Rapid Screening of Herbal Medicine Using Single Quadrupole LC-MS

Importance of the Topic

The chemical composition and quality of herbal medicines can vary widely based on factors such as geographic origin, cultivation conditions and harvesting time. A rapid, reliable screening method is essential for ensuring batch consistency, therapeutic efficacy and safety of complex natural products.

Study Objectives and Overview

This study demonstrates a high‐throughput approach for classifying and characterizing Kakkonto (a traditional Kampo herbal formulation) extracts using flow injection analysis combined with single quadrupole LC‐MS. Key aims included:

• Detecting a broad range of metabolites without chromatographic separation
• Applying multivariate statistical tools to differentiate sample types
• Identifying marker compounds via QTOF LC‐MS for enhanced accuracy

Methodology

Samples of five commercial Kakkonto granule products (A–E) were extracted with 50% methanol, followed by ultrasonic treatment and centrifugation. Diluted supernatants containing internal standards (reserpine, chloramphenicol) were subjected to direct injection (flow injection analysis) into a single quadrupole LCMS-2050. Positive and negative electrospray ionization generated mass spectra (m/z 50–2000) within a one‐minute analysis time.

Instrumentation Used

• Nexera™ XR HPLC system for flow injection at 0.1–1.0 mL/min
• LCMS-2050 single quadrupole mass spectrometer (ESI/APCI, dual polarity)
• LCMS-9030 QTOF system for MS/MS confirmation
• LabSolutions LCMS and eMSTAT Solution™ software for data processing and multivariate analysis

Main Results and Discussion

Flow injection LC‐MS detected a total of 290 peaks across positive (91 peaks) and negative (199 peaks) modes. Principal component analysis (PCA) separated sample types, revealing close clustering of products B and E, while sample A exhibited distinct metabolic profiles. A RandomForest discriminant model further classified an unknown sample F with 88.5% average score, grouping it near B and E.

Key marker peaks at m/z 255 (positive mode) and m/z 821 (negative mode) were subjected to QTOF MS/MS. High‐accuracy matching (<1 mDa error) against ChemSpider identified these as daidzein and glycyrrhizin, respectively.

Practical Benefits and Applications

• Sub‐minute analysis time enables high‐throughput screening of multiple herbal batches
• Simple flow injection protocol reduces instrument contamination risk due to LCMS-2050 robustness
• Multivariate statistical workflows allow non‐experts to classify and authenticate herbal products

Future Trends and Possibilities

Continued integration of high‐resolution QTOF and time‐of‐flight platforms will refine compound identification workflows. Machine learning algorithms applied to larger metabolomic datasets can improve predictive accuracy for herbal authenticity and origin. Miniaturized direct‐infusion devices linked to cloud‐based analytics may further accelerate quality control in field settings.

Conclusion

The combination of flow injection single quadrupole LC‐MS with multivariate analysis offers a rapid, robust and user‐friendly approach for screening complex herbal formulations. The demonstrated discriminant model and marker identification pave the way for streamlined quality assessment and authentication of Kampo and other natural product medicines.

References

Hattori T., Shibayama Y., Watabe Y. Rapid Screening of Herbal Medicine Using Single Quadrupole LC‐MS. Shimadzu Application News, First Edition: Apr. 2024.