A single-sample workflow for joint metabolomic and proteomic analysis of clinical specimens

Abstract

AbstractUnderstanding the interplay of the proteome and the metabolome aids in understanding cellular phenotypes. To enable more robust inferences from such multi-omics analyses, combining proteomic and metabolomic datasets from the same sample provides major benefits by reducing technical variation between extracts during the pre-analytical phase, decreasing sample variation due to varying cellular content between aliquots, and limiting the required sample amount. We evaluated the advantages, practicality and feasibility of a single-sample workflow for combined proteome and metabolome analysis. In the workflow, termed MTBE-SP3, we combined a fully automated protein lysis and extraction protocol (autoSP3) with a semi-automated biphasic 75% EtOH/MTBE extraction for quantification of polar/non-polar metabolites. Additionally, we compared the resulting proteome of various biological matrices (FFPE tissue, fresh-frozen tissue, plasma, serum and cells) between autoSP3 and MTBE-SP3. Our analysis revealed that the single-sample workflow provided similar results to those obtained from autoSP3 alone, with an 85-98% overlap of proteins detected across the different biological matrices. Additionally, it provides distinct advantages by decreasing (tissue) heterogeneity by retrieving metabolomics and proteomic data from the identical biological material, and limiting the total amount of required material. Lastly, we applied MTBE-SP3 to a lung adenocarcinoma cohort of 10 patients. Integrating the metabolic and proteomic alterations between tumour and non-tumour adjacent tissue yielded consistent data independent of the method used. This revealed mitochondrial dysfunction in tumor tissue through deregulation of OGDH, SDH family enzymes and PKM. In summary, MTBE-SP3 enables the facile and confident parallel measurement of proteins and metabolites obtained from the same sample. This workflow is particularly applicable for studies with limited sample availability and offers the potential to enhance the integration of metabolomic and proteomic datasets.