Quantitative Phosphoproteomics of CXCL12 (SDF-1) Signaling
Figure 4
Comparison of CXCL12-responsive phosphopeptides and corresponding genes to published phosphoproteomic and manually curated datasets.
Genes from various phosphoproteomic datasets and manually curated databases were compared to genes from CXCL12-responsive phosphosites using contingency tables (see methods). Fold-enrichments are listed at the base of each bar. Benjamini and Hochberg corrected p-values (Fisher's exact test) are plotted with p<0.05 considered statistically significant (solid line). (A) Genes from CD3-responsive phosphosites (TCR) [41], epidermal growth factor (EGF)-responsive phosphosites [42], and mitotic phosphosites [68] were compiled as published for the indicated time points. TCR (15 min) and EGF (5 min) are the most statistically significant, suggesting crosstalk between pathways. Unlike cancer-related genes [69], only HIV ‘interactors’ (http://www.ncbi.nlm.nih.gov/RefSeq/HIVInteractions) and mammalian target of rapamycin (mTOR) signaling genes [57] were statistically significant. (B) Phosphopeptides and corresponding genes of G protein-dependent (Ang II) and independent (SII-Ang II) signaling (3 min) were reanalyzed to reflect our experimental conditions and analysis criteria (see methods) [44]. Only the overlap of Ang II-responsive phosphosites and corresponding genes were statistically significant.