Abstract
Bacterial expressed kinesin motor domains hydrolyze ATP and promote microtubule-dependent motility. It has routinely been assumed that motor domain preparations are monomeric on the basis of the presumption that dimerization is mediated by the stalk region. However, experimental verification of the oligomeric state of the kinesin construct is required to interpret the results from single-molecule motility assays as well as presteady-state kinetic experiments. We have measured directly the state of assembly of three conventional kinesin motor domain constructs-K401, K366, and K341, comprising the N-terminal 401, 366, and 341 amino acids, respectively, of the Drosophila kinesin heavy chain-by sedimentation velocity and sedimentation equilibrium methods in an analytical ultracentrifuge. K401 (MW of ADP complex, 45,532) is a predominantly a dimer with a sedimentation coefficient, s020,w, of 5.06 S, but it is able to self-associate by means of a 1-2-4 mechanism into higher oligomers. Molecular weight measurements establish the dissociation constant for dimerization at 37 +/- 17 nM in the presence of ATP. The dissociation constant in the presence of ADP is 35 +/- 26 nM and in the presence of AMPPNP is 42 +/- 28 nM. The construct K366 (MW of ADP complex, 41,404) is a monomer (measured MW, 41,768 +/- 1219) at concentrations below 4 microM K366, with a sedimentation coefficient, s020,w, of 3.25 S. At higher concentrations, there is evidence for a weak association of K366 to a 1-2-4-8 model with a slight preference for octamer formation. The smallest construct, K341 (MW of ADP complex, 38,274), is a monomer (measured MW, 38,191 +/- 734) up to at least 10 microM total K341 concentration with a sedimentation coefficient, s020,w, of 2.9 S. Thus, the dimerization domain either is between amino acid residues 367 and 401 or is strongly affected by the removal of this region. Higher oligomers of K401 form by a mechanism involving dimers of dimers, and suggest that native kinesin may also undergo self-association. These results have important implications for the interpretation of ATP-dependent motility assays.
