Session: Parallel session 3 - AI and Bioinformatics

TWISTT-PRM: A Novel Real-Time Acquisition Method Enabling Targeted Proteomics with Increased Throughput

Zoltan UDVARDY¹, Anne GONZALEZ DE PEREDO¹, Yoann ROMBOUTS¹, Alexandre STELLA¹, Odile BURLET-SCHILTZ¹, Mathias WILHELM², David BOUYSSIÉ¹

¹Institut de Pharmacologie et de Biologie Structurale (IPBS), Université de Toulouse, CNRS, Université Toulouse III - Paul Sabatier (UT3), 31062 Toulouse, France, Toulouse, France
²Computational Mass Spectrometry, TUM School of Life Sciences, Technical University of Munich, Freising, Germany, Freising, Germany

Introduction

Discovery acquisition methods, such as DDA and DIA are used today to identify and quantify thousands of proteins in complex biological samples, but they cannot capture the least abundant species (e.g. cytokines in serum or bacterial proteins in infected cells). Although hypothesis-driven approaches like SRM and PRM offer higher sensitivity, the low number of peptides they can monitor restricts their use to certain applications. To mitigate this limitation, we propose TWISTT-PRM (TWo Injections with Smart Triggering of Targets for Parallel Reaction Monitoring), an intelligent PRM method leveraging real-time control of Thermo MS instruments.

Methodology

The proposed solution has two stages: a reference run and a targeted run. It applies an accurate RT adjustment procedure in real-time based on high-confidence peptide signals collected from the reference run, in order to increase the number of targets that can be monitored by PRM. The accuracy of the RT correction and sensitivity of the method were assessed using a tailored sample mixture containing 713 synthetic peptides (proteotypic peptides from cytokines and tumor associated antigens) spiked in an E. coli background at different concentrations (from 20 to 0.2 fmol/µL). The results were compared to acquisitions performed using a classical DIA method. All data were recorded on a Thermo Exploris 480 instrument.

Results

TWISTT-PRM exhibited higher sensitivity compared to DIA (+50% synthetic peptides detected at lowest concentration). Moreover the quantitative measurements contained less missing values and proved to be as precise as DIA (similar CV distributions) but more accurate (observed ratios closer to theoretical ones).

Conclusion

TWISTT-PRM demonstrates higher sensitivity, repeatability, precision and accuracy compared to DIA. Besides, it greatly improves the acquisition throughput compared to conventional PRM, thus opening promising perspectives for targeted proteomics studies.