MECHANISM FOR COUPLING FREE-ENERGY IN ATPASE TO THE MYOSIN ACTIVE-SITE

Acrylamide quenching of tryptophan 510 (Trp510) fluorescence in rabbit skeletal myosin subfragment 1 (S1) indicates the conformation of the probe binding cleft, containing the highly reactive thiol (SH1) and Tr p510, in the presence of nucleotides or nucleotide analogs trapped in the active site of S1 [Park et al. (1996) Biochim. Biophys. Acta 1296, 1-4]. The Trp510 quenching efficiency shows that the probe binding cl eft closes slightly in the presence of beryllium fluoride trapped MgAD P(MgADPBeF(x)-S1) and most tightly in the presence of vanadate trapped MgADP (MgADPVi-S1) with aluminum fluoride and scandium fluoride trapp ed MgADP (MgADPAlF(4)S1) and MgADPScF(x)S1) falling in between in the order eF(x)>MgADPAlF(4)>MgADPScF(x)>MgADPScF(x)>MgADPVi. These nucleot ide analogs are identified with sequential structural changes in MgATP during hydrolysis in the same order with beryllium fluoride occurring earliest in the ATPase cycle. Correlation of the separation distance of the gamma-phosphate analog metal from the oxygen connecting it to t he beta-phosphate in ADP, to the extent of cleft closure, suggests tha t this distance in the nucleotide transition state determines the conf ormation of the probe binding cleft. Trp510 quenching efficiency was a lso measured as a function of the base moiety of the vanadate trapped Mg-nucleotide diphosphate (MgNDPVi-S1). The extent of cleft closure is largest in the presence of the natural substrate NDP and follows the order MgADPVi>MgCDPVi>MgUDPVi>MgIDPVi>MgGDPVi with very little differe nce between MgADPVi and MgCDPVi. These data follow the order of fibers [Pate et al. (1993) J. Biol. Chem. 268, 10046-10053]. In both the fib er and S1, it appears that the 6-position amino group of the bases of ADP and CDP is required to properly anchor the nucleotide in the activ e site, possibly at tyrosine 135 as suggested by X-ray crystallographi c studies [Fisher et al. (1995) Biochemistry 34, 8960-8972]. Finally, the Trp410 quenching efficiency was measured as a function of the size of the divalent cation trapped in the active site of S1 with ADPVi. T hese data failed to show a correlation suggesting that the divalent ca tion is not involved with the propagation of influence from the active site to the probe binding cleft. The forgoing experiments suggest tha t the changing conformation of ATP during hydrolysis, parameterized by the increasing distance between the beta- and the gamma-phosphate, st resses the active site of S1 thought protein-nucleotide contacts at th e gamma-phosphate and nucleotide base. The stress-induced strain in th e cross-bridge may be the mechanism by which energy in ATP is transfer red to the myosin structure.