Cooperative Interactions Between Myosin II and Actin Cross-Linking Proteins to Actin Filaments

Tianzhi Luo,Douglas N. Robinson
DOI: https://doi.org/10.1016/j.bpj.2009.12.112
IF: 3.4
2010-01-01
Biophysical Journal
Abstract:Heavy meromyosin (HMM) of both muscle myosin II and non-muscle myosin II cooperatively binds to actin filaments in a manner that depends on divalent ion composition and concentration or the transition state of the myosin motor domain. Therefore, the cooperative binding may be due to the conformational changes in F-actin induced by two-headed HMM that facilitate the binding of other heads to nearby regions in the same actin filament. Recently, we found that myosin II and an actin crosslinker cortexillin I cooperatively accumulate to highly deformed regions in Dictyostelium cells and the accumulation extent increases with increasing forces. One possible mechanism of the cooperative accumulation is that the binding of myosin to actin enhances cortexillin binding to actin filaments. We suspect this kind of cooperative binding might also exist among other actin crosslinkers. We are testing this hypothesis by using novel designed in vitro systems and advanced molecular simulation schemes. We are experimentally investigating this mechanism by measuring the binding lifetimes of myosin II and actin crosslinkers to actin and their accumulations at different load conditions in reconstituted actin cytoskeletons. One of the reconstituted systems is a bilayer vesicle that has an actin cortex anchored to its inner lipid membrane. Another system is an actin meshwork crosslinked by various actin crosslinkers and stiffened by myosin II. It is positioned in a motor-driven stretcher, allowing for in situ TIRF microscopic measurements at different strains and different strain rates. In addition, we simulate the corresponding two-dimensional reaction-diffusion problems using coarse-grained kinetic Monte Carlo simulation. In our simulations, the diffusion coefficients vary in the range of 0.01∼100 μm2/s and the forward and backward characteristic time of the binding reactions is in the range of 100 s and 0.01 s.
What problem does this paper attempt to address?