Fig. 1

Characterisation of RNA–protein complexes, RBPs and RBDs. We characterised the Drosophila RBPome at four levels of resolution. a (green box) On the first two levels of resolution we focus on comprehensive identification of RBPs and RNA-associated protein complexes in Drosophila cells. First we used FA crosslinking to compile a dataset containing both direct and indirect RNA binders (FA-RBPome (1)). At the next level we used UV crosslinking to generate a dataset composed of direct RNA binders (UV-RBPome (2)). Both FA- and UV-RBPome datasets exhibit peptide coverage across the entire length of RBPs, which is illustrated by the green FP-peptides in the schematic in the lower panel. b (blue box) The next two levels of resolution focused on discovery of RNA–protein contact sites. We devised a new approach, termed CAPRI, which integrates two analysis pipelines to facilitate large-scale identification of RBDs. Briefly, UV-crosslinked RBPs are isolated using oligo-dT beads and digested with endoprotease Lys-C to release non-crosslinked peptides. The RNA-peptide moieties are eluted from the beads, transferred to a 30 kD filter and washed with 8 M urea to remove nonspecific background peptides. The crosslinked peptides are further digested with trypsin (cutting C-terminal to K/R) to release the adjacent peptides (coloured blue) as flowthrough. The RNA-peptide conjugates (orange-red) retained on the filter were released after degrading the RNA using a cocktail of nucleases. The MS/MS data from all crosslinked replicates are submitted to PEAKS for a combined search with custom defined monoisotopic masses of RNA adducts to identify XL-peptides. The non-heterconjugate peptides from all MS/MS data are analysed by MaxQuant. The statistically enriched peptides in the UV irradiated samples (adjacent-peptides) are reconstructed in silico to the original endoproteinase Lys-C-digested full peptide sequence (ADJ-peptides: blue). The ADJ-RBPome (3) and XL-RBPome (4) together give the CAPRI-RBPome. c (orange box) The FA-dom-RBPome provides complementary data on RBDs using FA as a crosslinker. Briefly, 0.1% formaldehyde-crosslinked RBPs are captured using oligo-dT beads and are further digested using Lys-C and trypsin. Following elution the RNA-peptide crosslinks are reversed by heating and later the nucleic acids are degraded by nucleases. The peptides are cleaned-up using SP3 purification and analysed using LC/MS