Label-Free Digital Quantification of Lipid Droplets in Single Cells by Stimulated Raman Microscopy on a Microfluidic Platform

Quantitative characterization of a single-cell phenotype remains challenging. We combined a scalable microfluidic array of parallel cell culture chambers and stimulated Raman scattering (SRS) microscopy to quantitatively characterize the response of lipid droplet (LD) formation to free-fatty-acid stimuli with single-LD resolution at the single-cell level. By enabling the systematic live-cell imaging with SRS microscopy in a microfluidic device, we were able to quantify the morphology of over a thousand live cells in 10 different chemical environments and with 8 replicates for each culture condition, in a single experiment, and without relying on fluorescent labeling. We developed an image processing pipeline for cell segmentation and LD morphology quantification using dual-channel SRS images. This allows us to construct distributions of the morphological parameters of LDs in the cellular population and expose the vast phenotypic heterogeneity among genetically similar cells. Specifically, this approach provides an analytical tool for quantitatively investigating LD morphology in live cells in situ. With this high-throughput, high-resolution, and label-free method, we found that LD growth dynamics showed considerable cell to cell variation. Lipid accumulation in nonadipocyte cells is mainly reflected in the increase of LD number, as opposed to an increase in their size or lipid concentration. Our method allows statistical single-cell quantification of the LD distribution for further investigation of lipid metabolism and dynamic behavior, and also extends the possibility to couple with other "omics" technologies in the future.