We have purified a protein from rabbit serum with a molecular weight of 90,000 that inhibits the polymerization of actin measured viscometrically and that we have named "brevin" (from the Latin breviare, to shorten). From the extent of purification, we estimate that this inhibitor constitutes 0.3% of the total protein in plasma and serum. Brevin is also present in sera from humans and rats. Almost all of the activity in blood is extracellular; only 1% is present in platelets or other cellular elements. Several lines of evidence indicate that brevin is the same protein as the factor described by Fagraeus and Norberg [Fagraeus, A. & Norberg, R. (1978) Curr. Top. Microbiol. Immunol. 82, 1-13] as an actin-depolymerizing factor (ADF). If ADF and brevin are identical, then "ADF" is an inappropriate name because we find that the protein shortens actin filaments without depolymerizing them. Thus, brevin causes little change in the critical concentration of monomeric actin, even though the inhibitor binds to monomeric actin complexed to DNase I-agarose. Binding of brevin to filaments was demonstrated by sedimenting the inhibitor with F-actin. From the amounts of actin and brevin sedimented, and from the lengths of filaments measured by electron microscopy, we calculated that the stoichiometry of binding is one brevin molecule per filament over a wide range of inhibitor concentrations. This stoichiometry suggests that brevin inhibits polymerization by binding at the end of elongating actin filaments, a mechanism similar to that proposed for several intracellular actin-binding proteins and for the cytochalasins. Its abundance suggests that brevin plays an important physiological role in serum, but one not directly concerned with intracellular motility. Therefore its relationship to cytoplasmic actin-binding proteins remains to be determined.
Read full abstract