Chemical cross-linking mass spectrometry

Apo or heme-bound REV-ERBβ LBD and NCoR RID protein samples were cross-linked in Hepes gel filtration buffer [25 mM Hepes (pH 7.5) at room temperature, 50 mM NaCl, and 0.5 mM TCEP]. Non–cross-linked negative controls were generated using the same procedure without the addition of DSSO (Thermo Fisher Scientific). Samples included 4 μM apo LBD with or without 2 μM RID and 4 μM heme-bound LBD with or without 2 μM RID; complexes were preincubated 1 hour on ice. Reactions were initiated by spiking DSSO in DMSO at a final concentration of 600 μM in 50-μl solutions. The final molar ratio of cross-linker:protein was 100:1. The reactions were incubated at 25°C for 45 min before being quenched by the addition of tris (pH 8.0) to a final concentration of 50 mM. Each cross-link reaction was done in triplicate, and the reaction replicates were pooled after quenching. Samples (10 μl) were loaded for SDS-PAGE analysis with Coomassie staining to confirm the presence of cross-linked protein. The remaining cross-linked and non–cross-linked samples were then acetone-precipitated at −20°C overnight, and the protein was pelleted by centrifugation at 16,000 relative centrifugal force (rcf) for 5 min at 4°C. After decanting, the pellets were dried for 15 min at room temperature before being resuspended in 12.5 μl of resuspension buffer [50 mM ammonium bicarbonate and 8 M urea (pH 8.0)]. ProteaseMAX (Promega) was added to 0.02%, and the solutions were mixed on an orbital shaker operating at 400 rpm for 5 min. After resuspension, 87.5 μl of digestion buffer [50 mM ammonium bicarbonate (pH 8.0)] was added. Protein solutions were digested using trypsin (Promega) at a ratio of 1:200 (w/w, trypsin:total protein) and incubated at 37°C. After overnight incubation at 37°C, the samples were acidified by the addition of trifluoroacetic acid (TFA; Thermo Fisher Scientific) to 1%. Samples were then frozen and stored at −20°C until analysis.

Peptide samples were thawed, vortexed, and spun down at 16,000 rcf for 5 min. Ten microliters of peptide samples were loaded in an UltiMate 3000 autosampler (Dionex, Thermo Fisher Scientific). Approximately 500 ng of each peptide sample was injected in triplicate. Peptides were trapped on a μPAC trapping column (PharmaFluidics) using a load pump operating at 20 μl/min. Load pump buffer contained 2% acetonitrile and 0.1% TFA. After a 3-min desalting period, peptides were then separated using linear gradients (2 to 30% solvent B over 1 to 60 min and 30 to 95% solvent B over 60 to 90 min) at 1 μl/min on a 50-cm μPAC C18 column (PharmaFluidics). Gradient solvent A contained 0.1% formic acid, and solvent B contained 80% acetonitrile and 0.1% formic acid. The liquid chromatography eluate was interfaced to an Orbitrap Fusion Lumos (Thermo Fisher Scientific) via nanospray ionization source. Cross-links were identified using a previously described method (43). Master scans of mass/charge ratio (m/z) 375 to 1500 were taken in the Orbitrap mass analyzer operating at 60,000 resolution at 400 m/z, with automatic gain control target set to 400,000. Maximum ion injection time was set to 50 ms, and advanced peak detection was enabled. Precursor ions with a charge state between 4 and 8 were selected via quadrupole for collision-induced dissociation fragmentation at 25% collision energy and 10-ms reaction time. Fragment ion mass spectra were taken on the Orbitrap mass analyzer operating at 30,000 resolution at 400 m/z, with an automatic gain control target of 50,000, and maximum injection time was set to 150 ms. Doublet pairs of ions with the targeted mass difference for sulfoxide fragmentation [31.9721 Da (44)] were selected for higher-energy collisional dissociation fragmentation at 35% collision energy. MS3 scans were collected in the ion trap operating in “rapid” mode at automatic gain control target and maximum ion injection time set to 20,000 and 200 ms, respectively.

Cross-links were identified using the XlinkX algorithm (45) implemented on Proteome Discoverer (version 2.2). Cross-links were considered for lysine, threonine, serine, and tyrosine residues, and the validation strategy was set to “simple” where relaxed and strict false discovery rates were set to 0.05 and 0.01, respectively. During database searches, the target databases only contained sequences for the proteins that were analyzed, and the decoy database contained the reverse sequences. The maximum missed cleavages were set to 8. Methionine oxidation was considered as a variable modification. Mass tolerances for precursor Fourier transform MS (FTMS), fragment FTMS, and fragment ion trap MS (ITMS) were set to 10 and 20 parts per million (ppm) and 0.5 Da, respectively. Peak areas for identified cross-links were quantified using Skyline (version 19.1) using a previously described protocol (46). Cross-link spectral matches found in Proteome Discoverer were exported and converted to sequence spectrum list format using Excel (Microsoft). Cross-link peak areas were assessed using the MS1 full-scan filtering protocol for peaks within 8 min of the cross-link spectral match identification. Peaks areas were assigned to the specified cross-linked peptide identification if the mass error was within 10 ppm of the theoretical mass, if the isotope dot product was greater than 0.95, and if the peak was not found in the non–cross-linked negative controls. Pairwise comparisons were made using “MSstats” package (47) implemented in the Skyline browser to calculate relative fold changes and significance (multiple testing adjusted P values). Significant changes were defined as −log₁₀; adjusted P value was greater than 1.3 (P value less than 0.05). The results from skyline were exported and are reported in an Excel spreadsheet. Dot plots were generated in R. SASD calculations were determined using Jwalk (48) using the heme-bound LBD structure as a template (PDB 3CQV). Analyzed XL-MS data are provided in data file S2.