Decoding of inter-organelle phospholipid transport by fluxome profiling

About This Grant

Project Summary/Abstract Cells continuously synthesize, degrade, and redistribute phospholipids. While vesicular trafficking was once considered the mechanism of this redistribution, recent studies have highlighted the critical roles of lipid transfer proteins (LTPs) in various contexts. For example, large quantities of phospholipids must be supplied to repair damaged lysosomal membranes or to form autophagosomes during starvation. However, it remains unclear why cells expend substantial energy to transport phospholipids even under normal physiological conditions. A major barrier to addressing this question has been the lack of methods to directly and comprehensively measure phospholipid transport in vivo without introducing bulky tags that perturb native lipid behavior. To circumvent this challenge, Aim 1 will establish a method to comprehensively profile inter-organelle phospholipid transport, referred to here as a “fluxome profile,” starting with phosphatidylcholine (PC). Using phospholipase D (PLD), I will perform organelle-specific pulse labeling by converting PC into deuterated PC (d- PC). After a chase period, d-PC distribution will be measured by stimulated Raman scattering (SRS) imaging, which detects vibrational modes of subcellular chemical bonds as well as the carbon–deuterium bond. This minimally perturbative label enables the observation of natural PC transport under physiological conditions. In parallel, I will create a “lipoprint map” by reclassifying organelles based on lipid-derived SRS spectral fingerprints. By projecting d-PC pulse-chase data from various organelles onto this map, I will obtain the fluxome profile, a comprehensive landscape of inter-organelle relationships defined by lipid composition and transport dynamics. In Aim 2, I will extend this strategy to other phospholipid classes by engineering PLD to alter its substrate specificity. This will involve three approaches: (i) an in vivo strategy that converts cellular PLD activities toward two different substrates into spectrally separated fluorescent signals for high-throughput screening; (ii) an in vitro method to pulldown PLD by phospholipids to enrich PLD sequences with desired substrate specificity; and (iii) an in silico deep learning model that predicts PLD activity for virtual screening. These approaches will enable selective deuterium labeling of major phospholipid classes and allow construction of fluxome profiles beyond PC. In Aim 3, I will first establish a method for the absolute quantification of organelle phospholipids. These data will enable quantitative comparison of phospholipid transport when different organelles serve as the pulse- labeling origin, providing the basis for a quantitative network model of inter-organelle phospholipid transport. I will then assess the structural robustness of this network model and experimentally validate it through LTP overexpression or knockout and engineered PLD-based perturbations. These analyses will test the hypothesis that the lipid transport network alone is a sufficient and efficient mechanism for maintaining organelle phospholipid compositions, uncovering fundamental design principles by which cells preserve membrane homeostasis under normal physiological conditions.