On the Aggregation of Bovine Serum Albumin in 0.1 M Salt Solutions

Abstract

The molecular weight of bovine serum albumin at neutral p H in 0.1 M salt solutions was determined with light scattering measurements. Simultaneously, preferential binding of a salt to the protein was calculated. It was found that the protein aggregates in KI and KSCN, but not in KCI, LiCI, and LiBr solutions. Analysis of the sedimentation behavior has confirmed aggregation as a molecular phenomenon, not an artifact.Le poids molCculaire de serum d'albumine de boeuf a etC determine par diffusion de la lumiere a p H neutre dans des solutions contenant 0.1 M de sel. Simultanement on a calcult I'attachement preferentiel d'un sel a la proteine. On a trouve que la proteine forme des aggrkgats dans ie K I et e KSCN mais pas des solutions de KCI, LiCl et LiBr. L'analyse des patrons de sedimentation a confirme que I'aggregation est un phtnomene moleculaire et non un artefact.[Traduit par le journal]Can. J. Chem. 51, 3781 (1973) Bovine serum album~n aggregates in 0.1 M and the refractive index increments are measured on solu- transition, albumin exists in a compact form, ponent systkm (water-1, protein-2. and salt-3), the plot albumin has been assumed to be a nin-associat: ing protein. For example, the studies of the binding of C1-, I-, and SCN-to albumin (4-6) all ignored the aggregation. In this paper we shall demonstrate that at neutral p H (ca. 5.2) albumin also forms aggregates in the presence of some though not all salts. The molecular weight of albumin at neutral pH, ionic strength 0.10 and room temperature was determined and the preferential binding of each salt to the protein was compared, using the technique of light scattering. Supplementary information about aggregation was obtained with sedimentation measurements. Experimental MaterialsArmour crystalline bovine serum albumin, Lot Nos. A 69908, F 71601, and D 71209, was deionized by Dintzis' method (7). The dimer content varied from 1% to 5% depending upon the lot as judged from sedimentation velocity analyses. N o significant differences were detected in light scattering experiments whichever lot was used.The salts used, KCI, KI, and KSCN were Baker analyzed reagents, while LiCl and LiBr were Matheson, Coleman and Bell reagent grade. Glass-distilled water was used in all the experiments. Light ScatteringAdvantage was taken on the theory of light scattering in multi-component systems that if the light scattering of HC~IAT vs. C2 would give the true molecu'l.ar weight M2, H being the Debye parameter, C2 the concentration in g/ml and AT theexcess turbidity of thesolution over the solvent. If the solutions are not at constant chemical potential but at constant concentration of the solvent system, then the plot of HC2/Ar us. C2 can only give M2', the apparent molecular weight of component 2 (protein), not the true molecular weight. Nevertheless the relation between M2 and M 2 ' enables one t o calculate a quantity (ag3/agZ)7,,,,,,,, which is the preferential binding of the number of grams of the salt to one gram of the protein on molar or molal basis, g being t...