Quantitative cross-linking/mass spectrometry
Dynamic proteins and multi-protein complexes govern most biological processes. Cross-linking/mass spectrometry (CLMS) is increasingly successful in providing residue- resolution data on static proteinaceous structures. Here we investigate the technical feasibility of recording dynamic processes using isotope-labelling for quantitation. We cross-linked human serum albumin (HSA) with the readily available cross-linker BS3-d0/4 in different heavy/light ratios. Wefound two limitations. First, isotope labelling reduced the number of identified cross-links. This is in line with similar findings when identifying proteins. Second, standard quantitative proteomics software was not suitable for work with cross-linking. To ameliorate this we wrote a basic open source application, XiQ. Using XiQ we could establish that quantitative CLMS was technically feasible. Biological significance Cross-linking/mass spectrometry (CLMS) has become a powerful tool for providing residue- resolution data on static proteinaceous structures. Adding quantitation to CLMS will extend its ability of recording dynamic processes. Here we introduce a cross-linking specific quantitation strategy by using isotope labelled cross-linkers. Using a model system, we demonstrate the principle and feasibility of quantifying cross-linking data and discuss challenges one may encounter while doing so.We then provide a basic open source application, XiQ, to carry out automated quantitation of CLMS data. Ourwork lays the foundations of studying themolecular details of biological processes at greater ease than this could be done so far.
Sample Processing Protocol
Fifteen microgram aliquots of 0.75 M human serum albumin (HSA) (Sigma) in cross-linking buffer (20 mM HEPES-KOH, 20 mM NaCl, 5 mM MgCl2, pH 7.8) were each cross-linked with mixtures of bis[sulfosuccinimidyl] suberate-d0 (BS3-d0) (Thermo Fisher Scientific) and its deuterated form bis [sulfosuccinimidyl] 2,2,7,7-suberate-d4 (BS3-d4) (Thermo Fisher Scientific). For the purpose of quantitation, BS3-d0 and BS3-d4 weremixedwith threemolar ratios, 1:1, 1:2 and 1:4. The ratio of BS3-d4:HSA was 4:1 (by mass) in all three mixing ratios. Three replicaswere prepared for eachratio.The cross-linking reaction was incubated at room temperature (~23 °C) for 1 hour, and quenched by addition of ammonium bicarbonate and incuba- tion for 30 minutes at room temperature. Cross-linked protein samples were isolated on SDS–PAGE gel, and in-gel digested using trypsin following a standard protocol . After digestion, peptide solutions were desalted using self-made C18-StageTips , following the published protocol  for subsequentmass spectrometric analysis. We used as analytical column a spray emitter (75-μm inner diameter, 8-μm opening, 250-mm length; New Objectives) that was packed with C18 material (ReproSil-Pur C18-AQ 3 μm; Dr Maisch GmbH, Ammerbuch-Entringen, Germany) by help of an an air pressure pump (Proxeon Biosystems) .Mobile phase A consisted of water and 0.1% formic acid. Mobile phase B consisted of acetonitrile and 0.1% formic acid. Peptides were loaded onto thecolumnwith1%B at 700nl/minflow rateand eluted at 300 nl/min flow rate with a gradient: 1 minute linear increase from 1% B to 9% B; linear increase to 35% B in 169 minutes; 5 minute increase to 85% B. The eluted peptides were directly sprayed into an LTQ-Orbitrap Velos mass spec- trometer (Thermo Fisher Scientific). Mass spectrometric analy- ses were carried out using a “high-high” acquisition strategy [7,9]. The survey scan (MS) spectrawere recorded in theOrbitrap at 100,000 resolution. In each acquisition cycle, the eight most intense signals in the survey scan were isolated with an m/z window of 2 Th and fragmented with collision-induced dissoci- ation (CID) in the ion trap. 1+ and 2+ ions were excluded from fragmentation. Fragmentation (MS2) spectra were acquired in the Orbitrap at 7500 resolution. Dynamic exclusionwas enabled with 90 seconds exclusion time and repeat count equal to 1.
