Chain Dynamics, Mesh Size, and Diffusive Transport in Networks of Polymerized Actin: A Quasielastic Light Scattering and Microfluorescence Study

Dynamical (chain excitations, reptational diffusion) and structural (mesh size) properties of semidilute solutions (gels) of polymerized actin and their concentration dependencies were studied by quasielastic light scattering (QELS) and microfluorescence experiments. By QELS we could measure the internal dynamics of single chains. The fact that only internal single-chain dynamics are observed by QELS is a consequence of the inverse of the scattering vector q (6 X 10⁴cm^-1< q < 3 X 10⁵cm^-1) being small compared to both the average contour length of the actin filaments (>30 μm), which was estimated from reptational diffusion, and the average mesh size ξ of the network. That QELS measures internal dynamics of single chains is also shown by the insensitivity of the measured dynamic structure factor to cross-linking by a-actinin. For the range of actin concentrations, 0.08 mg/mL < c_a< 0.37 mg/mL, and scattering angles, 20° < 0 < 150°, studied in this work, the dynamic structure factor S(q,t) of the polymer chains decays like S(q,t) <« exp(-Γ_q⁽⁰⁾£) at short times (Γ_q^(o)t « 1). At long times (Γ_q⁽⁰⁾t » 1) it follows the relation S(q,t) « exp(-(r_a⁽⁰⁾t)^2/3). The initial decay rate r_q⁽⁰' exhibits a power law of the form r_q⁽⁰⁾. a 2.760.1 _cn_nfilaments thus approximate the universal behavior predicted for a Rouse-Zimm chain by Dubois-Violette and de Gennes (Physics 1967, 3, 181). The deviation from the predicted q³dependence of the initial decay rate is similar to that found for synthetic polymers by several groups. Concentration-dependent deviations from Rouse-Zimm behavior, which are most prominent for r_q⁽⁰⁾£ » 1, are attributed to filament interactions. Translational diffusion coefficients of fully polymerized actin filaments were measured by fluorescence photobleaching (FRAP). We obtained a diffusion coefficient of Dtran8,----- 8 x 10^-11cm²/s for c_a= 1 mg/mL. Da scaled roughly like D<x c_a^_1£²as predicted by the reptation model for flexible chains. The average mesh size of actin networks as a function of actin concentration, c_a, was determined by measuring the translational diffusion coefficient, D(c_a,d), of latex spheres of various diameters, d, by both FRAP and microscopic videotracking (Perrin technique). Scaling laws were applied to relate first D(c_a,d) to the ratio d/£ and second the mesh size £ to the actin concentration c_a. D(c_a,d) scales as D(c_a,d) = D(0,d) explcddc/]⁵) with v »^l/_2i5^2, and a « 1 as hypothesized by de Gennes et al. for a non-cross-linked network of rodlike polymers. The Rouse-Zimm-like behavior of actin filaments, as observed by QELS, contrasts with the appearance of actin as a semiflexible polymer in microscopic end-to-end distance measurements, which have been reported in the literature. This scale-dependent difference in apparent flexibility is suggested to be a consequence of the complex structure of actin filaments as a “polymer of polymers”. We hypothesize that small-scale dynamics are dominated by stretching elasticity, i.e., by Rouse-Zimm-like excitations of filament subunits, whereas large-scale dynamics are determined by the bending elasticity of the filament. The crossover length between the two regimes would have to be of the order of micrometers. We point out the usefulness of actin gels as model systems to study fundamental properties of polymers on a mesoscopic scale with coil radii in the 100-/um regime. On this scale internal motions can be easily studied by quasielastic light scattering instead of neutron scattering.