Accurate Compound Identification from Complex Natural Product Samples Using SONAR

Significance of the Topic

The accurate identification of compounds in multi-herb natural product preparations is essential for ensuring safety, efficacy and quality control in pharmaceutical and nutraceutical industries. Complex matrices and co-eluting constituents often generate convoluted MS/MS data, making confident assignment of molecular structures challenging. The application of advanced data acquisition strategies such as SONAR can address these challenges by improving spectral clarity and reducing false positive identifications.

Study Objectives and Overview

The primary goal of this study was to demonstrate the advantages of the SONAR data independent acquisition (DIA) mode for the reliable identification of multiple co-eluting compounds in a traditional Chinese medicine formulation, Yu Ping Feng San (YPFS). YPFS consists of three herbs mixed in a defined ratio and represents a stringent test case due to its chemical complexity. A comparison was conducted between SONAR and a conventional DIA method (MSe) to assess improvements in spectral specificity and compound assignment.

Methodology

The YPFS formulation was prepared by decoction of Rhizoma Atractylodis Macrocephalae, Radix Astragali and Radix Saposhnikoviae in a 2:2:1 ratio. After extraction, the combined aqueous decoction was dried, reconstituted in methanol (10 mg/mL), centrifuged and filtered. Chromatographic separation employed a reversed-phase ACQUITY UPLC HSS T3 column (2.1 × 100 mm, 1.8 µm), using a typical gradient of aqueous and organic mobile phases. Mass spectrometric data were acquired on a Xevo G2-XS QTof system.

Instrumentation

• UPLC: ACQUITY UPLC HSS T3 Column (2.1 × 100 mm, 1.8 µm)
• Mass Spectrometer: Xevo G2-XS QTof operating in SONAR mode
• SONAR Settings: quadrupole isolation window 10 Da, scanning 50–2000 m/z at 0.1 sec per scan
• Comparison Method: Conventional DIA (MSe)

Main Results and Discussion

Representative base peak chromatograms revealed a dense profile of co-eluting constituents. For the marker compound prim-O-glucosylcimifugin, traditional DIA yielded 97 high-energy fragment ions, many arising from adjacent precursors and complicating identification. In contrast, SONAR provided a narrow precursor window and generated only eight relevant fragment ions specific to prim-O-glucosylcimifugin. This reduction in spectral complexity enhances confidence in matching fragments to the correct parent ion and reduces the risk of false positives.

Benefits and Practical Applications

• Improved spectral specificity facilitates accurate compound assignments in highly complex natural product matrices.
• Reduction in false positive identifications enhances reliability of non-targeted screening workflows.
• Compatibility with existing UPLC-QTof platforms enables straightforward adoption in QA/QC and research laboratories.

Future Trends and Opportunities

Emerging applications of SONAR may include lipidomics, metabolomics and proteomics analyses where complex isomeric and isobaric species co-elute. Integration with spectral libraries, machine learning-driven deconvolution and real-time data processing will further expand its utility. Narrower isolation windows and higher scan rates will continue to improve selectivity and throughput.

Conclusion

SONAR data independent acquisition significantly enhances the accuracy and confidence of compound identification in complex natural product samples. By isolating narrow precursor windows during fragmentation, SONAR reduces spectral congestion and false positive rates compared to conventional DIA methods, making it a powerful tool for advanced analytical workflows.

References

1. Chinese Pharmacopoeia Commission, Pharmacopoeia of the People’s Republic of China, Volume I, China Medical Science and Technology Press, 2015.
2. Chen L.G., Yen K.Y., Yang L.L., Determination of six bioactive components in Yu-Ping-Feng-San by high-performance liquid chromatography, J. Liq. Chromatogr. Relat. Technol., 22 (1999) 1149-1159.
3. Heywood D., Craven K., SONAR – Delivering MS/MS Data from a DIA Experiment, Waters White Paper 720006033EN, 2017.
4. Gethings L.A. et al., Lipid profiling of complex biological mixtures by LC/MS using a scanning quadrupole DIA strategy, Rapid Commun. Mass Spectrom., 31(19) (2017) 1599-1606.