"Mechanistic insights into protein-mediated lipid transfer" by Zhouping Hong

Date of Award

Spring 2023

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Cell Biology

First Advisor

Reinisch, Karin

Abstract

In eukaryotes, most lipids are synthesized in the endoplasmic reticulum (ER) and then redistributed from there to other organelles, each of which has a distinct membrane lipid composition. We have long known that lipids can be trafficked between organelles by vesicles. More recently, it has become clear that there is another major means of lipid redistribution mediated by lipid transport proteins (LTPs) that localize to so-called membrane contact sites (MCSs), where two organelles come into close apposition. LTPs have a hydrophobic cavity to solubilize the lipid fatty acyl tail. They work as either shuttles that ferry 1-2 lipid molecules at a time between membranes, not necessarily leading to net transfer, or bridges that form conduits allowing for bulk lipids to flow between membranes and net transfer.My first project was to characterize a new MCS protein, mitoguardin-2 (MIGA2), which localizes to either lipid droplet (LD)-mitochondria or ER-mitochondria contacts. Although MIGA2 was shown to support LDs expansion and maintain mitochondrial integrity, its molecular function and how it regulates these organelles were mysterious. I determined the crystal structure of the C.elegans MIGA soluble portion, showing that the C-terminal domain of MIGA2 has a hydrophobic cavity that binds lipids. Mass spectrometry analysis reveals that both glycerophospholipids (GPL) and fatty acids co-purify with MIGA2 from cells, and each MIGA2 can accommodate up to 2 lipids. Using in vitro transfer assays, I found that MIGA2 transfers GPLs between liposomes as well as between liposomes and artificial lipid droplets, and the lipid transfer ability of MIGA2 is required to preserve mitochondria morphology and lipid droplet formation. Collectively, this study established MIGA2 as a new shuttle-like LTP, which could transfer phospholipids at ER-mitochondria contact sites and phospholipids or fatty acids at mitochondria-LD contacts, and whose lipid transfer ability is critical for its function in cells, both for mitochondria and LDs. For the second part of my thesis, I focused on the bridge-like LTPs, particularly VPS13. VPS13 proteins are proposed to function at contact sites as bridges for lipids to move directionally and in bulk between organellar membranes. The molecular mechanisms of VPS13-mediated lipid transfer, however, are still unclear: 1) how is the direction of lipid flow regulated at MCSs? 2) what is the mechanism for rapid lipid extraction and insertion? 3) how do Vps13s support rapid membrane expansion during organelle biogenesis? One intriguing idea is that Vps13 proteins could collaborate with integral membrane proteins for channeling lipids. In particular, as VPS13s transfer lipids between only the cytosolic leaflets at apposed membranes, membrane expansion via bulk lipid transfer requires a mechanism to re-equilibrate lipids between the leaflets to prevent bilayer asymmetry. VPS13s are anchored between membranes via interactions with receptors, including both peripheral and integral membrane proteins. In yeast, Mcp1p is the only multi-span integral membrane protein so far known to interact directly with Vps13. In humans, an integral membrane protein, XK, interacts with VPS13A. Using in vitro reconstitution and biochemical assays, I demonstrated that yeast Mcp1p and human XK are scramblases. My finding supports the model of a partnership between bridge-like bulk lipid transport proteins and scramblases.

Share

COinS