ATP drives the normal actomyosin cross-bridge cycle in which M.ADP.P_i binds to actin and undergoes a working stroke during the release of products and formation of the rigor state (A.M). In recent years optical tweezers have been used by several laboratories to measure this working stroke at the single molecule level. The consensus value is around 6 nm. Other ligands such as pyrophosphate (PP_i) and ADP, which do not go through a hydrolytic cycle and are not a source of energy, do however weaken the interaction between myosin and actin, although to a lesser extent than ATP. When myosin states such as M.ADP and M.PP_i bind to actin do they still perform a working stroke? In the experiments, an actin filament is suspended between two optically trapped polystyrene beads and stretched taut, forming a ‘dumb-bell’. The dumb-bell is then positioned over a fixed bead carrying a low density of S1 molecules. The images of the trapped beads are projected onto quadrant detector photodiodes to record bead displacements. In the absence of interaction the position signal is the noise due to the Brownian motion of the dumb-bell controlled by the trap stiffness, whereas when the S1 attaches to the actin filament, the S1 stiffness comes into play and reduces the position noise. In this way it is possible to distinguish between bound and unbound states and to measure the S1 working stroke.

Experiments with ATP (5 mM) gave working strokes in the range 5Ð6 nm and interaction half-lives of about 40 ms. With ADP (500 mM) interactions were much more prolonged (t_1/2 ~1.4 s) particularly when hexokinase (1 mg ml^-1) was used to eliminate ATP contamination. Our estimate of the working stroke was 0.05 ± 0.7 nm.

With PP_i we found that the rate of M.PP_i binding to actin was higher than with M.ADP.P_i and it was necessary to add 40 mM KCl to reduce this rate to allow the characterisation of single interactions. The length of interactions was dependent upon [PP_i], the second-order binding constant being 10⁵ M^-1 s^-1, which shows that during the interaction the A.M state must have been formed. The estimate of the working stroke was 0.2 ± 0.5 nm.