The 1997 Albany Conference: Biomolecular Motors and Nanomachines
![[1997 Albany Conference]](logo2.gif)
COMPUTER MODELLING OF ASSEMBLIES OF CILIARY MOTOR MOLECULES
Ernestina Guevara (1), Michael E.J. Holwill (1,2), Toshikazu Hamasaki (2) and Peter Satir (2)
(1) Department of Physics, Kings College London, Strand, London WC2R 2LS, UK
(2) Albert Einstein College of Medicine, 1300 Morris Park Avenue, Bronx, NY 10461
Computer simulations have been developed to represent experiments carried out by Hamasaki et al. (Biophys. J. 69:2569-2579, 1995), in which dynein molecules removed from cilia, when reactivated using ATP, were found to transport microtubules unidirectionally. The relationship between microtubule length and velocity can be
fitted by an hyperbola. When cAMP was introduced into the assay, up to 10% of the dynein molecules became phosphorylated, and the overall microtubule translocation velocity increased by 50%. The phosphorylated dynein molecules are modelled with a shorter
force-generating phase than untreated dynein, as they are responsible for the increase in speed. The computer models
simulate both stochastic and coordinated behavior, and, together with mathematical analyses, provide a tool which can be used to
simulate the transport of microtubules of any length. In the coordinated model, dynein arms are activated sequentially, so
that activity moves across the substrate in a wave-like fashion. In the stochastic model, dynein molecules are activated randomly,
but each molecule cycles continuously after initial activation. The microtubule velocity increases hyperbolically with
microtubule length in both systems, but in the metachronal system, a sharp cut off in velocity occurs at a specific length
related to the dynein arm separation and cycle time of the system. The mean experimental dynein arm separation was 0.06
micrometers, and a range of separations including this value are used in the models. The results show that both the stochastic and
the metachronal models can fit the observed data. The spread of the data, and the biologically implausible assumptions made in
the metachronal case, suggest that the stochastic model is the appropriate representation of in vitro dynein behavior.