Proteomic mapping of organ secretomes using in vivo proximity labeling
Authors: Justin A. Bosch 1; Pierre M. J. Beltran 2; Cooper Cavers 1; Thai LaGraff 1; Randy Melanson 2; Steven Carr 2; Norbert Perrimon 1,3
Affiliations: 1) Department of Genetics, Blavatnik Institute, Harvard Medical School, Boston, MA; 2) Broad Institute, Cambridge, MA; 3) Howard Hughes Medical Institute
Keywords: n. proteomics; t. other (inter-organ communication)
Secreted proteins regulate essential biological processes and are attractive for developing therapeutics. Due to their low abundance in extracellular fluids, discovering the complete set of secreted proteins (the “secretome”) - as well as their tissue of origin – can be extremely challenging. To address this, we developed a proteomic method to label proteins in specific organs in Drosophila larvae using the biotin ligase TurboID, and subsequently enrich and identify biotinylated proteins in the hemolymph. Using this tool, we constructed a secretome map of 540 proteins from 10 major cell types (e.g. fat body, muscle). In addition to identifying most known blood proteins (e.g. Lsp2, apolpp), we discovered hundreds of uncharacterized proteins, in many cases derived from a single cell type. Furthermore, while most blood proteins originate from the fat body, we identify secreted proteins from less appreciated cell-types, such as oenocytes and glia. Using CRISPR/Cas9 knock-in flies to validate selected tissue-specific uncharacterized proteins, we found that most were expressed in the presumed tissue and some were detected in circulation by western blot. One protein, CG6867, the single fly homologue of human Olfactomedin proteins, is specifically expressed in somatic muscle and secreted into larval hemolymph. Interestingly, CG6867 accumulates on the basal surface of most tissues (e.g. imaginal discs), suggesting CG6867 may be a novel circulating basement membrane protein. Our secretome map will serve as a resource to investigate blood protein function, discover organ communication factors, and compare with homologues of human biomarkers.