An informatic framework for decoding protein complexes by top-down mass spectrometry

Abstract

Efforts to map the human protein interactome have resulted in information about thousands of multi-protein assemblies housed in public repositories, but the molecular characterization and stoichiometry of their protein subunits remains largely unknown. Here, we report a computational search strategy that supports hierarchical top-down analysis for precise identification and scoring of multi-proteoform complexes by native mass spectrometry.

Figure 1

Computational platform and workflow for characterization of human multi-proteoform complexes (MPCs). In step 1, two databases are created, the “CORUM-Proteoform” database, (which contains Swiss-Prot entries also present in the CORUM database that are combinatorially expanded into candidate proteoforms), and the CORUM-MPC (which contains candidate MPCs from all subunit combinations in CORUM and their known isoforms contained within Swiss-Prot). In step 2 of the workflow, a proteoform is retrieved using the mass values from MS² (subunit) and MS³ (backbone fragment ions) by searching the CORUM-Proteoform database. In step 3, the identified proteoform and the MS¹ (intact complex) mass value are used to search against the CORUM-MPC database and generate a candidate MPC list. In step 4, a MPC-score is calculated for each member of the candidate MPC list by incorporating MS¹ intact mass information and the quality of proteoform characterization for the subunit ejected from the complex.

Figure 2

Characterization of toyocamycin nitrile hydratase (TNH), a hexameric multi-proteoform complex purified to homogeneity. The 3-tiered approach to tandem top-down mass spectrometry for the TNH complex is illustrated with the (a) MS¹, (b) MS² and (c) MS³ fragment ion maps of each of the ejected proteoforms. Masses are reported as the average +/− the S.D. of the mass measured from each of the most abundant charge states. (d) The identity of a specific multi-proteoform complex (α₂β₂γ₂) obtained from database searching that combines information from MS¹, MS², and three MS³ spectra is shown below the subunit graphical fragment maps.

Figure 3

A summary of the MPCs identified in this study. The complex name, intact mass, and number of observed MPCs is noted next to each complex, with modifications and endogenous cleavages on subunits specified in the key in the upper right. (a) Additional combinatorial MPCs were detected for pyruvate kinase (see Supplementary Fig. 4 and the Supplementary Discussion for greater detail). (b) Broad charge states in the case of ferritin indicate a distribution of bound iron (in the thousands) and therefore the possibility of additional MPCs lying beneath this molecular polydispersity. (c) The overall stoichiometry of the human 20S proteasome is known to be (α-1)₂(α-2)₂(α-3)₂(α-4)₂(α-5)₂(α-6)₂(α-7)₂(β-1)₂(β-2)₂(β-3)₂(β-4)₂(β-5)₂(β-6)₂(β-7)₂.