Cellular heterogeneity is inherent within biological systems, yet conventional metabolomic analyses typically rely on bulk measurements that obscure differences between individual cells. Elucidating single-cell metabolic variation across a population is particularly important in host-pathogen interactions, where infected cells and neighbouring uninfected bystander cells may exhibit distinct metabolic responses. This thesis describes the development, optimisation and application of novel mass spectrometry (MS)-based workflows for spatially resolved single-cell metabolomics, with the aim of investigating macrophage responses to mycobacterial infection.

The development of single-cell metabolomics workflows is presented in this thesis as three journal articles, the first of which has been peer-reviewed and published. The first article establishes methodology for single-cell metabolomics using nano-electrospray ionisation (nano-ESI)-MS and liquid chromatography (LC)-MS, revealing that, while nano-ESI-MS enabled rapid single-cell analysis, LC-MS provided significantly greater metabolic coverage. Using microscopy to preserve spatial context, Mycobacterium bovis Bacillus Calmette-Guérin (BCG) infected macrophages were selectively aspirated from live cultures and compared to unexposed control cells and blank samples. LC-MS showed improved discrimination between the groups and was therefore established as the primary analytical platform for subsequent work.

In the second article, microflow LC-MS methods were developed to improve analyte sensitivity, metabolite coverage and annotation confidence. Two complementary chromatographic methods significantly expanded the metabolite coverage. Application of these workflows to BCG-infected macrophage cultures enabled direct comparison of infected cells, uninfected neighbouring bystanders, and unexposed control macrophages to test the hypothesis that infection induces a bystander effect. Infection-associated metabolic differences were observed, including pathway-level changes in both infected and bystander cells relative to controls, supporting our hypothesis that neighbouring uninfected macrophages undergo metabolic shifts in response to infection. In the third article, further microflow LC-MS optimisation (improved chromatographic separation, ion transfer, and precursor ion scheduling) addressed key analytical challenges associated with single-cell metabolomics, and established the methods suitable for future studies of infection-induced heterogeneity across a cellular population.

Overall, this thesis establishes LC-MS-based spatial single-cell metabolomics as a powerful tool for answering important biological questions.