Intelligent Algorithm System Tested for Natural Product Mass Spectrometry

Scientists from the Ganjiang Chinese Medicine Innovation Center in Nanchang, China recently tested a new method for conducting natural product mass spectrometry (MS). In this study, the scientists introduced MS-SMART, a system that integrates three intelligent algorithms: automatic mining of diagnostic ions, rapid filtration of alkaloids from untargeted MS/MS data, and structural recommendations for filtered components. Their findings were published in the Journal of Chromatography A (1).

New glass vials for instrumental analysis on a white background. analytical vials prepared for chemical analysis | Image Credit: © Elena - stock.adobe.com

Natural products (NPs) are small molecules naturally produced by any organism that contains primary and secondary metabolites (2). These products have diverse chemical structures and specific biological functions, with scaffolds and bioactivity inspiring the development of numerous new drugs. Traditional Chinese medicines (TCMs), which act as a rich repository of plant-derived NPs, show a broad range of pharmacological activities, becoming vital resources for potential drugs or lead compounds. Efficiently mining and identifying new compounds is essential for accelerating the discovery of lead compounds, with one of the most powerful tools for discovering new compounds in TCMs being liquid chromatography tandem high-resolution mass spectrometry (LC-HRMS); this is due to its superior separation capability and remarkably sensitivity. Regardless, the complexity of TCM chemical constituents has led to tens of thousands of mass spectra being generated, which can prove challenging for analyzing and identifying novel NPs.

Further, efficiently mining and identifying new compounds from extensive tandem mass spectrometry (MS/MS) data sets of plants has proven challenging, due to structural diversity and compositional complexity inherent in NPs. Various data post-processing techniques have been used to simplify MS/MS data interpretation; however, these often have limited specificity and precision. As for structure annotation following data post-processing, this can be especially time-consuming.

To show how feasible this approach was for rapidly discovering novel compounds, the scientists applied the system to berberine-type alkaloids. Berberine, an isoquinoline alkaloid derived from the roots and stem bark of the Berberis L. plant, is an antihyperglycemic drug that inhibits disaccharidase efficiency, reducing glucose transport through the intestinal epithelium (3). First, diagnostic ions were automatically extracted and validated using available reference data. Next, the berberine-type compounds were filtered from raw MS/MS data. Finally, the target components’ structures were recommended using building blocks derived from berberines found in various plants.

In total, 103, 198, 60, 80, and 51 berberines were efficiently identified in diverse families and genera, including Stephaniae Epigaeae Radix, Coptidis Rhizoma, Phellodendri Chinensis Cortex, Phellodendri Amurensis Cortex and Corydalis Decumbentis Rhizoma, with 99, 169, 50, 64 and 40 new compounds identified, respectively. Of these, 8, 14, 8, 7, and 12 berberines were confirmed by reference compounds, respectively.

With this study, the scientists successfully designed an intelligent strategy for efficiently mining and identifying new NPs using various algorithms. Using berberine-type alkaloids as an example, their strategy allowed for diagnostic ions to be automatically extracted from reference compounds, in addition to the selective filtering of berberines from complex plant extracts and intelligent annotation of the structures of the target components. Altogether, 422 new berberines were discovered in isoquinoline alkaloid-rich plants from different families. Because of this new strategy, experts can reduce their dependence on liquid chromatography–tandem mass spectrometry (LC–MS/MS) data processing, save time and resources, and significantly facilitate new compound discoveries. Further, this method is currently being tested for application in medicinal plant research, showing the beginnings of a new research paradigm for rapidly identifying NPs in complex samples.

References

(1) Yu, W.; Zheng, X.; Li, X.; Zhu, J.; Liu, H.; et al. An Algorithm-Driven Intelligent Mining and Identification Strategy for Natural Product Mass Spectrometry. J. Chromatogr. A 2024, 1734, 465288. DOI: 10.1016/j.chroma.2024.465288

(2) Natural Products Articles from Across Nature Portfolio. Nature Portfolio 2024. https://www.nature.com/subjects/natural-products (accessed 2024-10-3)

(3) Berberine. ScienceDirect 2023. https://www.sciencedirect.com/topics/chemistry/berberine (accessed 2024-10-3)