Automated Quantitative Phosphopeptide Enrichment using Titania-packed Capillary Columns and Its Application to Human Tissues for Label-Free Phosphoproteomics

摘要

Phosphorylation is a reversible post-translational modification affecting both the folding and function of proteins. Due to the sub-stoichiometric nature of protein phosphorylation, an enrichment stop is often performed prior to mass spectrometric analysis. The most commonly used enrichment technique, immobilized metal affinity chromatography is incompatible with many reagents commonly used in biological experiments, including alkaline metal salts and EDTA. For this reason, there has been a rapid increase in the used of TiO_(2) materials to enrich for phosphorylated species. Our laboratory previously developed a enrichment protocol using TiO_(2) particles in a spin column format (Soderblom, 2011). While this approach has been demonstrated to be both quantitative and reproducible, it is a very manual process. A transition to an automated setup was needed to handle the increase in study size and potentially improve the reproducibility. This required the transition from a spin column to a capillary column. Based on the information gained from the matrix experiment performed using the spin column, we were able to target a few variable to more efficiently evaluate the capillary column. Similar qualitative identifications were made with both setups, while the specificity and reproducibility were improved in the automated enrichments. From the matrix experiments performed previously, 80percent specificity was achieved for both qualitative and quantitative analyses. Regarding quantitative reproducibility, the average (percent)CV for all phosphopeptides was 17.2percent with 80percent of phosphopeptides having variations below 25.4percent. Using the automated, capillary column setup, the specificity was slightly higher at 84percent. The main difference between the two methods is in the quantitative reproducibility. The average and median (percent)CV (10.7 and 8.1percent, respectively) in the reproducibility study using the automated setup were approximately two times lower than that achieved using the spin column. This technique was applied to the analysis of left ventricle cores taken from seven patients during Left Ventricular Assisted Device (LVAD) implant surgery and seven left ventricle cores taken from the same patients during heart transplant surgery. The total amount of protein enriched from each sample was 1 mg. A total of 637 phosphopeptides, corresponding to 300 phosphoproteins, were identified across the entire dataset. Using each patient as their own control, only 31 phosphopeptides were significant in at least 3 patients at a fold change greater than 2 and p-value less than 0.005.