Key findings
- The original isolation separated a single-chain protein of apparent mass 35,000 that is highly enriched in cysteines and shares no sequence homology with the heterodimeric inhibins, and also recovered a second form of apparent mass 32,000 present at much lower concentration. [1]
- The crystal structure of the activin complex shows two follistatin molecules encircling the ligand, burying one third of its residues, and an N-terminal domain that folds into a mimic of a universal type I receptor motif and occupies that receptor site. [2]
- The way the protein grips its ligands differs between them: the N-terminal domain rearranges conformationally to bind myostatin, while activin B makes direct contacts with the third follistatin domain and disrupts the head-to-tail cooperativity seen with activin A. [3][4]
- Heparin binding is coupled to ligand binding in both directions: complexes bind heparin far more tightly than free protein, and heparin chains above a threshold length dimerise the protein and slow the association of ligands without changing dissociation. [5][6]
Identity and structure
- Form and apparent mass
- A single polypeptide chain highly enriched in cysteines with an apparent mass of 35,000, with a second form of apparent mass 32,000 present at much lower concentration that may differ in glycosylation or carboxyl-terminal truncation [1]
- Distinct from the inhibins
- The isolated protein has no sequence homology with the follicular fluid inhibins, which are heterodimeric proteins of apparent mass 32,000 with the same activity in the assay used [1]
- Isoforms used in the cited work
- The structural and biophysical papers use the 288-residue isoform, and one of them compares it against the 315-residue isoform; the two behave differently with heparin [3][6]
- Fusion constructs
- Dimeric follistatin Fc fusion proteins circulate in the same literature and are distinguished from naturally occurring isoforms only by methods that resolve the intact protein [7]
- Form as supplied
- Sterile lyophilized powder
Mechanism as studied
The antagonism is steric and complete. Two molecules of the protein wrap the activin dimer and bury roughly a third of its residues along with both receptor binding surfaces. Earlier expectations had been that the type I receptor site would stay open, but the structure shows the N-terminal domain adopting an unexpected fold that mimics a universal type I receptor motif and filling that site. [2]
Ligand identity changes the details of that grip. With myostatin the N-terminal domain undergoes conformational rearrangement and acts as the site of specificity, and the prehelix region of myostatin closely resembles that of the TGF-beta class. With activin B the fingertips of the ligand contact the third follistatin domain directly, which disrupts the intermolecular head-to-tail cooperativity that stabilises the activin A complex and shifts stabilisation onto individual domain affinities. [3][4]
Heparin is the third variable and it works in both directions. Binding myostatin creates a continuous electropositive surface that raises heparin affinity and strengthens interaction with the cell surface, while preincubation with heparin chains long enough to dimerise the protein slows the association of both myostatin and activin A without altering the dissociation rate or the antagonistic activity. [3][6]
Research findings
Analytical
- System
- Porcine ovarian follicular fluid, fractionated by heparin-Sepharose affinity chromatography, gel filtration on Sephacryl S-200 and multiple high performance liquid chromatography steps, assayed in rat anterior pituitary monolayer culture
- Measured
- Apparent molecular mass, chain composition, cysteine content, sequence homology against the follicular fluid inhibins, and selectivity for follicle-stimulating hormone against luteinizing hormone secretion
- Reported
- A single-chain protein of apparent mass 35,000, highly enriched in cysteines and with no sequence homology to the heterodimeric inhibins, specifically inhibited basal follicle-stimulating hormone secretion but not luteinizing hormone secretion. A second form of apparent mass 32,000 was isolated at much lower concentration. [1]
Structural
- System
- Crystal structure of activin A bound to follistatin
- Measured
- Arrangement and stoichiometry of the complex, the proportion of ligand residues buried, occupancy of the type I and type II receptor binding sites, and the fold of the N-terminal domain
- Reported
- Two follistatin molecules encircle activin, burying one third of its residues and both of its receptor binding sites. The N-terminal domain adopts an unexpected fold that mimics a universal type I receptor motif and occupies that site, contrary to earlier expectations that type I receptor binding would remain open. [2]
Structural
- System
- Crystal structure of myostatin in complex with the 288-residue isoform, with a chimeric activin A carrying the myostatin prehelix region and with the 315-residue isoform in the degradation assays
- Measured
- Conformation of the myostatin prehelix region against TGF-beta class members, signalling through the non-canonical type I receptor Alk5 after the region was swapped into activin A, conformational rearrangement of the N-terminal domain, the electropositive surface created on complex formation, heparin affinity, and ligand degradation
- Reported
- The myostatin prehelix region closely resembled that of TGF-beta class members and alone conferred Alk5 signalling when swapped into activin A. The N-terminal domain rearranged to bind myostatin and acted as the site of specificity, complex formation created a unique continuous electropositive surface that significantly raised heparin affinity, and this translated into stronger cell surface interaction and enhanced myostatin degradation with either isoform. [3]
Structural
- System
- Crystal structure of activin B bound to the 288-residue isoform at 2.7 Angstrom resolution, with computational comparison against activin A and GDF8
- Measured
- Contacts between the ligand fingertips and the third follistatin domain, the intermolecular head-to-tail cooperativity of the follistatin dimer, and the relative flexibility of activin B, activin A and GDF8
- Reported
- Activin B engages the antagonist in a modified mode: its fingertips form direct contacts with the third follistatin domain, disrupting the head-to-tail cooperativity normally seen in the dimer and shifting stabilisation onto individual ligand-domain affinities. Computation indicated activin B is more flexible than activin A and GDF8, and that activin A relies more on the intermolecular interactions. [4]
Structural
- System
