The ClpAP protease from E. coli is composed of a proteolytic component (ClpP) and a regulatory ATPase component (ClpA). Ultracentrifugation has been conducted at pH 7.5 (0.1-0.3 M KCl +/- 10% glycerol) and 4 or 20 degreesC to determine the macromolecular size and shape of the ClpAP components. Without nucleotide, ClpA (subunit MW of 84,150) exists in a monomer-dimer equilibrium with log K = 5 (apparent s/20,w = 8.7 S). With the non-hydrolyzable analogue adenosine-5'-O-(3-thiotriphosphate) (ATPgammaS) present, ClpA associates to a hexamer with MW = 505,000 +/- 3000 and s/20,w = 17.2 S. The inactive pro-ClpP(SA) (13.2 S) and mature ClpP (12.2 S) have identical subunits of 23,164 and 21,558 MW, respectively, and appear as two rings of seven subunits in electron microscopic image reconstructions (M. Kessel et al., J. Mol. Biol. 250, 587-594, 1995). Ultracentrifugal studies confirm that both the pro-Clp(SA) mutant and ClpP are 14-mers with molecular weights of 324,000 and 302,000 +/- 5%, respectively. Nucleotide-bound, hexameric ClpA is required for association with ClpP to form the active ClpAP protease; log K = 8.5 /(M ClpA hexamer or M ClpP tetradecamer) under assay conditions. Sedimentation velocity experiments in the presence of ATPgammaS, Mg(II), and varying molar ratios of hexameric ClpA to tetradecamer ClpP show two complexes: s/20,w = 21.0 +/- 0.4 S and 27.3 +/- 0.7 S. These correspond to 1:1 and 2:1 complexes of hexameric ClpA and oligomeric ClpP, with MW approximately 807,000 (frictional ratio = 1.6) and MW approximately 1,310,000 (frictional ratio = 1.7), respectively. Both complexes are active in casein or propeptide degradation and both are seen in electron micrographs. Recently, it has been established that ClpP undergoes a temperature- dependent, reversible dissociation of the two rings (containing seven subunits/ring) in 0.1 M sodium sulfate at pH 7. At 5 _C, the protein sedimented as a monodisperse boundary corresponding to the size of half molecules of ClpP (s/20,w = 7.9 S). Warming from 5 to 20 _C, produced complete reassociation of ClpP to its original size (s/20,w = 12.2 S). Cooling from 20 to 5 degreesC and centrifuging at 52,000 rpm within 1 h, produced two boundaries (approximately 8 and 12 S/20,w), which suggests that dissociation is slower than reassociation kinetics. Dr. Zolkiewski has initiated studies on the oligomeric structure of E. coli ClpB, a protein homologous to ClpA. ClpB, like ClpA, has protein activated ATPase activity but unlike ClpA, ClpB does not interact with ClpP. The sizes of ClpB complexes at different protein concentrations in the absence and presence of nucleotides (ATP, ATPgammaS, and ADP) are being determined by sedimentation velocity, sedimentation equilibrium, HPLC, gel filtration, and chemical cross linking with glutaraldehyde.
