Label-Free Imaging of Lipid Droplets in Prostate Cells Using Stimulated Raman Scattering Microscopy and Multivariate Analysis

microscopy is a powerful imaging modality for the analysis of biological systems. Here, we report the application of k-means cluster analysis (KMCA) of multi-wavelength SRS images in the high-wavenumber region of the Raman spectrum as a robust and reliable method for the segmentation of cellular organelles based on the intrinsic SRS spectrum. KMCA has been applied to the study of the endogenous lipid biochemistry of prostate cancer and prostate healthy cell models, while the corresponding SRS spectrum of the lipid droplet (LD) cluster enabled direct comparison of their composition. The application of KMCA in visualizing the LD content of prostate cell models following the inhibition of de novo lipid synthesis (DNL) using the acetyl-coA carboxylase inhibitor, 5-(tetradecyloxy)-2-furoic acid (TOFA), is demonstrated. This method identified a reliance of prostate cancer cell models upon DNL for metabolic requirements, with a significant reduction in the cellular LD content after treatment with TOFA, which was not observed in normal prostate cell models. SRS imaging combined with KMCA is a robust method for investigating drug???cell interactions in a label-free manner.

Automated summary

This study reports the application of k-means cluster analysis in microscopy for the segmentation of cellular organelles. By analyzing the SRS spectrum, the composition of different organelles can be directly compared. This method can be used for studying drug-cell interactions.