(11b) Quantitative Phosphoproteomics and in-Depth Protein Profiling of Melanoma Cell Lines

Analysis of kinase signaling processes in diseases such as melanoma requires quantitative information on the phosphorylation sites of a large number of proteins. In addition, the proteins themselves must be quantified, in order to distinguish changes in stoichiometry vs changes in protein levels (for instance, from relocalization between cell compartments). Methods involving enrichment of phosphopeptides are difficult to reproduce and do not capture all phosphopeptides. Instead, we have developed a shotgun proteomics protocol exploiting scanning for phosphate loss during negative ion MS/MS in an ABI 4000QTrap for phosphopeptides (-79 scanning). This is combined with parallel profiling of the proteins on a ThermoFisher LTQ/Orbitrap. We are using this approach to analyze MAPK kinase signaling in melanoma cell lines; this pathway is constitutively activated in ~75% of melanomas. Our current studies compare protein profiles and phosphopeptide features in a metastatic cell line treated with MAPK kinase inhibitor vs the control cells, extracting the soluble proteins and resolving them on monoQ into ~50 fractions. New computational methods based on OpenMS were developed to match features between individual LC/MS analyses of each fraction, allowing quantification of the phosphopeptide ion intensities in the negative ion QTrap scans. The number of phosphopeptide features detected in the QTrap (~5000) was similar to the number of proteins identified in the Orbitrap profiles (~4,300 proteins, with evidence of saturation sampling). These numbers suggest that we are likely sampling almost all the detectable phosphorylation sites, because they are consistent with estimates that ~1/3 of proteins are phosphorylated and that the average phosphoprotein has about 3 phophorylation sites. Of the QTrap detected phosphopeptide features, ~6% changed intensity upon treatment of the cells with MAPK kinase inhibitor. MS/MS and MS3 data collected in the same experiments provides sufficient information to identify some of these peptides, and many of these are consistent with known MAPK kinase pathway targets. We are now developing methods to target the remaining unidentified phosphopeptide features.