Performance Characteristics of Electron Transfer Dissociation Mass Spectrometry

Abstract

We performed a large scale study of electron transfer dissociation (ETD) performance, as compared with ion trap collision-activated dissociation (CAD), for peptides ranging from ϳ1000 to 5000 Da (n ϳ 4000). These data indicate relatively little overlap in peptide identifications between the two methods (ϳ12%). ETD outperformed CAD for all charge states greater than 2; however, regardless of precursor charge a linear decrease in percent fragmentation, as a function of increasing precursor m/z, was observed with ETD fragmentation. We postulate that several precursor cation attributes, including peptide length, charge distribution, and total mass, could be relevant players. To examine these parameters unique ETDidentified peptides were sorted by length, and the ratio of amino acid residues per precursor charge (residues/ charge) was calculated. We observed excellent correlation between the ratio of residues/charge and percent fragmentation. For peptides of a given residue/charge ratio, there is no correlation between peptide mass and percent fragmentation; instead we conclude that the ratio of residues/charge is the main factor in determining a successful ETD outcome. As charge density decreases so does the probability of non-covalent interactions that can bind a newly formed c/z-type ion pair. Recently we have described a supplemental activation approach (ETcaD) to convert these non-dissociative electron transfer product ions to useful c-and z-type ions. Automated implementation of such methods should remove this apparent precursor m/z ceiling. Finally, we evaluated the role of ion density (both anionic and cationic) and reaction duration for an ETD experiment. These data indicate that the best performance is achieved when the ion trap is filled to its space charge limit with anionic reagents. In this largest scale study of ETD to date, ETD continues to show great promise to propel the field of proteomics and, for small-to medium-sized peptides, is highly complementary to ion trap CAD. Electron transfer dissociation (ETD), 1 a relatively new peptide/protein fragmentation method, holds great promise to advance the field of protein mass spectrometry (1-3). As compared with the conventional technique, collision-activated dissociation (CAD), ETD offers a more robust method to characterize post-translational modifications (PTMs) and to interrogate large peptides or even whole proteins (4 -7). Because of these attributes and the fact that it generates c-and z-type products, instead of b-and y-type, many propose that ETD is highly complementary to CAD. ETD reactions, of course, are generally conducted within the confines of ion trap mass spectrometers where sequential CAD and ETD experiments are easily performed. Most proteomics experiments, however, are coupled with on-line chromatographic separations, and analysis time, per peptide, is ideally minimized to increase dynamic range (8). Thus, to extract the most information from a given experiment, knowledge of how these two dissociation techniques complement one another ...

Analysis of the trypanosome flagellar proteome using a combined electron transfer/collisionally activated dissociation strategy

“…In a limited number of cases, these peptides show some overlap (i.e., they result from missed enzymatic cleavage sites); these remain independent recognitions of distinct peptide analytes. The findings reported here are similar to those found in a recent paper from Coon and coworkers, which described analyses performed upon Arabidopsis thaliana and Saccharomyces cerevisiae using CAD/ETD switching methods [20]. The reported extent of new peptide recognitions arising from performing ETD analysis, and the extent of overlap between peptide recognitions made via ETD and CAD was substantially lower than that found in this study (12% of peptides, with overall 0.48% FDR versus 28% at 1% FDR in the current analysis).…”

“…specific fragmentations, with a far greater diversity of detectable fragmentation pathways in comparison to CAD. Similar trends in the fragmentation efficiency of multiply charged peptidic precursors by have been reported previously by others [20,26]. The net solution-phase charge of the peptides has been investigated and indicates that, as expected, basic peptides are more predominant in the set of peptides identified using ETD than in those identified by CAD (see Supplementary Figure 2).…”

“…The distribution of peptide length for ETDrecognized peptide analytes is skewed somewhat towards longer peptides than that of analytes identified by CAD (mean 15 residues for ETD versus 14 for CAD). Similarly, peptides recognized by ETD are more basic in nature than those recognized by CAD ( Supplementary Figures 1 and 2), showing similar trends to the published Arabidopsis and yeast data [20]. A similar strategy employing normalization of FDR for ETD and CAD data was recently reported by Pandey and coworkers [27], in their case employing a different search engine upon a proprietary mixture of 50 proteins.…”

“…The results of the database search, using varying FDR are summarized in Figure 2, which illustrates the number of (nonredundant) peptides matched at a FDR of either 1% (Figure 2a) or 5% (Figure 2b). The FDR was determined by examination of the number of hits matching to "decoy" sequences within a concatenated forward-reverse database search [20]. A sub-set of the data was subjected to database searching following filtering to remove m/z values consistent with the presence of residual and charge-reduced precursor ion species; this additional processing step was found to have no significant influence upon the resultant peptide recognitions made (data not shown).…”

“…The efficiency of the ETD process may be optimized by adjustment of the precursor ion population and the interaction time with the reagent anion [17,19]. A recent article indicating the complementary value of ETD to conventional CAD analysis was published by Good et al during preparation of this study, and will be discussed in further detail below [20].…”