Analytical Ultracentrifugation: Macromolecular Assoc.

Heart, Lung, And Blood Institute

Investigators

Abstract

Collaborations with several groups have been pursued to further develop the use of analytical ultracentrifugation as a broadly applicable technique for analyzing biological systems in terms of reversible associations among their constituent biopolymers. Collaboration with K. Remmert, Ph.D. and John A. Hammer III, Ph.D : Purified CARMIL (subunit Mr = 121610) preparations in the absence of Capping Protein are homogeneous as evidenced in sedimentation velocity experiments which show a single symmetrical boundary. At 9.4 micromolar subunit, CARMIL has a sedimentation coefficient of 7.1 S, under which conditions the protein is >90% dimer. This indicates that the CARMIL dimer is quite asymmetric with a frictional coefficient of ca. 2.0 or about 2-fold greater than that of a spherical particle. Global fitting of sedimentation equilibrium data at 4 C obtained at 34 and 36 h at 7500 rpm for two CARMIL preparations gave excellent fits to a monomer - dimer equilibrium model with an association constant (log K) of 6.0 /(M monomer) at pH 7.0. No higher oligomeric species than dimer could be detected. Apparently, the dimerization constant is greater at 20 C since only dimer is detected in sedimentation velocity time-derivative profiles for 4-7 micromolar. CARMIL subunit at the higher temperature. Once CARMIL and the Capping heterodimer protein (CP) had been separately purified and the extinction coefficients determined, these proteins were mixed in a known ratio to give 1.7-fold CP:CARMIL for a sedimentation equilibrium experiment. After reaching equilibrium, the data have been analyzed (assuming three models) for the different components present. Only one model was consistent with the fact that CARMIL monomer could not be detected and that the presence of CP promotes dimerization of CARMIL. For the mixture, interference data are fitted well to a model with free CP (calculated from [3.28 micromolar total - complexed CP]), 13-15% CARMIL dimer, and 85-87% CARMIL dimer complexed to one equivalent of CP. Moreover, the calculated association constant, log K = 6.4 (/M) for CP binding to CARMIL dimer from the sedimentation equilibrium data is in agreement with the value measured for the affinity of CP for CARMIL in a BioCore assay. The fact that only one equivalent of CP interacts with the CARMIL dimer suggests that there is severe negative cooperativity in binding the second equivalent of CP or, as is more likely, the dimerization of CARMIL excludes a second binding site for CP; e.g., a tail to head interaction of CARMIL with CP binding to either the head or tail regions of CARMIL. Collaboration with S. G. Kang, Ph.D. & M. R. Maurizi, Ph.D.: In both sedimentation velocity and sedimentation equilibrium experiments, purified human ClpP (hClpP) was shown to be a monodisperse heptamer of 24,166 Mr subunits (7.0 S). This result was surprising since in E. Coli, the two 7-membered rings of ClpP associate to form a stable tetradecamer. Since in the mitochondria hClpP is complexed to hClpX, an approach to equilibrium at low speeds was used to examine mixtures of hexameric hClpX with heptameric hClpP (0.25, 0.50, and 1.0 equiv of hexameric ClpX to 14-mer ClpP in the presence of ATPgammaS. A 2:1 complex between hexameric hClpX and tetradecameric hClpP accounted for ca. 46% of the ClpP present when 1.0 equiv of hexameric ClpX to 14-mer ClpP was present (with ca. 8% free heptameric ClpP). Other possible subassemblies (such as 14-mer ClpP, heptameric ClpP complexed to hexameric ClpX, or 14-mer ClpP complexed to one equiv of hexameric ClpX ) were not present in sufficient quantities to be detected. Thus, the interaction of hexameric ClpX with heptameric ClpP strongly stabilizes the tetradecameric structure of ClpP in a complex flanked by hexameric rings of ClpX (analogous to the E. Coli ClpAP complex). Further analysis of the nucleotide factor-dependent association of ClpXP rings promises to provide unique insights into the effects of structural changes during the catalytic cycle of proteosome-like protease complexes. Collaboration with P.Reddy, Ph.D. and A. Peterkofsky, Ph.D: We were approached to conduct sedimentation equilibrium experiments to detect a monomer-dimer equilibrium, since the elution position of Dr. Reddy s cloned calmodulin-like protein from gel filtration columns indicated a larger size than expected for a monomeric protein. Sedimentation equilibrium studies of this protein showed no evidence of association of monomers in the absence or presence of 1 mM Ca(II) at either 4 or 25 C. A global fit of the data obtained for the protein in the presence of 1 mM Ca(II) at 4 C after 42, 44, and 46 h at 37,000 rpm gave a molecular weight of 4950 or nearly the same value as calculated from the amino acid composition (4962 Mr). A quick mass spectral analysis of this sample gave a biomass of 4980. We conclude that the monomeric protein must have a higher Stokes radius than usual for a particle of only 5000 Mr.