| To better understand the precise mechanism of F-actin bundle formation in cellular environments, we have used a combination of traditional physical and biochemical techniques to measure the phase boundary between isotropic F-actin an F-actin bundles as a function of polyamine concentration. While the actin system proved to be too complicated for light scattering measurements, pelleting assays were able to reveal much information about the nature of bundle formation. To do this, we first grew actin filaments (F-actin) by adding MgCl2 to G-actin (globular actin). F-actin was then incubated with spermine or spermidine (low-binding-energy cationic linkers and actin-bundling factors) overnight, and then the samples were spun at low speeds to separate bundles from unbundled F-actin. The relative amounts of actin in the pellet and supernatant were determined via gel electrophoresis, yeilding a description of the bundling transition as a function of actin and linker concentrations. With this approach, we have mapped the phase boundary between bundled F-actin and isotropic unbundled F-actin for two F-actin/polyamine linker systems. Surprisingly, the dependence of bundle formation on actin concentration is small to non-existent. At the actin concentrations we studied (4.5, 9, 18 and 36muM), actin tends to form bundles at or above a single linker concentration, regardless of the valence of the linker. In order to understand the iteractions holding F-actin together in bundles, we used NMR to determine where the polyamines were with respect to the bundled and unbundled phases of actin. Surprisingly, the spermine and spermidine bind so tightly to the actin that they are completely immobilized and thus invisible to NMR. This work has shown that in the F-actin system, explanations of the mechanism of bundle formation involving only electrostatics are not complete. |
