We have used electron paramagnetic resonance to study the orientation of myosin heads in the presence of nucleotides and nucleotide analogs, to induce equilibrium states that mimic intermediates in the actomyosin ATPase cycle. We obtained electron paramagnetic resonance spectra of an indane dione spin label (InVSL) bound to Cys 707 (SH1) of the myosin head, in skinned rabbit psoas muscle fibers. This probe is rigidly immobilized on the catalytic domain of the head, and the principal axis of the probe is aligned nearly parallel to the fiber axis in rigor (no nucleotide), making it directly sensitive to axial rotation of the head. On ADP addition, all of the heads remained strongly bound to actin, but the spectral hyperfine splitting increased by 0.55 ± 0.02 G, corresponding to a small but significant axial rotation of 7°. Adenosine 5′-(adenylylim-idodiphosphate) (AMPPNP) or pyrophosphate reduced the actomyosin affinity and introduced a highly disordered population of heads similar to that observed in relaxation. For the remaining oriented population, pyrophosphate induced no significant change relative to rigor, but AMPPNP induced a slight but probably significant rotation (2.2° ± 1.6°), in the direction opposite that induced by ADP. Adenosine 5′-O-(3-thiotriphosphate) (ATPγS) relaxed the muscle fiber, completely dissociated the heads from actin, and produced disorder similar to that in relaxation by ATP. ATPyS plus Ca induced a weak-binding state with most of the actin-bound heads disordered. Vanadate had negligible effect in the presence of ADP, but in isometric contraction vanadate substantially reduced both force and the fraction of oriented heads. These results are consistent with a model in which myosin heads are disordered early in the power stroke (weak-binding states) and rigidly oriented later in the power stroke (strong-binding states), whereas transitions among the strong-binding states induce only slight changes in the axial orientation of the catalytic domain.
Bibliographical noteFunding Information:
We thank Robert L. H. Bennett for development of EPR data analysis software and spectrometer maintenance, Edmund C. Howard for improve- ments in EPR fitting and simulation programs, and Franz L. Nisswandt and Nicoleta Cornea for development and maintenance of other computational hardware and software. All of the above are affiliated with the University of Minnesota Medical School (Minneapolis, MN). We are especially indebted to Dr. Kalman Hideg, University of Hungary at Pecs, for gener-ously providing us with the spin label InVSL. This work was supported by grants to D. Thomas from the National Institutes of Health (AR32961), the Muscular Dystrophy Association, and the Minnesota Supercomputer Institute.