Comparison of liquid chromatography-high-resolution tandem mass spectrometry (MS2) and multi-stage mass spectrometry (MS3) for screening toxic natural products

Ruben Yiqi Luo; Kate Comstock; Caroline Ding; Alan H B Wu; Kara L Lynch

doi:10.1016/j.jmsacl.2023.09.002

Comparison of liquid chromatography-high-resolution tandem mass spectrometry (MS²) and multi-stage mass spectrometry (MS³) for screening toxic natural products

J Mass Spectrom Adv Clin Lab. 2023 Sep 30:30:38-44. doi: 10.1016/j.jmsacl.2023.09.002. eCollection 2023 Nov.

Authors

Ruben Yiqi Luo¹, Kate Comstock², Caroline Ding², Alan H B Wu³, Kara L Lynch³

Affiliations

¹ Department of Pathology, Stanford University, Stanford, CA, USA.
² Thermo Fisher Scientific, San Jose, CA, USA.
³ Department of Laboratory Medicine, University of California San Francisco, San Francisco, CA, USA.

Abstract

Background: Liquid chromatography-high-resolution mass spectrometry (LC-HR-MS) has emerged as a powerful analytical technology for compound screening in clinical toxicology. To evaluate the potential of LC-HR-MS³ in detecting toxic natural products, a spectral library of 85 natural products (79 alkaloids) that contains both MS² and MS³ mass spectra was constructed and used to identify the natural products. Samples were analyzed using an LC-HR-MS³ method and the generated data were matched to the spectral library to identify the natural products.

Methods: To test the performance of the LC-HR-MS³ method in different sample matrices, the 85 natural product standards were divided into three groups to separate structural isomers and avoid ion suppression effects caused by co-elution of multiple analytes. The grouped analytes were spiked into drug-free serum and drug-free urine to produce contrived clinical samples.

Results: The compound identification results of the 85 natural products in urine and serum samples were obtained. The match scores using both MS² and MS³ mass spectra and those using only MS² mass spectra were compared at 10 different analyte concentrations. The two types of data analysis provided identical identification results for the majority of the analytes (96% in serum, 92% in urine), whereas, for the remaining analytes, the MS²-MS³ tree data analysis had better performance in identifying them at lower concentrations.

Conclusion: This study shows that in comparison to LC-HR-MS (MS²), LC-HR-MS³ can increase the performance in identification of a small group of the toxic natural products tested in serum and urine specimens.

Keywords: Clinical toxicology; LC-HR-MS3; Natural products; Spectral library.