Data Processing Protocol
The raw mass spectrometric data files were processed into peak lists using MaxQuant version 18.104.22.168  with default parameters, except “Top MS/MS Peaks per 100 Da” was set to 200. The peak lists were searched against the sequences of HSA using Xi software (ERI, Edinburgh) for identification of cross-linked peptides. Search parameters were as follows: MS accuracy, 6 ppm; MS/MS accuracy, 20 ppm; enzyme, trypsin; specificity, fully tryptic; allowed number of missed cleavages, four; cross-linker, BS3-d0/d4; fixed modifications, carbamidomethylation on cysteine; variable modifications, oxidation on methionine. The linkage specificity for BS3 was assumed to be at lysine, serine, threonine, tyrosine and protein N-termini. Identified candidates of cross-linked peptides were validated by Xi, and only auto-validated cross-linked peptides were used for subsequent quantitation. Distribution of these cross-links was visualised in the crystal structure of HSA (PDB| 1AO6)  using PyMOL . Distances between alpha-carbons (C-α distances) of cross-linked residues were measured and compared to the maximumcross-linker length, which allowed for further validation of these identified cross-links . Five cross-linked peptide pairs were quantified manually for each mixing ratio. The cross-links were selected as being amongst those with highest identification confidence in all three replicas. For each cross-linked peptide, the summed intensities of the first three isotope peaks in the isotope cluster of heavy signals and light signals were used to calculate the signal ratio of BS3-d0 cross-linked (light) to BS3-d4 cross-linked (heavy) peptides. We excluded the monoisotopic peak because the relatively small mass difference of 4 Da used here can lead to an overlap of isotope clusters for light and heavy peptides. Peak intensitywas defined as the peak area for each isotope peak, and was derived from raw data using Peak detection in Xcalibur (version 2.1.0, Thermo Scientific). Visual inspection ensured that any overlap between light and heavy signals was minimal and that no other signals interfered with the quantitation. MaxQuant currently does not support quantitation of cross- linked peptides or indeed of any third-party identifications. Therefore we had to rely on first doing a quantitation that considers doublet signals detected by MaxQuant and thenmatching these to our results, a process that omits the re-quantitation routine of MaxQuant for identified peptides. We analysed our raw-files using MaxQuant (version 22.214.171.124) allowing for 5missed cleavages, using a single protein fasta file (HSA), no contaminants, disabled “I= L” and “Filter labelled amino acids,” and multiplicity was set to two. We selected “Lys4” as the heavy label, which corresponds to themass shift introduced by the labelled cross-linker used in our study. After MaxQuant finished,we concentrated on the allPeptides.txt file. This contains all found “features” in the raw-files and the assigned ratios. Thesewere thenmapped back to the identified cross-linked peptides. The ratio for each cross-link site was taken as the median of all supporting ratios for that site. The Xi quantitation application (XiQ) was implemented in C++, following and automating the manual quantitation procedure. It is used in on-going projects in our lab while we are trying to convince the makers of other freely available and more powerful quantitation packages, e.g. MaxQuant  and Skyline  to make the necessary amendments to their tools that allow cross-link data to be used. XiQ application is available open source at http://xiq.rappsilberlab.org.
Sven Giese, Department of Bioanalytics, Institute of Biotechnology, Technische Universität Berlin, 13355 Berlin, Germany
Juri Rappsilber, 1. Department of Bioanalytics, Institute of Biotechnology, Technische Universität Berlin, 13355 Berlin, Germany, 2. Wellcome Trust Centre for Cell Biology, School of Biological Sciences, University of Edinburgh, Edinburgh EH9 3JR, United Kingdom; ( lab head )
Fischer L, Chen ZA, Rappsilber J. Quantitative cross-linking/mass spectrometry using isotope-labelled cross-linkers. J Proteomics. 2013 Aug 2;88:120-8 PubMed: 23541715