Comparisons of different peptide fractionation methods for in-depth phosphoproteomic profiling

Abstract

Phosphorylation which is one of the most common post-translational modification (PTM) is important and widespread occurrence on biological cellular processes, but the identification of sites of protein phosphorylation is still a challenge even when performed on highly purified protein. To overcome the limit of phosphoproteome study, technics for phosphopeptide enrichment with fractionation methods have been advanced in field of phosphoproteomics. In this study, we sequentially introduced batch mode protocol for phosphopeptide enrichment and LC-based fractionation methods for phosphoproteome study. The selectivity of our batch mode TiO2 enrichment has approximately range of 80% to 89% from sub-milligram to milligram scale of mouse brain tissue lysate. With optimizing TiO2 enrichment, three kinds of different pH (low pH, medium pH, high pH) LC-based off-line fractionation were employed. Low pH reverse phase fractionation is superior to other pH conditions for phosphoproteomic profiling. Low pH reverse phase fractionation identified higher number of phosphopeptides (4,151), and phosphoproteins (1,604) than medium pH (4,061/1,520), and high pH (2,001/862) fractionation, but the enrichment efficiency (~70%), and the separation of unique phosphopeptides per fraction (1 fraction by 79%, 76%, 80%, ~2 fractions by 94.4%, 92.5%, 93.5%) had similar pattern from the starting material of 1 milligram mouse brain tissue lysate. We also obtained the trend of 2D-orthogonality meaning the unique effective area of 2D separation space covered by the eluting peaks. Theoretically, high pH fractionation was predicted to shown optimal broaden 2D separation space covered eluting peaks (RP-RP), but it was not. Interestingly, Low pH fractionation has an aspect that most of the phosphopeptides are loaded to C18 resin identifying just six phosphopeptides on flow-through fraction (0.1%), but, high pH fractionation is considerably eluted earlier on flow-through fraction (41.2%) meaning that most of the phosphopeptides are not strongly binding to C18 resin, or weakly binding to C18 resin. Medium pH fractionation shows the pattern that not only considerable number of phosphopeptides are eluted on flow-through fraction (16 %), but also most of phosphopeptides are eluted on each concatenated fractions. It is slightly inferior to low pH fractionation according to the concept of well separated, and more identifying the number of phosphopeptides. Our data shows the linear trend of retention time shifts that phosphopeptides elute earlier on high pH condition as decreasing to medium pH, and low pH condition on LC-based off-line fractionation, and specifically high pH fractionation method can cause potential sample losses of phosphopeptides in LC instrument comprised of metal components. We conclude that the low pH fractionation method is better for phosphoproteomic profiling than medium pH fractionation method, and high pH fractionation method.