The myosin head consists of a globular catalytic domain and a light chain binding domain (LCBD). The coupling efficiency between ATP hydrolysis and myosin‐induced actin movement is known to decline as the LCBD is truncated or destabilized. However, it was not clear whether the observed alteration in the production of force and movement reflects only the mechanical changes to the length of the LCBD or whether these changes also affect the kinetic properties of the catalytic domain. Here we show that replacement of the LCBD with genetically engineered domains of similar rigidity and dimensions produces functional molecular motors with unchanged kinetic properties. The resulting single‐chain, single‐headed motors were produced in Dictyostelium discoideum and obtained after purification from a standard peptone‐based growth medium at levels of up to 12 mg/l. Their actin motility properties are similar or greater than those of native myosin. Rates of 2.5 and 3.3 microm/s were observed for motor domains fused to one or two of these domains, respectively. Their kinetic and functional similarity to the extensively studied myosin subfragment 1 (S1) and their accessibility to molecular genetic approaches makes these simple constructs ideal models for the investigation of chemo‐mechanical coupling in the myosin motor.

The transient kinetic properties of the recombinant myosin head fragments M761 and M781, which both lack the light chain binding domain (LCBD) and correspond to the first 761 and 781 residues of Dictyostelium discoideum myosin II, were compared with those of the subfragment 1-like fragment M864 and a shorter catalytic domain fragment M754. The properties of M761, M781, and M864 are almost identical in regard to nucleotide binding, nucleotide hydrolysis, actin binding, and the interactions between actin and nucleotide binding sites. Only the rate of the hydrolysis step was significantly faster for M761 and the affinity of M781 for actin significantly weaker than for M864. This indicates that the LCBD plays no major role in the biochemical behavior of the myosin head. In contrast, loss of the peptide between 754 and 761 produced several major changes in the property of M754 as documented previously [Woodward, S. K. A., Geeves, M. A., & Manstein, D. J. (1995) Biochemistry 34, 16056-16064]. We further show that C-terminal extension of M761 with one or two alpha-actinin repeats has very little effect on the behavior of the protein. The recombinant nature of M761 and the fact that it can be produced and purified in large amounts make it an ideal construct for systematic studies of the structure, kinetics, and function of the myosin motor.

The dissociation constant for actin binding to myosin and its subfragments (S1 & HMM) is <<1 microM at physiological ionic strength. Many of the methods used to measure such affinities are unreliable for a Kd below 0.1 microM. We show here that the use of phalloidin to stablise F-actin and fluorescently labelled proteins allows the affinity of actin for myosin S1 to be measured in a simple transient kinetic assay. The method can be used for Kd's as low as 10 nM and we demonstrate that the Kd's can be estimated using only microgram quantities of material. Furthermore we suggest how this method may be adapted for ng quantities of protein. This will allow the affinity of actin for myosin fragments to be estimated for proteins which are difficult to obtain in large quantities i.e. from biopsy material or from proteins expressed in baculovirus.