MALDI mass spectrometry imaging across spatial scales in toxicology: tissue evidence, cellular heterogeneity, and emerging single-cell opportunities.

Toxicological responses are inherently spatially heterogeneous, yet conventional analytical strategies generally rely on homogenized samples and therefore provide limited information on where xenobiotics accumulate, which tissue regions or cellular populations are preferentially affected, and how local molecular perturbations relate to pathology and mechanism. Mass spectrometry imaging (MSI) addresses this limitation by enabling label-free, multiplexed, and spatially resolved detection of xenobiotics, metabolites, lipids, peptides, and proteins directly in biological specimens. In this review, we examine the role of MSI in spatial toxicology, with particular emphasis on matrix-assisted laser desorption/ionization mass spectrometry imaging (MALDI-MSI) as a practical core platform for many current tissue-level and emerging cellular-scale toxicology studies. We first discuss why MSI is analytically well suited for spatial toxicology and compare the major MSI platforms in terms of molecular coverage, spatial resolution, and toxicological applicability. We then summarize representative applications at the tissue level, where MSI has already shown clear value in mapping xenobiotic localization, lesion-associated molecular remodeling, and organ-specific toxicity. Next, we discuss cellular-scale MSI and the transition toward true single-cell analysis, emphasizing that current progress is driven more by workflow development, multimodal integration, and computational advances than by routine toxicological deployment. Finally, using per- and polyfluoroalkyl substances as a focused case study, we illustrate how MSI can connect tissue burden, regional bioaccumulation, and localized biochemical response. Overall, MSI is becoming a key analytical framework for mechanistic spatial toxicology, although broader impact will depend on further advances in standardization, annotation confidence, quantification, and multimodal interpretation.