Use of gel chromatography for the determination of the Stokes radii of proteins in the presence and absence of detergents. A reexamination.

In the course of a routine investigation of the complex between the erythrocyte membrane protein spectrin and sodium dodecyl sulfate, we observed a large discrepancy between the true Stokes radius (178 A, measured by hydrodynamic methods) and the apparent value derived from gel chromatography (107 A). In attempting to resolve this discrepancy, we have experiments that indicate that all large asymmetric particles may be subject to a similar discrepancy; e.g., native fibrinogne has a true Strokes radius of 108 A, whereas the value derived by column chromatography after calibration with globular proteins is only 71 A. The simplest interpretation is that end-on insertion of asymmetric particles into the gel pores contributes to their retardation. The phenomenon clearly limits the usefulness of gel chromatography as a quantitative measure of the hydrodynamic Stokes radius. Incidental data obtained in the course of this work indicate that spherical viruses may have weak chemical affinity for the porous gel. Chromatography of large proteins in the presence of detergents produced no effects ascribable to absorption of the detergents, but the results suggest a need for further study of possible interaction between detergents and small gel pores.