1. The effects of MgADP on cross-bridge kinetics were investigated using laser flash photolysis of caged ATP (P3-1(2-nitrophenyl) ethyladenosine 5'-triphosphate), in guinea-pig portal vein smooth muscle permeabilized with Staphylococcus aureus alpha-toxin. Isometric tension and in-phase stiffness transitions from rigor state were monitored upon photolysis of caged ATP. The estimated concentration of ATP released from caged ATP by high-pressure liquid chromatography (HPLC) was 1.3 mM. 2. The time course of relaxation initiated by photolysis of caged ATP in the absence of Ca2+ was well fitted during the initial 200 ms by two exponential functions with time constants of, respectively, tau 1 = 34 ms and tau 2 = 1.2 s and relative amplitudes of 0.14 and 0.86. Multiple exponential functions were needed to fit longer intervals; the half-time of the overall relaxation was 0.8 s. The second order rate constant for cross-bridge detachment by ATP, estimated from the rate of initial relaxation, was 0.4-2.3 x 10(4) M-1 s-1. 3. MgADP dose dependently reduced both the relative amplitude of the first component and the rate constant of the second component of relaxation. Conversely, treatment of muscles with apyrase, to deplete endogenous ADP, increased the relative amplitude of the first component. In the presence of MgADP, in-phase stiffness decreased during force maintenance, suggesting that the force per cross-bridge increased. The apparent dissociation constant (Kd) of MgADP for the cross-bridge binding site, estimated from its concentration-dependent effect on the relative amplitude of the first component, was 1.3 microM. This affinity is much higher than the previously reported values (50-300 microM for smooth muscle; 18-400 microM for skeletal muscle; 7-10 microM for cardiac muscle). It is possible that the high affinity reflects the properties of a state generated during the co-operative reattachment cycle, rather than that of the rigor bridge. 4. The rate constant of MgADP release from cross-bridges, estimated from its concentration-dependent effect on the rate constant of the second (tau 2) component, was 0.35-7.7 s-1. To the extent that reattachment of cross-bridges could slow relaxation even during the initial 200 ms, this rate constant may be an underestimate. 5. Inorganic phosphate (Pi, 30 mM) did not affect the rate of relaxation during the initial approximately 50 ms, but accelerated the slower phase of relaxation, consistent with a cyclic cross-bridge model in which Pi increases the proportion of cross-bridges in detached ('weakly bound') states.(ABSTRACT TRUNCATED AT 400 WORDS)