Fluctuation correlation in the sliding movement generated by protein motors in vitro.

The fluctuation in the sliding distance of cytoskeletal filaments driven to move by protein motors in vitro does not depend on the filament length. This is in sharp contrast to the case of Brownian movement of filamentous particles in their longitudinal directions, in which the positional fluctuation is proportional to the inverse of the length (L) of filaments. This latter 1/L dependence is a direct consequence of the central limit theorem: the statistical independence and randomness of the solvent molecule collisions with filaments, the collisions of which cause the random Brownian movement. The above length-independence in the sliding distance fluctuation found in the in vitro motility indicates the presence of correlation in the fluctuation. A possible explanation for the correlation is to assume that there is an extended time-correlation in the sliding movement, a correlation which could be produced by the actions on a sliding filament of protein motors with their heads randomly oriented in the in vitro motility assay system. We have checked this possibility by using long myosin thick filaments of molluscan smooth muscles, on which myosin heads are uniformly oriented, and have found that even with such myosin filaments with oriented myosin heads, the positional fluctuation of actin sliding distance does not depend on the actin filament length. This result thus indicates that the actions of protein motors on a sliding filament are not statistically independent or random, so that the positional fluctuation of filaments in the motor-generated sliding movement does not depend on the filament length.