Automated sample preparation with SP3 for low-input clinical proteomics

Abstract

High-throughput and streamlined workflows are essential in clinical proteomics for standardized processing of samples originating from a variety of sources, including fresh frozen tissue, FFPE tissue, or blood. To reach this goal, we have implemented single-pot solid-phase-enhanced sample preparation (SP3) on a liquid handling robot for automated processing (autoSP3) of tissue lysates in a 96-well format, performing unbiased protein purification and digestion, and delivering peptides that can be directly analyzed by LCMS. AutoSP3 eliminates hands-on time and minimizes the risk of error, reduces variability in protein quantification and improves longitudinal performance and reproducibility. We demonstrate the distinguishing ability of autoSP3 to process low-input samples, reproducibly quantifying 500-1000 proteins from 100-1000 cells (<100 ng protein). Furthermore, we applied this approach to a cohort of clinical FFPE pulmonary adenocarcinoma (ADC) samples, and recapitulate their separation into known histological growth patterns based on proteome profiles. Collectively, autoSP3 provides a generic, scalable, and cost-effective pipeline for routine and standardized proteomic sample processing that should enable reproducible proteomics in a broad range of clinical and non-clinical applications.