Abstract
Proximity-dependent biotinylation (PDB) is a powerful means of exploring
the cellular environments in which proteins reside. Expressing a protein
of interest (bait) fused to a biotin ligase and adding biotin induces
the covalent biotinylation of proximal partners, which are recovered on
streptavidin beads and identified by MS. However, a major technical
limitation of PDB is peptide carryover into subsequent MS runs. This is
mitigated via lengthy inter-sample washing, considerably lowering
throughput. This study aims to optimize PDB sample acquisition using an
EvoSep LC system coupled to a timsTOF mass spectrometer, which has
higher throughput and sensitivity than our current system, with less
carryover. Our efforts resulted in an ~15-fold increase in
throughput using the 60 samples-per-day (SPD) gradient with better
sensitivity, identifying nearly double the proteins found by our
previously standardized workflow. Significance scoring also revealed
more sensitive detection of high-confidence proximal interactions
(~1.5-fold) for five well-characterized baits, validating
the new experimental workflow. Importantly, carryover was extremely
limited, even without inter-sample washing, and limited to easily
filterable abundant proteins. Without washing, this method can process
60 samples per day, using 1/16th of the sample amount previously
required.