In this study, untargeted metabolic profiling by Gas Chromatography Mass Spectrometry (GC-MS) was leveraged to identify differences between Perchloric acid extraction (PCA) and 3:3:2 acetonitrile/isopropanol/water (IPA) extraction for liver metabolomics research. Perchloric acid extraction (PCA), a common means of precipitating excess protein through acidification, has been used in physiology literature for decades and is considered as the standard method for extracting water soluble metabolites. However, PCA extraction commonly requires tedious pH neutralization with potassium hydroxide (KOH) and subsequent desalting. While IPA extraction can also precipitate protein it is pH neutral and bypasses the need for neutralization and is more amenable for high throughput sampling. Several studies have demonstrated that IPA has a high extraction efficiency amongst other organic solvent based extractions such as 50:50 acetonitrile water and 80% methanol water in mammalian fecal extracts [Yang, Y., Yin, Y., Chen, X. et al Scientific Reports. 2019, 9(12017)] and plasma extracts [Jiye, A. et al. 2005 Anal. Chem. 77, 8086–8094] for GC-MS. Similarly, plasma extractions for Nuclear Magnetic Resonances (NMR) have demonstrated that 50:50 acetonitrile water extractions, a variant of organic extraction similar to IPA, are reproducible and are capable of eliminating excess protein from extracts while preserving low concentration metabolites [Daykin et al. Analytical Biochemistry 2002, 304, 220 –230]. In this study we expand the current knowledge of IPA versus PCA extractions in the field of liver metabolomics. Liver is a potential target for study, due to the increasing prevalence of Non-Alcoholic Fatty liver disease (NAFLD) and Non-alcoholic Steatohepatitis (NASH), which now affects a combined 80-100 million Americans [Perumpail et al. World J Gastroenterology. 2017, 23(47), 8263]. Due to the multifaceted nature of NAFLD approaches including NMR, mass spectrometry and various extraction variants have aimed to identify metabolic changes that would shed light on the mechanism of pathogenesis. To achieve a relevant background for analyzing the suitability of PCA or IPA for liver metabolomics in NAFLD, we extracted ex vivo C57/BLKS perfused mouse livers under fed, fasted and pooled conditions. We utilize C57/BLKS mice as they are prone to type 2 diabetes, atherosclerosis and obesity, all of which are significant cofactors for NAFLD development. Through our untargeted metabolomics panel several previously unidentified metabolites were assessed by principle components analysis and determined to be similar (Figure I). From our targeted metabolomics panel of extracted standards we also observed that several metabolites are largely similar in extraction efficiency with only the ketone 3-hydroxybutyrate having a higher extraction efficiency by PCA rather than IPA (Figure I). In conclusion, the extraction efficiency and statistical comparison of PCA and IPA extractions for GC-MS metabolomics assessments demonstrate their suitability and equivalence for all but a few metabolites.
Read full abstract