Metabolic phenotyping using the iKnife identifies IRS2 as a driver of lipid metabolism reprogramming in colorectal cancer

Abstract

The advent of precision medicine is changing the clinicians' approach to therapeutic decisions. It also implies that the identification of biomarkers and discriminating factors is necessary for more effective treatment strategies and tumours can no longer be considered by their tissue of origin of TNM stage alone. Colorectal cancer (CRC) is the second leading cause of cancer-related deaths in the Western world. Despite numerous existing classifications, there is a real need for a better understanding of the clinically relevant features that stratify patients with CRC, ultimately leading to more personalised therapies. Metabolic reprogramming is recognised as one of the main cancer hallmarks. As such, the metabolic signatures of tumours can be a rich source of information for better patient stratification. To investigate lipid metabolism reprogramming in CRC, an emerging mass spectrometry technique called the iKnife (or intelligent knife) was used on a panel of 60 CRC cell lines. Unsupervised hierarchical clustering of the cell lines based on their REIMS profile suggested a stratification of the disease into two subtypes, with profound changes in the overall abundance of lipids. Subsequent bioinformatic analysis identified alterations in Insulin-Receptor Substrate 2 (IRS2) as a potential driver of the lipidome-based stratification of CRC. IRS2 overexpression resulted in a shift in cells' metabolic behaviour, with an increase in glycolytic activity and de novo lipogenesis. Newly synthesised fatty acids were found to be stored in lipid droplets, where an increase in both triglyceride and choloresterol ester content was observed. The investigation of the enzymes involved in lipid droplet biogenesis and fatty acid storage revealed that DGAT1 expression was increased in IRS2 overexpressing cells. Mechanistically, this increase in the enzyme responsible for the storage of free fatty acids in triglycerides was found to be regulated downstream of AKT and mTOR signalling. Interestingly, IRS2 overexpression also led to an increase in ACSS2, a crucial enzyme for the conversion of acetate to acetyl-CoA, under nutrient-poor conditions, suggesting a high degree of plasticity for fatty acid acquisition downstream of IRS2. In conclusion, the investigation of the CRC liposome using the iKnife suggested an IRS2-driven stratification associated with a profound remodelling of the lipid-related metabolic activity which could affect both tumour progression and response to therapy.