When the sliding filament hypothesis was proposed in 1953-1954, existing ev
idence showed that (1) contributions to tension were given by active sites
uniformly distributed within each zone of filament overlap and (2) each sit
e functioned cyclically. These sites were identified by electron microscopy
as cross-bridges between the two filaments, formed of the heads of myosin
molecules projecting from a thick filament and attaching to a thin filament
. The angle of these cross-bridges was found to be different at rest and in
rigor, suggesting that the event causing relative motion of the filaments
was a change of the angle of the cross-bridges. At first, it seemed likely
that the whole cross-bridge rotated about its attachment to actin, but when
the atomic structures of actin and myosin were obtained by X-ray crystallo
graphy, a possible hinge was found between the "catalytic domain" which att
aches to the actin filament and the "light-chain domain" which appears to a
ct as a lever arm. Two attitudes of the lever arm are now well established,
the transition between them being driven by a conformational change couple
d to some step in the hydrolysis of ATP, but several recent observations su
ggest that this is not the whole story: a third attitude has been shown by
X-ray crystallography; a non-muscle myosin has been shown to produce its wo
rking stroke in two steps; and there are suggestions that an additional dis
placement of the filaments is produced by a change in the attitude of the c
atalytic domain on the thin filament. (C) 2000 Published by Elsevier Scienc
e Ltd.