Profiling of Japanese Green Tea Metabolites by GC-MS

Importance of the topic

Metabolomic profiling of green tea provides deep insight into the chemical composition that underlies tea quality and flavor. Understanding the metabolite patterns associated with high-grade teas enables objective evaluation, quality control, and breeding strategies. GC-MS–based metabolomics offers comprehensive coverage of primary metabolites (amino acids, organic acids, sugars) that influence taste and health benefits.

Study objectives and overview

This study aimed to (1) extract and profile metabolites from nine varieties of high-grade Japanese green tea leaves ranked in a national competition, (2) identify marker compounds distinguishing ranking levels, and (3) develop a predictive model correlating metabolite profiles with competition ranking.

Methodology and instrumentation

Sample preparation and derivatization:

• Tea leaf (30 mg) ground in liquid nitrogen.
• Extraction with water/methanol/chloroform (1:2.5:1) plus ribitol internal standard.
• Phase separation, centrifugation, solvent removal, freeze-drying.
• Methyloximation (methoxyamine in pyridine) and trimethylsilylation (MSTFA) for GC-MS analysis.

GC-MS analysis:

• Instrument: Shimadzu GCMS-QP2010 Plus with AOC-20i+s auto-injector.
• Column: Restek Rtx-5MS (30 m × 0.25 mm i.d., 0.25 µm).
• Temperature program: 80 °C (2 min) to 320 °C at 15 °C/min (20 min hold).
• Electron impact ion source (200 °C), interface 250 °C, scan range m/z 50–1000.

Data processing and multivariate analysis:

• Peak identification via Shimadzu GC/MS Metabolites Spectral Database and NIST 2008 library.
• Quantitation based on peak area ratios to ribitol.
• SIMCA-P software for PCA and PLS regression.

Main results and discussion

• Approximately 100 peaks detected; 71 metabolites identified, including amino acids (e.g., glutamine, theanine), organic acids (e.g., citric, malic), sugars (e.g., sucrose, glucose), and caffeine.
• PCA score plot separated high- and low-ranking teas along principal component 1 (R2X[1] = 0.605, R2X[2] = 0.146, Q2 = 0.618), indicating distinct metabolite patterns.
• Loading plot revealed that higher-ranked teas were enriched in glutamine, arabinopyranose, and caffeine; lower-ranked teas showed higher sucrose, theanine, quinic acid, fructose, and glucose.
• PLS regression model using training ranks (5, 10, 20, 30, 40, 45) achieved R2Y = 0.977, Q2 = 0.972, RMSEE = 2.39; validation on ranks 15, 25, 35 gave RMSEP = 4.56, demonstrating robust ranking prediction.

Benefits and practical applications

• Objective, rapid evaluation of green tea quality based on metabolite fingerprints.
• Guidance for tea breeding programs to select favorable metabolite traits.
• Quality control in tea processing to ensure consistent flavor profiles.
• Potential integration into sensory analysis for consumer preference studies.

Future trends and potential applications

• Expansion of metabolite libraries with retention indices and custom spectral entries to cover more specialized compounds.
• Integration of GC-MS metabolomics with LC-MS and NMR for broader metabolome coverage.
• Automated sample preparation and real-time data processing for high-throughput tea screening.
• Application of machine learning to improve predictive accuracy for quality grading and flavor forecasting.
• Use of portable GC-MS systems for on-site quality monitoring in tea farms and processing facilities.

Conclusion

This study demonstrates that GC-MS–based metabolomic profiling, combined with multivariate analysis, effectively distinguishes high- and low-ranking green teas and enables reliable ranking prediction. The approach supports objective quality assessment, breeding decisions, and quality control in the tea industry. Continued advances in instrumentation, spectral databases, and data analysis will further enhance metabolomics applications in food science.

References

1. Fukusaki E. Possibilities and Technological Obstacles in Metabolomics. Biotechnology 2006, 84, 231–234.
2. Pongsuwan W.; Fukusaki E.; Bamba T.; Yonetani T.; Yamahara T.; Kobayashi A. Prediction of Japanese Green Tea Ranking by Gas Chromatography/Mass Spectrometry–Based Hydrophilic Metabolite Fingerprinting. J. Agric. Food Chem. 2007, 55, 231–236.