Analyzing chromatin remodeling complexes using shotgun proteomics and normalized spectral abundance factors

Abstract

Mass spectrometry-based approaches are commonly used to identify proteins from multiprotein complexes, typically with the goal of identifying new complex members or identifying post-translational modifications. However, with the recent demonstration that spectral counting is a powerful quantitative proteomic approach, the analysis of multiprotein complexes by mass spectrometry can be reconsidered in certain cases. Using the chromatography-based approach named multidimensional protein identification technology, multiprotein complexes may be analyzed quantitatively using the normalized spectral abundance factor that allows comparison of multiple independent analyses of samples. This study describes an approach to visualize multiprotein complex datasets that provides structure function information that is superior to tabular lists of data. In this method review, we describe a reanalysis of the Rpd3/Sin3 small and large histone deacetylase complexes previously described in a tabular form to demonstrate the normalized spectral abundance factor approach.

Figure 1. Purification of Rpd3/Sin3 Large and Small Complexes

Whole cell extracts (section 2.1) from 6L of the RPD3-TAP strain culture were subjected to Ig-Sepharose followed by TEV elution (section 2.2). TEV eluates were fractionated on MonoQ ion exchange chromatography to separate the small and large Rpd3/Sin3 complexes (section 2.3). A. MonoQ fractions 26 through 30 corresponding to Rpd3/Sin3L (as defined by Superose 6 gel filtration, [10]) were pooled and further subjected to Calmodulin-sepharose. EGTA elutions 1 through 10 were resolved on 8% SDS-PAGE followed by silver staining. Bands known to correspond to different subunits are shown. B. MonoQ fractions 19 through 23 containing Rpd3/Sin3S were pooled and further subjected to Calmodulin-sepharose. EGTA eluates 1 through 10 were resolved on 8% SDS-PAGE followed by silver staining. Bands known to correspond to different subunits are shown.

Figure 2. Base Peak Chromatograms from a MudPIT Analysis of Rpd3/Sin3L

A peptide mixture generated from Rpd3/Sin3L (section 3.1) obtained from affinity and MonoQ purifications of Sin3-TAP was resolved on a triphasic microcapillary column (section 3.2) using the six step multidimentional gradients described in section 3.3.

Figure 3. Summary Percentile Statistics of NSAF-Based Ranks for Rpd3/Sin3 Subunits

Specific proteins detected within a particular run were ranked based on their NSAF values calculated for each of the six analyses reported here (section 4.3, eq. 2). The distribution of these ranks for the 14 proteins belonging to Rpd3/Sin3 complexes was plotted as a box plot representation, where the 25^th and 75^th percentiles are represented by the upper and lower boundaries of the box, the median being the line dissecting the box, and the mean being the small square in each box. The 5^th and 95^th percentiles are shown with lines attached to the box, the ‘X’ represents the 1^st and 99^th percentiles, and the stand alone ‘–’ represents the complete range. The number (n) of runs in which each protein was detected is shown within arrows below the graph.

Figure 4. Spectral Abundance Factors Normalized against Subunits of Rpd3/Sin3 Complexes

NcSAFs values for each of the 14 proteins belonging to Rpd3/Sin3 complexes were calculated (section 4.3, eq. 3) for whole TAP preparations (A), and large (B) and small (C) complex fractionations of Rpd3-TAP (black bars) and Sin3-TAP (gray bars) pull-downs.