Poster S1 - Spatially correlated metabolic nanoscopy to study aging and diseases

Introduction: The signals of various molecular distributions and biochemical reactions at the sub-cellular organelle level are the most direct clues for detecting diseases. Cells and tissues are composed of various biomolecules, and these molecules undergo diverse chemical reactions in different sub-cellular organelles, such as OXPHOS in mitochondria and signal transduction on cellular membranes. These chemical reactions are crucial for proper bodily functions, including the generation of mechanical forces and chemical energies.
The perturbation of these chemical reactions can be linked to the development of diseases. To detect the changes in bio-chemical reactions, recent advancements in microscope techniques have allowed for the observation of a large number of biomolecules in nanoscopic regions. However, the resolving power is still insufficient to show the various chemical reactions and their correlations at the nanoscopic range of sub-cellular organelles. Conventional imaging techniques, such as MRI, PET, Fluorescence, and Mass Spectrometry, have inherent limitations, including spatial resolution. While super-resolution fluorescence microscopes have great power to spatially resolve molecules, the resolving capability of molecular components is limited.
To overcome these challenges, a super-resolution multimodal imaging platform combining stimulated Raman Scattering (SRS), multiphoton fluorescence (MPF), second harmonic generation (SHG), and super-resolution deconvolution algorithm has been developed to map the correlations between different metabolic activities in cells and tissues with subcellular organelle level. With additional algorithms to detect subcellular organelles and to calculate spatial correlations between different metabolic activities, the disturbed chemical reactions related to aging and diseases were detected.

Materials and

Methods: Imaging platform setup: Our custom-designed multimodal microscopy system has integrated stimulated Raman scattering (SRS), second harmonic generation (SHG), and multiphoton fluorescence (MPF) imaging modalities. Furthermore, we have integrated a new method that combines deuterium isotope probing and Stimulated Raman Scattering microscopy to visualize metabolic dynamics in live animals. The enzymatic incorporation of deuterium (D) into biomolecules generates carbon-deuterium (C-D) bonds in macromolecules. Within the broad vibrational spectra of C-D bonds, we discovered lipid-, protein-, and DNA-specific Raman shifts and developed spectral unmixing methods to obtain C-D signals with macromolecular selectivity.
Analysis: After acquiring the multimodal imaging dataset, we processed the data into super-resolution images using our custom-designed super-resolution deconvolution platform. We then generated quantitative ratiometric images displaying the spatial interactions and correlations between different metabolic activities, including redox ratio dynamics, lipid/protein turnovers, extracellular matrix changes, and lipid saturation heterogeneity. The correlations between each ratiometric image were mapped out using a spatial correlation coefficient calculation algorithm.
Sample: For aging study, from Drosophila, we have collected ovaries from two age groups: 5 days and 35 days. Additionally, we have collected xenograft cancer tissues from mice to study two different breast cancer subtypes, BT474 and HCC1806. All the samples were measured using the multimodal imaging setup, and the images were analyzed following the analysis workflow of super-resolution deconvolution algorithm, sub-cellular organelle detection, and spatial correlation analysis algorithm.

Results, Conclusions, and Discussions: Results
The measured multimodal label free image set includes spatial distribution signals from old proteins, old lipids, NADH, Flavin, newly synthesized proteins, newly synthesized lipids, unsaturated lipid, and saturated lipid channels.
Aging study: Ovaries from young and old Drosophila were imaged and analyzed. In nurse cells of the ovaries, the metabolic correlations between different signal channels were mapped and quantified for studying the metabolic changes during the aging processes.
Cancer metabolism study: Metabolic activities in two different types of breast cancer were imaged in situ and analyzed with spatial correlation coefficient calculation approach. The changes of the spatial correlations between the metabolic activities were significant only in extracellular matrix.
Discussions and conclusions
The spatially correlated super-resolution multimodal imaging method we have developed can provide nanoscopic spatial distribution information of metabolites and macromolecules simultaneously. Since the intensity values of each image are directly proportional to the concentrations of the respective components, the ratiometric images generated from this data are intrinsically linked to the specific rate constants of various biochemical reactions. Additionally, as some biochemical reactions share common molecular components, their spatial correlations should be discerned. To the best of our knowledge, no other technique has achieved the capability to image multiple metabolic reaction rate constants and their spatial correlations from the same field of view in a single focal plane.
To understand the mechanisms governing biological functions, the collective and integrated signals from this comprehensive approach are essential. This ability to visualize the spatial distribution of the correlated various metabolic reactions is both critical and novel. By leveraging this technology, we can define the characteristic biochemical signatures associated with specific disease states, such as cancer and aging. These disease-specific biochemical signals can then serve as robust standards to enable early detection and improved management of these debilitating conditions in clinical settings.

Acknowledgements (Optional):