Under standard conditions, G-actin has been submitted to nine proteases of varying specificity, and in each case the pattern of fragments produced has been studied by NaDodSO4 gel electrophoresis. The results suggest that the actin monomer consists of a large region (ca. 33 kilodaltons) and a small, easily degraded region (ca. 9 kilodaltons). The COOH terminus is in the large region. Consideration of primary sequence homologies, medium resolution maps of actin crystals, and certain reactions of actin suggests that the NH2 terminus is in the small region, as is the negative sequence to which a divalent metal cation is normally chelated, but that the nucleotidebinding site is on the large region near the junction between the regions. From analysis of these results, numerous properties of actin are understandable.Actin is one of the two major proteins comprising the contractile system of muscle. The structure of actin monomers and their assembly into F-actin filaments are being studied by various physical methods (1, 2). Here our methods and conclusions are chemical, but the latter seem to have structural implications. Guided by previous studies of actin proteolysis (3-7), we have submitted monomeric actin in solution to a battery of proteases in a test for the possible existence of domains within the monomer (see ref. 8 for a similar study of myosin S1). Analysis of the results shows that actin consists of two regions-a large, protease-resistant region of 33 kilodaltons (kDa) and a small, protease-vulnerable region of 9 kDa. The NH2 terminus and the divalent metal cation (M2+)-binding regions are probably in the smaller region, whereas the COOH terminus is in the larger region, and the nucleotide-binding site is probably in the larger region near the junction with the smaller. From this knowledge and certain steric considerations, many characteristic properties of Gand F-actin can be rationalized.
MATERIALS AND METHODSPreparation of Actin and Myosin. Rabbit skeletal muscle Factin was prepared as described by Eisenberg and Kielley (9), and its concentration was estimated by using Aj1.8 = 11.0 (10).G-actin was obtained by three successive dialyses of Factin (3-6 mg/ml) with a 5-mM 2-{[tris(hydroxymethyl)methyl]amino}ethanesulfonic acid (Tes) buffer (pH 7.6) containing 0.2 mM ATP, 0.2 mM CaCl2, 0.1 mM dithiothreitol, and 0.1 mM NaN3 (buffer A). Its concentration was then estimated by using Al" = 6.37 (11).For proteolysis, we always used G-actin solutions no older than 1 week, and we verified that these solutions were not denatured by adding 5 mM MgCI2 and observing (viscometrically) F-actin polymerization as rapid as with fresh G-actin.When labeled F-actin was used, the dye was N-iodoacetyl-N'-(5-sulfo-1-naphthyl)ethylenediamine (1,5-IAEDANS), known to react specifically with cysteine-374 (12). Then this fluorescent F-actin was dialyzed against dye-free solvent for 48 hr, like native actin, to obtain fluorescent G-actin.Rabbit skeletal myosin was prepared according to Offer et al. (13); subfragment 1 (Si) was...