- The 288-residue isoform preincubated with heparin or heparan sulfate of defined chain length, with myostatin and activin A as ligands
- Measured
- Association and dissociation kinetics by surface plasmon resonance as a function of heparin chain length, dimerisation of the protein, and the solution configuration by small-angle X-ray scattering
- Reported
- Preincubation with heparin or heparan sulfate significantly slowed the association kinetics for both myostatin and activin A with no apparent effect on dissociation, and this happened only with chains longer than a degree of polymerisation of 10. Heparin above that length induced dimerisation, and scattering showed a dimer configuration similar to the ligand-bound state, with the antagonistic activity unchanged. [6]
In vitro
- System
- Free 288-residue isoform and its complexes with activin A and with myostatin, against heparin polysaccharide and heparin oligosaccharides of defined length, across salt concentrations and pH values
- Measured
- Binding affinity for heparin by surface plasmon resonance, chain length dependence using oligosaccharide competition, and sensitivity of each interaction to salt and to pH
- Reported
- The complexes bound heparin far more tightly than the free protein, and binding depended on chain length, with full-length heparin or oligosaccharides of 18 to 20 sugar residues showing the highest activity for the free protein and the activin complex while smaller chains could still engage the myostatin complex. Unbound protein was the most sensitive to salt above 150 millimolar, heparin binding to the activin complex was disrupted at 500 millimolar while that of the myostatin complex was strengthened, and acidic pH enhanced binding except at pH 4 for the myostatin complex. [5]
Structural
- System
- The isolated complex of inhibin A with the 288-residue isoform, examined against the known homodimeric ligand complexes and against activin receptor type IIb
- Measured
- Stoichiometry of the complex, its low-resolution structure by small-angle X-ray scattering with modelling, heparin affinity, dissociation on receptor binding by surface plasmon resonance, and antagonism in an activin-responsive luciferase assay
- Reported
- The complex formed at a one to one ratio rather than the one to two ratio of the homodimeric ligand complexes, with binding through the shared activin beta chain and the alpha chain left free and flexible. It bound heparin with lower affinity than the free protein or the activin A complex, and it readily dissociated on binding activin receptor type IIb, so both antagonists remained able to inhibit activin signalling. [8]
Analytical
- System
- Serum and plasma spiked with monomeric protein and with dimeric Fc fusion constructs as model compounds
- Measured
- Detection by immunoaffinity purification, tryptic digestion and liquid chromatography with high resolution tandem mass spectrometry targeting tryptic signature peptides, and confirmation by immunoaffinity purification, SDS-PAGE and Western blotting of the intact proteins
- Reported
- Both assays were highly specific with an estimated detection limit of 10 nanograms per millilitre, and the Western blotting confirmation distinguished synthetic Fc fusion constructs from the naturally occurring isoforms, which the mass spectrometry step alone cannot do because the protein is also produced endogenously. [7]
Handling for in-vitro work
- Heparin affinity in cited work
- The protein was purified on heparin-Sepharose in the original isolation, and heparin or heparan sulfate in a buffer changes both its oligomeric state and its ligand association kinetics [1][6]
- Ionic strength in cited work
- Unbound protein was the most sensitive to salt concentrations above 150 millimolar in the cited surface plasmon resonance work, while ligand-bound complexes behaved differently from one another [5]
- Storage
- Lyophilized at -20 °C, dark and dry; reconstituted aliquots kept cold and used promptly
Open questions
- The 288-residue and 315-residue isoforms behave differently with heparin in the cited work, and a listing giving only the family name does not say which isoform was supplied.
- The original isolation recovered two forms that may differ in glycosylation or carboxyl-terminal truncation, so the expression system and glycosylation state of a supplied lot remain open questions.
- Fc fusion constructs are present in the same literature and the cited analytical work needed an intact-protein method to tell them from naturally occurring isoforms.
- The structural work characterises complexes with activin A, activin B, myostatin and inhibin A, and shows the binding mode differs between them, so a result for one ligand does not carry to another.
Clinical literature associated with this protein concerns engineered constructs and finished formulations rather than a research material, and is out of scope for this profile.
Lot records
Check the record for the exact material you order. A published paper and a batch certificate answer different questions.
No published lot is available for this exact compound name.
Read a certificate of analysis ↗References
- Ueno N, Ling N, Ying SY, et al. Isolation and partial characterization of follistatin: a single-chain Mr 35,000 monomeric protein that inhibits the release of follicle-stimulating hormone. Proceedings of the National Academy of Sciences of the United States of America. 1987.
- Thompson TB, Lerch TF, Cook RW, et al. The structure of the follistatin:activin complex reveals antagonism of both type I and type II receptor binding. Developmental cell. 2005.
- 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.
- Hok L, Walker RG, Howard JA, et al. The crystal structure of the activin B:Fst288 complex and computational insights into the broad antagonistic activity and specificity of follistatin. The Journal of biological chemistry. 2026.
- Zhang F, Beaudet JM, Luedeke DM, et al. Analysis of the interaction between heparin and follistatin and heparin and follistatin-ligand complexes using surface plasmon resonance. Biochemistry. 2012.
- Walker RG, Kattamuri C, Goebel EJ, et al. Heparin-mediated dimerization of follistatin. Experimental biology and medicine (Maywood, N.J.). 2021.
- Walpurgis K, Weigand T, Knoop A, et al. Detection of follistatin-based inhibitors of the TGF-β signaling pathways in serum/plasma by means of LC-HRMS/MS and Western blotting. Drug testing and analysis. 2020.
- Kappes EC, Kattamuri C, Czepnik M, et al. Follistatin Forms a Stable Complex With Inhibin A That Does Not Interfere With Activin A Antagonism. Endocrinology. 2023.
Publication records fetched from PubMed on 2026-09-20. Profile text reviewed 2026-09-20.