Taking aim at shotgun phosphoproteomics

Abstract

Shotgun phosphoproteomics employs liquid chromatography-coupled tandem mass spectrometry (LC-MS/MS) to analyze phosphopeptides from complex protein mixtures, allowing detection and quantification of phosphorylation events on a global scale. Within the past few years, this powerful technique has been used to uncover novel phosphorylation sites as well as explore changes in protein phosphorylation during cellular signaling in various systems. This review presents a general overview of current phosphoproteomic methodologies, including summaries of various approaches to phosphopeptide enrichment, alternative MS fragmentation strategies, and the latest software for analysis of phosphopeptide data sets.

Figure 1

A standard protocol for LC-MS/MS-based phosphoproteomic analysis. Proteins are digested with a site-specific protease, usually trypsin. Phosphorylated peptides are enriched from non-phosphorylated peptides using one or more of the approaches listed. Phosphopeptide samples are then separated on a reversed-phase liquid chromatography (RP-HPLC) column which is connected on-line to the mass spectrometer. Full-length peptide (parent) ions are measured in the initial scan (MS¹) and selected individually for fragmentation by collision-induced dissociation (CID). This process often results in “neutral loss” of phosphoric acid (98 Da; H₃PO₄) from the parent ion and very little backbone fragmentation is detectable in the MS² spectrum. The neutral loss peak can be selected for further CID fragmentation and detection in the MS³ spectrum to increase the overall phosphopeptide identification rate or to confirm poor quality MS² identifications.

Figure 2

Effect of sample preparation on the distribution of phosphoprotein identifications. Renal collecting duct membrane protein pellets were resuspended in either 50mM ammonium bicarbonate (NH₄HCO₃), 6M guanidine, or 1% SDS prior to trypsinization and analysis by LC-MS/MS.

Figure 3

Comparison of the quality of MS² and MS³ spectra for a typical phosphopeptide. A. Amino acid sequence of a phosphopeptide identified as water channel aquaporin-2 (AQP2) with a single phosphorylated residue, Ser-256 (*). The region that is highlighted in red includes 5 b and 5 y fragment ions that represent so-called “site-determining ions” that could distinguish the actual site of phosphorylation (i.e. Ser-256) from another potential site, Ser-261. B. MS² spectrum of the peptide shown in A. The blue neutral loss peak (NL) is predominant while the number and abundance of b and y ions, particularly the site-determining ions (red arrows) is low. Neutral loss peak with loss of 1 water (NL - H₂O). C. Higher quality MS³ spectrum generated by fragmentation of the neutral loss peak shown in B. There is an increase in the number and abundance of b and y ions relative to the MS² spectrum.

Figure 4

Analyzing phosphopeptide data sets in the post-acquisition phase. After MS spectra are acquired, the appropriate protein sequence database must be searched using specific criteria in order to generate a list of potential peptide matches. By using various filtering parameters based on a predetermined false discovery rate (FDR), only the highest quality spectra corresponding to predicted phosphopeptides are selected from this list. At this point both MS² and MS³ data sets can be combined and analyzed to determine the total number of phosphoproteins identified as well as the exact site(s) of phosphorylation on each corresponding peptide. Assigning phosphorylation sites can be accomplished using computer programs (automated) or by personal examination of spectra (manual).