Background/Objectives: Metabolic profiling of tissue samples via liquid-state nuclear magnetic resonance (NMR) requires the extraction of polar metabolites in a suitable deuterated solvent. Such methods often prioritise metabolite recovery over protein removal due to the relatively low sensitivity of NMR metabolomics and the routine use of methods able to supress residual protein signals. However, residual protein may impact metabolite integrity and the metabolite stability after NMR sample preparation is often overlooked. This study aimed to investigate the effect of residual protein contamination in rodent brain extracts and identify a reproducible extraction method that optimises metabolite recovery while ensuring sample stability. Methods: The performance of acetonitrile/water (50–100% MeCN), methanol/water (50–100% MeOH), and methanol/water/chloroform (MeOH/H2O/CHCl3) were assessed for extraction efficiency, reproducibility, residual protein contamination, and metabolite stability up to eight hours post NMR sample preparation. Results: Aspartate and glutamate deuteration were observed in 50% MeCN, 50% MeOH, and 67% MeOH extractions along with the conversion of N-acetyl aspartate to aspartate and acetate in 50% MeCN and 50% MeOH extractions. Both observations correlated with residual protein contamination and, thus, are a result of inadequate protein precipitation, as confirmed by ultrafiltration. MeOH/H2O/CHCl3 extraction preserved the stability of these metabolites while maintaining good extraction efficiency and reproducibility. Conclusions: Thus, we recommend MeOH/H2O/CHCl3 extraction for untargeted brain NMR metabolic profiling due to its effective protein precipitation and reliable performance. Nonetheless, the performance of detecting metabolites prone to oxidation such as ascorbate and glutathione is not improved by this method.
Read full abstract