Periodic Reporting for period 1 - MEMTOR (Understanding mechanisms of membrane tension loss and recovery using small-molecule tools)
Okres sprawozdawczy: 2021-05-01 do 2023-04-30
The plasma membrane is a network of lipids and proteins that serves as a physical barrier between the external environment and the interior of the cell. The maintenance of plasma membrane tension homeostasis is vitally important to prevent cancer metastasis, neurodegeneration, and metabolic syndrome. However, the study of this important process has lagged due to a lack of tools to measure and manipulate the behavior of lipids in the context of cellular membranes.
Starting from the observation that a small molecule, palmitoylcarnitine (PalmC), discovered in a screen in our lab, is able to induce membrane tension loss and plasma membrane invaginations, and inhibit the activity of the signaling complex that serves as a master regulator of membrane tension (TORC2), our aim was to characterize the mechanism by which PalmC alters plasma membrane tension (Aim 1). Our studies have revealed that PalmC, in addition to other small amphipathic molecules, acts directly on the plasma membrane and achieves its effects in a sterol-dependent manner.
During the course of this work, we also made a serendipitous discovery while purifying native TORC2 protein from yeast cells (Aim 2). We isolated a membrane-bound structure of the eisosome, a unique plasma membrane microdomain found in yeast which senses membrane stress and initiates signaling to TORC2. These isolated native eisosomes, scaffolded by the BAR-domain proteins Pil1 and Lsp1, were bound to a plasma membrane bilayer (See figure). By solving cryoEM structures of these native eisosomes, we were able to observe that the bound membrane has a remarkably well-organized structure with signatures of specific lipid species discernable within the membrane bilayer. These native eisosomes and their high-resolution structures have provided us a unique window into the organization of the lipids within a membrane microdomain.
Starting from the observation that a small molecule, palmitoylcarnitine (PalmC), discovered in a screen in our lab, is able to induce membrane tension loss and plasma membrane invaginations, and inhibit the activity of the signaling complex that serves as a master regulator of membrane tension (TORC2), our aim was to characterize the mechanism by which PalmC alters plasma membrane tension (Aim 1). Our studies have revealed that PalmC, in addition to other small amphipathic molecules, acts directly on the plasma membrane and achieves its effects in a sterol-dependent manner.
During the course of this work, we also made a serendipitous discovery while purifying native TORC2 protein from yeast cells (Aim 2). We isolated a membrane-bound structure of the eisosome, a unique plasma membrane microdomain found in yeast which senses membrane stress and initiates signaling to TORC2. These isolated native eisosomes, scaffolded by the BAR-domain proteins Pil1 and Lsp1, were bound to a plasma membrane bilayer (See figure). By solving cryoEM structures of these native eisosomes, we were able to observe that the bound membrane has a remarkably well-organized structure with signatures of specific lipid species discernable within the membrane bilayer. These native eisosomes and their high-resolution structures have provided us a unique window into the organization of the lipids within a membrane microdomain.
To understand the action of PalmC on membranes, we observed additional similar small amphiphilic molecules with slightly differing chemical structures and found that some are capable of causing similar effects to PalmC. Lipidomic analysis of our PalmC-resistant strain, initially identified by a SAturated Transposon Analysis in Yeast (SATAY) screen, suggested that this strain has a modified plasma membrane vulnerability to PalmC. Finally, we found that addition of PalmC alters sterol dynamics and that addition or reduction of membrane sterol content altered the sensitivity of (m)TORC2 activity to PalmC. Collectively, these results led us to conclude that PalmC acts directly on the plasma membrane.
During the process of biochemical optimization of native purification of TORC2 (Aim 2), we discovered large filamentous structures of yeast eisosomes bound to native plasma membrane (See figure). Solving the atomic resolution structures of these filaments by helical reconstruction enabled us to observe a unique patterning of membrane voids in the protein-bound cytoplasmic leaflet beneath an amphipathic helix of the Pil1/Lsp1 proteins, as well as lipid headgroups bound within a charged binding pocket. Using in vitro reconstitution with lipid mixtures of known composition and purified Pil1 protein, we were able to assign lipid identities to each of the signatures we saw within the membrane density, which we could further verify with molecular dynamics simulations. We could also observe changes in the dynamics of lipids in membranes bound by Pil1 protein, validating that these proteins indeed mediate the formation of a membrane microdomain. Finally, we observed a dynamic stretching within the Pil1/Lsp1 lattice, observed by 3D variability analysis, that correlates with a loss of bound lipid headgroups and the loss of this pattern of voids within the cytoplasmic leaflet.
During the process of biochemical optimization of native purification of TORC2 (Aim 2), we discovered large filamentous structures of yeast eisosomes bound to native plasma membrane (See figure). Solving the atomic resolution structures of these filaments by helical reconstruction enabled us to observe a unique patterning of membrane voids in the protein-bound cytoplasmic leaflet beneath an amphipathic helix of the Pil1/Lsp1 proteins, as well as lipid headgroups bound within a charged binding pocket. Using in vitro reconstitution with lipid mixtures of known composition and purified Pil1 protein, we were able to assign lipid identities to each of the signatures we saw within the membrane density, which we could further verify with molecular dynamics simulations. We could also observe changes in the dynamics of lipids in membranes bound by Pil1 protein, validating that these proteins indeed mediate the formation of a membrane microdomain. Finally, we observed a dynamic stretching within the Pil1/Lsp1 lattice, observed by 3D variability analysis, that correlates with a loss of bound lipid headgroups and the loss of this pattern of voids within the cytoplasmic leaflet.
Our studies with PalmC reveal new information on how this molecule and its amphiphilic relatives act on membranes. We showed that PalmC acts directly by inserting into membranes and identified an unanticipated role of sterol in mediating these membrane-altering effects. In addition, the crosstalk between plasma membrane sterol content and PalmC is valuable information for both gaining a mechanistic understanding of its mode of action and for tailoring any potential clinical utilization of these molecules.
The composition, organization, and very existence of so-called “lipid rafts” or membrane microdomains has remained controversial for over 50 years. Our eisosome structures contribute fundamental knowledge by giving us an unprecedented look at the structure of a native membrane microdomain. We have provided mechanistic detail into the membrane lipid dynamics that occur within the specialized membrane stress-sensing scaffold structure of the eisosome in near-atomic detail. In addition, the eisosomes have a functional relative in humans: the stress-sensing caveoli. While the proteins and organization of these mammalian membrane microdomains is not conserved, it is likely that the principles of stress-sensing (e.g. amphiphathic helices inserted into the membrane, altered lipid dynamics at the site of a membrane-bound protein scaffold) may be similar in these structures.
The composition, organization, and very existence of so-called “lipid rafts” or membrane microdomains has remained controversial for over 50 years. Our eisosome structures contribute fundamental knowledge by giving us an unprecedented look at the structure of a native membrane microdomain. We have provided mechanistic detail into the membrane lipid dynamics that occur within the specialized membrane stress-sensing scaffold structure of the eisosome in near-atomic detail. In addition, the eisosomes have a functional relative in humans: the stress-sensing caveoli. While the proteins and organization of these mammalian membrane microdomains is not conserved, it is likely that the principles of stress-sensing (e.g. amphiphathic helices inserted into the membrane, altered lipid dynamics at the site of a membrane-bound protein scaffold) may be similar in these structures.