Sensitive detection and structural characterisation of UV-induced cross-links in protein-RNA complexes using CLIR-MS

Abstract

AbstractCross-linking coupled with mass spectrometry is an increasingly popular methodology for elucidating structural information from biological complexes. Whilst protein-protein cross-linking workflows are widely used and well characterised, adoption of protein-RNA cross-linking workflows for structural studies is less widespread, and data produced from such experiments remains less well understood. The cross-linking of stable isotope labelled RNA coupled to mass spectrometry (CLIR-MS) workflow uses isotope labelled RNA to simultaneously confirm that peptides are cross-linked to RNA and aid cross-link localisation in an RNA sequence. For broader application of CLIR-MS as part of the structural analysis of ribonucleoproteins, the method must be sensitive, robust, and its reaction products need to be well characterised. We enhanced our previously published workflow, improving coverage and sensitivity. We used it to infer common properties of protein-RNA cross-links such as cross-linking distance, and to assess the impact of substitution of uracil with 4-thio-uracil in structural proteomics experiments. We profiled the compositional diversity of RNA-derived peptide modifications, and subsequently defined a more inclusive data analysis approach which more than doubles the number of cross-link spectrum matches compared with our past work. We defined distance restraints from these cross-links, and with the aid of visualisation software, demonstrated that on their own they provide sufficient information to localise an RNA chain to the correct position on the surface of a protein. We applied our enhanced workflow and understanding to characterise the binding interface of several protein-RNA complexes containing classical and uncommon RNA binding domains. The enhanced sensitivity and understanding demonstrated here underpin a wider adoption of protein-RNA cross-linking in structural biology.