Differential proteomics based on 18O labeling to determine the cyclin dependent kinase 9 interactome

Although enzyme catalyzed 18O labeling has been used as a tool in quantitative proteomics, this type of labeling has not yielded the same impact yet as alternative techniques for quantitation like SILAC or labeling with chemical mass tags. The practical difficulties involved in 18O labeling, most importantly the occurrence of incomplete labeling and, as a result, the difficulties in data analysis and interpretation have hampered its implementation in high-throughput comparative proteomics protocols. In this paper, we have optimized the 18O labeling procedure to such an extent that complete labeling can be achieved in a routine manner. We have implemented this approach into a protein-protein interaction analysis pipeline to differentiate between bona fide interaction partners of the low-level expressing cell cycle regulator cyclin-dependent kinase 9 (Cdk9) and nonspecifically binding or background proteins. Previously known as well as novel interaction partners of Cdk9 were found, among which most notably the Mediator complex and several other proteins involved in transcriptional regulation. We show here that a differential proteomics approach based on 18O labeling provides a valuable method for high-confidence determination of protein interaction partners and is easily implemented in protein network analysis workflows.