Rapid determination of quaternary protein structures in complex biological samples

Abstract

The understanding of complex biological systems is still hampered by limited knowledge of biologically relevant quaternary protein structures. Here, we demonstrate quaternary structure determination in biological samples using a combination of chemical cross-linking, high-resolution mass spectrometry and high-accuracy protein structure modeling. This approach, termed targeted cross-linking mass spectrometry (TX-MS), relies on computational structural models to score sets of targeted cross-linked peptide signals acquired using a combination of mass spectrometry acquisition techniques. We demonstrate the utility of TX-MS by creating a high-resolution quaternary model of a 1.8 MDa protein complex composed of a pathogen surface protein and ten human plasma proteins. The model is based on a dense network of cross-link distance constraints obtained directly in a mixture of human plasma and live bacteria. These results demonstrate that TX-MS can increase the applicability of flexible backbone docking algorithms to large protein complexes by providing rich cross-link distance information from complex biological samples.

ipaA triggers vinculin oligomerization to strengthen cell adhesion during Shigella invasion

“…We used the constraints obtained from the cross-linking MS data (Suppl. Tables 1-3) to guide the protein structure modeling using the TX-MS protocol as described by Hauri, Khakzad et al 29 . In short, TX-MS uses the Rosetta comparative modeling protocol (RosettaCM) 30 , and the flexible backbone docking protocol (RosettaDock) 31 to generate models and evaluate how well each model explains the MS constraints using a novel scoring function.…”

Structural determination ofStreptococcus pyogenesM1 protein interactions with human immunoglobulin G using integrative structural biology

“…For cross-linking, pooled normal human plasma was adsorbed onto the surface of S. pyogenes bacteria, as described [4,15]. The S. pyogenes M1 serotype strain SF370 from the American Type Culture Collection (ATCC; strain reference 700294), was grown at 37 °C, 5% CO2 to mid-exponential phase (OD620nm ∼ 0.4) in TH broth supplemented with 0.3% (w/v) yeast extract.…”

“…To elucidate the Fc-binding interface, we used TX-MS based on hrMS1 and DDA data that we combined with previously determined crystal structures of human IgGs and the partial peptides of M1 (A-and S-domains) [15] (see Methods). As demonstrated in Figure 2, all IgG subclasses bind to the M1 protein through a specific site in the IgG CH3 domain, which is essential for binding to FcγR receptors.…”

“…In the family of M proteins, antigenic variation has resulted in more than 200 serotypes, but only a few are frequently associated with invasive disease, with M1 being the most prevalent serotype [14]. Moreover, to physically protect its vulnerable antigenic epitopes, GAS scavenges human plasma proteins to form a surrounding coat-like barrier [4,15,13], and among these interactions specifically binds IgG-molecules via their Fc-domains rendering these inaccessible for Fc-receptors [16].…”

“…Targeted cross-linking mass spectrometry (TX-MS) relies on computational structural models to score sets of targeted cross-linked peptide signals acquired using a combination of mass spectrometry acquisition techniques such as data-dependent acquisition (DDA), data-independent acquisition (DIA) and high-resolution MS1 (hrMS1). We have previously demonstrated the utility of TX-MS by creating a high-resolution quaternary model of a 1.8 MDa protein complex composed of the M1 protein and ten human plasma proteins [15]. Among these interactions, the M1-IgG binding is found to be environment-specific, binding via Fab-domains under antibody-rich conditions such as plasma, or via Fc-domains in an antibody-poor environment such as saliva [16].…”