Myosin isoform determines the conformational dynamics and cooperativity of actin filaments in the strongly bound actomyosin complex

Abstract

We used transient phosphorescence anisotropy to detect the microsecond rotational dynamics of erythrosin-iodoacetamide-labeled actin strongly bound to single-headed fragments of muscle myosin subfragment 1 (S1) and non-muscle myosin V (MV). The conformational dynamics of actin filaments in solution are markedly influenced by the isoform of bound myosin. Both myosins increase the final anisotropy of actin at substoichiometric binding densities, indicating long-range, non-nearest neighbor cooperative restriction of filament rotational dynamics amplitude, but the cooperative unit is larger with MV than with muscle S1. Both myosin isoforms also cooperatively affect the actin filament rotational correlation time, but with opposite effects: muscle S1 decreases rates of intrafilament torsional motion, while binding of MV increases the rates of motion. The cooperative effects on the rates of intrafilament motions correlate with the kinetics of myosin binding to actin filaments such that MV binds more rapidly and muscle myosin binds more slowly to partially decorated filaments than to bare filaments. The two isoforms also differ in their effects on the phosphorescence lifetime of the actin-bound erythrosin iodoacetamide: while muscle S1 increases the lifetime, suggesting decreased aqueous exposure of the probe, MV does not induce a significant change. We conclude that the dynamics and structure of actin in the strongly bound actomyosin complex are determined by the isoform of the bound myosin in a manner likely to accommodate the diverse functional roles of actomyosin in muscle and non-muscle cells.

Fig. 1

The effect of muscle S1 and MV on TPA decays of actin in the absence of nucleotide. S1/A=1.2, KMg50, 0.5 U/ml apyrase. Smooth lines represent fits to the sum of two exponential terms (Eq. 2).

Fig. 2

The effect of increasing fraction of strongly bound muscle S1 and MV on the final anisotropy of actin.

Fig. 3

The effect of increasing fraction of strongly bound muscle S1 and MV on the average correlation time 〈ϕ〉 of actin.

Fig. 4

The effect of strongly bound muscle S1 and MV on the average phosphorescence lifetime of actin-conjugated ErIA. T he 〈τ〉 in the presence of muscle S1 is fitted to the linear lattice model Eq. 1, and 〈τ〉 in the presence of MV is represented by the linear fit for visualization.

Fig. 5

The effect of the pre-bound S1 binding density on the relative k_on of myosin binding to pyrene actin for muscle S1 (squares) or MV in the absence of nucleotide (filled circles) or MV in the presence of 20 µM MgADP (empty circles). Data points represent fits to averages of 2–4 individual time courses at indicated pre-bound S1 density. The concentration dependence is fitted to the linear lattice model (Eq. 1). Uncertainties represent errors in the fit.

Fig. 6

Structural alignment of the proposed regions of interaction of muscle S1 (MS1) and MV with actin. Numbers in parentheses show charges of the peptides calculated using Innovagen’s Peptide Property Calculator program.