The ultracentrifugal analysis of normal and pathological serum fractions.

IN this paper, 'which is the concluding one of the series, we are concerned with the fractionation, by means of a neutral electrolyte and electrodialysis, of a few sera which have already been investigated in the ultracentrifuge [McFarlane, 1935, 1, 2, 3]. The particular objects in view have been (1) to investigate the composition of the globulin fractions obtained by a single precipitation with ammonium sulphate; (2) to compare the analytical results of the fractionation procedures with the results of ultracentrifugal analysis of the same serum; and (3) to discover in the case of certain sera whether new molecular types which have been shown to be present are precipitated or remain in solution on half saturation with ammonium sulphate. For several reasons we have relied on ammonium sulphate as protein precipitant. It was advisable to use only one neutral electrolyte in order to be able to compare the results for one serum with those for another. Ammonium sulphate has been the most widely used electrolyte in the past in the fractionation of sera, although sodium sulphate is becoming increasingly used as a result of the work of Howe [1923]. It may be noted, however, that Howe finds the values for the albumin and globulin concentrations in cow sera obtained by a single precipitation with ammonium sulphate to agree with those obtained with sodium sulphate. Wu [1922] obtains the same analytical results with saturated magnesium sulphate as with half-saturated ammonium sulphate. Cullen and Van Slyke [1920] and S0rensen [1925; 1930] and others have used ammonium sulphate extensively in fractionating serum proteins. It has been show-n [1935, 1] that when horse serum albumin and globulin, prepared by ammonium sulphate fractionation, are mixed together in concentrated solution and the mixture is analysed in the ultracentrifuge, the concentration of the albumin fraction is found to be abnormally high and that of the globulin fraction correspondingly low. These proteins exist in concentrated solution in a form of equilibrium with each other. The same has been shown indirectly to be true for cow and human serum proteins by ultracentrifugal analysis of the concentrated and diluted sera. Since we know very little about the action of ammonium sulphate as a protein precipitant and in view of this difference in the state of the proteins in concentrated and diluted sera, it was decided in the first place to investigate the composition of the globulin precipitates obtained by half saturating a concentrated and a diluted cow serum with ammonium sulphate. It is not proposed to repeat here the numerous details of the technique of an ultracentrifugal experiment which will be found in the earlier papers referred to. In the following tables we show the concentrations of the various protein fractions in g./100 ml. calculated from measurements of protein refraction in-