Identification of gene products that control lipid droplet size in yeast using a high-throughput quantitative image analysis

Lipid droplets (LDs) are important organelles involved in energy storage and expenditure. LD dynamics has been investigated using genome-wide image screening methods in yeast and other model organisms. For most studies, genes were identified using two-dimensional images with LD enlargement as readout. Due to imaging limitation, reduction of LD size is seldom explored. Here, we aim to set up a screen that specifically utilizes reduced LD size as the readout. To achieve this, a novel yeast screen is set up to quantitatively and systematically identify genes that regulate LD size through a three-dimensional imaging-based approach. Cidea which promotes LD fusion and growth in mammalian cells was overexpressed in a yeast knockout library to induce large LD formation. Next, an automated, high-throughput image analysis method that monitors LD size was utilized. With this screen, we identified twelve genes that reduced LD size when deleted. The effects of eight of these genes on LD size were further validated in fld1 null strain background. In addition, six genes were previously identified as LD-regulating genes. To conclude, this methodology represents a promising strategy to screen for players in LD size control in both yeast and mammalian cells to aid in the investigation of LD-associated metabolic diseases. •An automated and quantitative LD morphology screen is established.•Novel imaging algorithm differentiates large LDs from small LD-clusters.•Yeast genome-wide non-biased screen for genes that reduce LD size•Cidea expression in yeast enables accurate LD size measurement.•Identification of eight genes that reduce LD size in fld1 deletion yeast strain