Mass spectrometry-based lipidomics is the primary tool for the structural analysis of lipids but the effective localization of carbon–carbon double bonds (C=C) in unsaturated lipids to distinguish C=C location isomers remains challenging. Here, we develop a large-scale lipid analysis platform by coupling online C=C derivatization through the Paternò-Büchi reaction with liquid chromatography-tandem mass spectrometry. This provides rich information on lipid C=C location isomers, revealing C=C locations for more than 200 unsaturated glycerophospholipids in bovine liver among which we identify 55 groups of C=C location isomers. By analyzing tissue samples of patients with breast cancer and type 2 diabetes plasma samples, we find that the ratios of C=C isomers are much less affected by interpersonal variations than their individual abundances, suggesting that isomer ratios may be used for the discovery of lipid biomarkers.

Electrospray ionization (ESI) has become a powerful tool for the analysis of biomolecules by mass spectrometry (MS). The process of ESI is difficult to control, and side reactions such as electrochemical reactions can occur during the ESI process because of the high voltages applied. Herein, a novel on-demand MS analysis method was developed based on discontinuous ion injection-induced ESI on a miniature MS system. Highly efficient ionization was enabled under low voltages (<300 V) using a discontinuous atmospheric pressure interface. On-demand ionization showed comparable sensitivity with regular nanoESI for the analyses of a series of compounds. It was found to be softer than regular ESI or nanoESI methods for ionization of proteins such as myoglobin and cytochrome C. As the ionization finished as soon as the interface was closed, the sample consumption was observed to reduce significantly for MS analysis, allowing single-cell analysis with multiple MS and MS/MS measurements.

The development of miniature mass spectrometry (MS) systems with simple analysis procedures is important for the transition of applying MS analysis outside traditional analytical laboratories. Here, we present Mini 14, a handheld MS instrument with disposable sample cartridges designed based on the ambient ionization concept for intrasurgical tissue analysis and surface analysis. The instrumentation architecture consists of a single-stage vacuum chamber with a discontinuous atmospheric interface and a linear ion trap. A major effort in this study for technical advancement is on making handheld MS systems capable of automatically adapting to complex conditions for in-field analysis. Machine learning is used to establish the model for autocorrecting the mass offsets in the mass scale due to temperature variations and a new strategy is developed to extend the dynamic concentration range for analysis. Mini 14 weighs 12 kg and can operate on battery power for more than 3 h. The mass range exceeds m/z 2000, and the full peak width at half-maximum is Δm/z 0.4 at a scanning speed of 700 Th/s. The direct analysis of human brain tissue for identifying glioma associated with isocitrate dehydrogenase mutations has been achieved and a limit of detection of 5 ng/mL has been obtained for analyzing illicit drugs in blood.

Mass spectrometry (MS) has become a powerful tool for metabolome, lipidome, and proteome analyses. The efficient analysis of multi-omics in single cells, however, is still challenging in the manipulation of single cells and lack of in-fly cellular digestion and extraction approaches. Here, we present a streamlined strategy for highly efficient and automatic single-cell multi-omics analysis by MS. We developed a 10-pL-level microwell chip for housing individual single cells, whose proteins were found to be digested in 5 min, which is 144 times shorter than traditional bulk digestion. Besides, an automated picoliter extraction system was developed for sampling of metabolites, phospholipids, and proteins in tandem from the same single cell. Also, 2 min MS2 spectra were obtained from 700 pL solution of a single cell sample. In addition, 1391 proteins, phospholipids, and metabolites were detected from one single cell within 10 min. We further analyzed cells digested from cancer tissue samples, achieving up to 40% increase in cell classification accuracy using multi-omics analysis in comparison with single-omics analysis. This automated single-cell MS strategy is highly efficient in analyzing multi-omics information for investigation of cell heterogeneity and phenotyping for biomedical applications.

No abstract