Super-resolution non-linear multimodal imaging platform for studying spatially correlated metabolic reactions.

Abstract

Metabolism is a collective biochemical process in living organisms, which involves numerous molecules and consists of various reaction steps. To understand the multi-step biochemical reactions composed of various components, it is thus essential to elucidate the correlations between different types of molecules, and the different molecules should be directly imaged in situ. Here, we established a nonlinear multimodal imaging platform that integrates stimulated Raman Scattering (SRS), multiphoton fluorescence (MPF), and second harmonic generation (SHG) for studying correlations between different metabolic activities in cells and tissues. We further developed an Adam-based Pointillism Deconvolution (A-PoD) algorithm to achieve super-resolution multimodal imaging of metabolism. Combining with heavy water-probing, we not only directly visualized metabolic changes of a variety of biomolecules in Drosophila ovaries during aging processes, but also distinguished the metabolic heterogeneity in triple-negative breast cancer tissues. The protein and lipid turnover rates, lipid unsaturation rate, and optical redox ratio were quantitatively analyzed in situ at subcellular scale, and the spatially correlated distributions of various metabolites were also studied by using a Pearson correlation coefficient mapping method. This super-resolved multimodal imaging platform, together with the analytic method enable us to quantitatively image collective molecular events in the same region of interest simultaneously. It offers powerful tools potentially for early stage disease detection, prognosis, and therapeutic effects, as well as for mechanistic understanding of scientific fundamentals in aging and biomedicine.