Proteomic analysis of human plasma: failure of centrifugal ultrafiltration to remove albumin and other high molecular weight proteins.

Determination of specific low abundance proteins, usually by radiolabelled or enzyme-linked immunoassays in serum or plasma is widely used in diagnostic medicine. Substitution of these assays by a proteomic approach has been suggested, but this methodology has far from realised its potential as a diagnostic tool. The main protein fractions of plasma represent more than 80% of total protein, making the hundreds or even thousands of other proteins difficult to detect by two-dimensional electrophoresis (2-DE). Thus, loading sufficient sample to detect trace proteins invariably means excessive loading of albumin and other high abundance proteins. The aim of this study was to determine whether centrifugal ultrafiltration of whole plasma could be used to eliminate proteins exceeding a desired molecular weight cut-off. Cellulose filters with a 30 kDa molecular weight cut-off were used with whole plasma, and total protein was determined before and after ultrafiltration. Samples were processed by routine methods for 2-DE using 18 cm, pH 3-10 isoelectric focusing strips for the first dimension and 7-15% gradient gels for the second dimension followed by silver staining. Gel analysis of the retentate fraction (> 30 kDa) revealed a typical 2-DE plasma profile with most of the major landmark proteins in place and as expected, the gels lacked many of the smaller (< 30 kDa) proteins. Comparison with gels of the filtrate fraction (> 30 kDa) revealed very little difference. Not only were many of the higher molecular weight proteins still present, but some of the smaller < 30 kDa landmark proteins were absent. Overall, gels of both the retentate and filtrate fractions were less informative than gels of whole plasma.