1H-NMR metabolomics-based classification of Japanese sake and comparative metabolome analysis by gas chromatography–mass spectrometry

Abstract

Six categories of Japanese sake have been established by the National Tax Agency of Japan. In this system, the rice polishing ratio and the addition of alcohol are the main criteria for classification. The most common nuclear magnetic resonance (NMR) spectrometry method is 1H-NMR, and has higher throughput than gas chromatography-mass spectrometry (GC-MS) or liquid chromatography-mass spectrometry (LC-MS) analysis due to its short measurement time, easy sample preparation, and high reproducibility. However, owing to the production of dominant ethanol signals, metabolome analyses have not been used for classifying Japanese sake using 1H-NMR. In this study, a technique to selectively suppress ethanol signals was used to classify Japanese sake by 1H-NMR, and a model was constructed to predict the rice polishing ratio. The results were compared to those obtained by GC-MS. The suppression of ethanol signals enabled the detection of trace components by 1H-NMR. In a principal component analysis (PCA) score plot of 1H-NMR spectra with ethanol signal suppression, PC1 was associated with both the addition of alcohol and the rice polishing ratio. Additionally, the separation of samples observed was similar when PCA score plots of 1H-NMR and GC-MS data were compared. Similarly, to predict the rice polishing ratio using partial least squares regression analysis, a model was constructed using 1H-NMR data, and showed nearly similar values for precision and predictive performance with the model constructed using GC-MS data. These results suggest that metabolomic analyses of Japanese sake based on 1H-NMR spectral patterns may be useful for classification.