin mutation associated with gross muscle hypertrophy in a child Schuelke M, Wagner KR, Stolz LE, Hubner C, Riebel T, Komen W, Braun T, Tobin JF, Lee SJ, New England Journal of Medicine, 2004 · PMID 15215484

  The human case. A child with a myostatin mutation and gross muscle hypertrophy - the proof that the pathway operates the same way in people, and the origin of essentially all the marketing in this category.

• Myostatin deficiency but not anti-myostatin blockade induces marked proteomic changes in mouse skeletal muscle Salzler RR, et al., Proteomics, 2016 · PMID 27214824

  The correction to the enthusiasm. Lifetime genetic deficiency and pharmacological blockade in an adult do not produce the same muscle.

• Downregulation of myostatin pathway in neuromuscular diseases may explain challenges of anti-myostatin therapeutic approaches Mariot V, Joubert R, Hourde C, Feasson L, Hanna M, Muntoni F, Maisonobe T, Servais L, Bogni C, Type R, Hogrel JY, Roux AL, Domingos J, Bassez G, Butler-Browne G, Bonne G, Voit T, Mouly V, Dumonceaux J, Nature Communications, 2017 · PMID 29192144

  Explains why anti-myostatin drugs underperform in the diseases they were designed for: the pathway is already downregulated, so there is less to block.

• UniProtKB O14793 (GDF8_HUMAN) - Growth/differentiation factor 8 UniProt Consortium, UniProt Knowledgebase

  Source of the 375-residue precursor architecture, the 23-residue signal peptide, the 243-residue propeptide and the 109-residue mature chain sequence given above.