A combinatorial approach to studying protein complex composition by employing size-exclusion chromatography and proteome analysis

The genome sequences of numerous organisms are available now, but gene sequences alone do not provide sufficient information to accurately deduce protein functions. Protein function is largely dependent on the association of multiple polypeptide chains into large structures with interacting subunits that regulate and support each other. Therefore, the mapping of protein interaction networks in a physiological context is conducive to deciphering protein functions, including those of hypothetical proteins. Although several high-throughput methods to globally identify protein interactions have been reported in recent years, these approaches often have a high rate of nonspecific or artificial interactions detected. For instance, the fraction of false positives of the protein interactions identified by yeast two-hybrid assay has been predicted to be of the order of 50%. We have developed a strategy to globally map Bacillus subtilis protein-protein interactions in a physiological context by fractionating the cell lysates using size-exclusion chromatography (SEC), followed by proteome analysis. Components of both known and unknown protein complexes, multisubunits and multiproteins, have been identified using this strategy. In one case, the partners of the B. subtilis protein complex have been coexpressed in Escherichia coli, and the formation of the overexpressed protein complex has been further confirmed by a pull-down assay.