Mapping Ku70 Protein Interactions Using Proximity-Dependent Biotin Identification
Abstract
The Ku heterodimer, composed of subunits Ku70 and Ku80, is a highly abundant protein complex, known for its affinity for double-stranded DNA ends. Accordingly, Ku is most well-studied for repairing double-stranded DNA breaks through the non-homologous end-joining (NHEJ) DNA repair pathway. Aside from NHEJ, Ku has also been implicated and studied in various other cellular processes including V(D)J recombination and telomere maintenance.
Numerous protein interactions have been mapped to the C-terminal region of Ku70, although few have been mapped to its N-terminal von Willebrand A-like (vWA) domain. Here, we used the high-throughput proteomic technique, proximity-dependent biotin identification (BioID, BioID2) to identify proteins that interact with full-length Ku70, the Ku70 vWA domain, and with a novel phosphorylation site, serine 155 (S155), found within the Ku70 vWA domain.
Using BioID, a proximity-based in vivo protein labelling technique, and a second high-throughput proteomics technique, affinity purification coupled to mass spectrometry (AP-MS), we were able to establish the Ku protein interactome and infer potential protein complexes and cellular pathways that may involve Ku.
Using BioID2 with various segments of the Ku70 protein including full-length Ku70, Ku70 ΔvWA, and the vWA domain targeted to the nucleus alone, we were able to map domain-specific interactions, test the specificity of the BioID2 technique, and validate RNF113A and Spindly as proximal and/or transient Ku70 vWA-specific interactors. Finally, we also used the BioID2 technique to identify Ku70 phospho-S155-specific interactors by comparing the BioID2 proteomic datasets for Ku70 S155A to S155D. We identified TRIP12, CHD3, and Borealin as candidate proximal interactors of Ku70 S155D, and we were able to further verify a proximity-based association between Ku70 S155D and TRIP12.
In sum, this research is the first to utilize the BioID and BioID2 techniques to identify in vivo candidate proximity-based interactors for the Ku heterodimer, the Ku70 vWA domain, and Ku70 S155, allowing us to infer novel functions that may involve Ku70. In addition, we demonstrated the versatility of the BioID2 technique by using it to identify full-length, domain-specific, and residue-specific candidate interactors, laying the foundation for future studies.