Spot overlapping in two-dimensional maps: a serious problem ignored for much too long

Abstract

In the analysis of a neuroblastoma xenograft implanted in mice using two-dimensional maps, some 85 proteins were found to be up- or down-regulated (out of a total of 264 detected by a medium-sensitivity colloidal Coomassie stain). When these spots were eluted and analysed by mass spectrometry in a quadrupole time of flight mass spectrometer, a number of spots were found to be envelopes of different polypeptide chains. Out of a total of 74 proteins identified, 52 (71%) were found to be singlets, 14 (19%) were doublets, 6 (8%) were triplets, 1 was a quadruplet and 1 a quintuplet. Analysis of the DeltapI and DeltaMr of all species contained in a single gel segment eluted helped point out potential errors in protein identification. This was a unique case, in that very minute bioptic sample loads were applied to the gel. In normal cases, where sample loads of ca. 1 mg of total protein are applied and typically at least 1000 spots are visualised, the singlets will be the minority, rarely exceeding 30% of all spots analysed. The experimental data on the abundance of overlapping spots were in excellent agreement with theoretical data calculated on the basis of the statistical theory of spot overlapping, originally proposed by Davis and further developed by some of the authors. Ways and means for minimizing spot overlap and visualising a greater number of spots in a two-dimensional map are discussed.