A Method of Scanning Paper Electrophoresis Strips and Its Application to the Study of Plasma Proteins*

Although paper electrophoresis has many distinct advantages over the free boundary method, the most important being that a very small amount of material is required, there is difficulty in the quantitative determination of the concentrations of the separated components. The most commonly used method of detection at the present time involves staining the protein with a dye such as bromphenol blue or amido black, and then either scanning with a densitometer (1, 2) or else cutting out the stained areas and eluting them for determination in a spectrophotometer. Although they provide quantitative results, these methods have several shortcomings. First, they require considerable time; second, there is no selectivity, so that if two components of a mixture have the same or nearly the same mobility, they are not separable; and third, the optical methods have limited resolution. In 1955 Sharpsteer? discovered that paper electrophoresis strips of blood plasma could be scanned with a movable electrode (a fixed electrode being maintained at one end of the wet paper strip) and a voltage record of the separated components obtained. This very significant observation seemed to offer, even in its crudest form, the possibility of a higher resolution of components in such a protein mixture than had heretofore been achieved by optical methods. The voltages obtained by scanning the strip in the manner indicated were later shown to arise from electrochemical reactions, i.e. the paper strip is essentially a series of electrolyte concentration cells, the variable voltage arising from gradients in the concentration of hydrogen and buffer ions.2 However, because of instrumental difficulties (e.g. poor signal to noise ratio, inability to obtain a well defined electrode reaction because of oxide coating on the platinum electrode, etc.) quantitative results could not be obtained to test the validity of the method in order to develop a scanning procedure. Since proteins bind various metal ions it was suggested3 that the combination of a metal ion to the protein on the paper strip, followed by scanning with an electrode of the same metal, might provide a well defined reversible electrode and might also overcome the problem of a poor signal to noise ratio. Sharpstee& carried out such experiments using silver and blood plasma in a procedure involving photochemical reduction of silver ion to metallic silver, before scanning, and obtained a significant improvement in the appearance of the scanning records. Having been apprised of this initial success, we proceeded with a sys-