Doctor of Philosophy in Cell and Developmental Biology (PhD)
Characterisation of novel targets of metabolic acid stress resistance in malignant cells
We currently have projects in the areas of genetic networks, cell signalling, membrane contact sites, cell polarity, cancer metabolism and autism. We use multiple model systems to study these topics including budding yeast for genetic network analysis, model human cell lines for cell signalling and microscopy, and knockout mouse genetic models for in vivo functional analysis.
G+PS regularly provides virtual sessions that focus on admission requirements and procedures and tips how to improve your application.
Membrane contact sites are sites of close apposition between two subcellular membranes that are believed to facilitate calcium signaling and lipid transport between organelles. The endoplasmic reticulum (ER) forms contacts with many organelles as well as the plasma membrane (PM). ER-PM contacts have been proposed to be held together by many families of protein tethers, and in the budding yeast Saccharomyces cerevisiae, proposed tethering proteins include the vesicle-associated membrane protein-associated protein homologues Scs2/22, the extended synaptotagmin homologues Tcb1/2/3, transmembrane protein Ice2, the lipin Pah1, and the transmembrane protein 16 (TMEM16) homologue Ist2. While these proteins are believed to function redundantly in creating ER-PM contacts, many of these proteins also possess functional modules such as lipid transfer domains, suggesting that they may also have specific functions in addition to tethering, with the identification of these functions being an area of current research. Here, by leveraging existing high-throughput datasets as a starting point, we identified Ist2 as a yeast ER-PM tether that participates in transport of the lipid phosphatidylserine (PS) from the ER to the PM through a physical interaction with the oxysterol-binding protein-related protein homologues and lipid transfer proteins Osh6 and Osh7. We found that Ist2 binds to both Osh6/7 through a binding site located in its disordered linker region between its ER-anchored TMEM16 domain and its C-terminal PM-binding helix. As well, we uncovered genetic evidence that Ist2 and Osh6/7 function in the same pathway as the PS decarboxylase Psd2, with loss of Ist2 or both Osh6/7 resulting in a strong growth defect and decreased cellular phosphatidylethanolamine levels when combined with loss of a redundant PS decarboxylase Psd1. Thus, these findings identify a physiologically relevant molecular link between members of the ORP and TMEM16 family of proteins and lays the groundwork for future studies to uncover the function of these proteins at membrane contact sites.
It is imperative for cell survival and function to maintain proper steady-state lipid levels, or lipid homeostasis. This has significant physiological consequences, as lipid homeostasis is disrupted in metabolic diseases including obesity and diabetes, which necessitates a greater understanding of this cellular phenomenon. The lipin family of phosphatidic acid phosphatases are conserved enzymes that control the cellular balance of phospholipid and triglyceride synthesis, and mammalian lipins can also regulate lipid synthesis through interacting with transcription factors in the nucleus. Unsurprisingly, lipins are tightly regulated enzymes and a conserved mechanism of lipin regulation is phosphorylation by kinases, which can control the subcellular localization of lipins from the cytoplasm to other cellular compartments. To date, various kinases have been identified that phosphorylate lipins including the mechanistic target of rapamycin complex 1 (mTORC1), which controls lipin 1 localization from the cytoplasm to the nucleus and the ability of lipin 1 to repress sterol regulatory element binding protein (SREBP) target-gene transcription and thus cholesterol and fatty acid biosynthesis and uptake. A high-throughput screen seeking novel kinase regulators of lipins has never been performed. In this work, we designed an overexpression screen in yeast and identified Mck1, a glycogen synthase kinase 3 (GSK3) kinase, as a novel regulator of lipins and lipid homeostasis. We further discovered that this relationship was conserved from yeast to mammals by characterizing that mammalian GSK3 phosphorylates lipin 1 directly. GSK3 activity, downstream of the PI3K/Akt pathway, towards lipin 1 was found to control its localization, and in the absence of GSK3 activity, lipin 1 translocated to the nucleus and repressed SREBP target-gene expression. We observed that regulation of SREBP-target gene expression in this pathway was dependent on lipin 1 and additionally that both GSK3 paralogs, GSK3α and GSK3β, appeared to be involved. Finally, we characterized the role of GSK3 in lipid metabolism using mouse models and found that mice lacking GSK3α or GSK3β in the liver demonstrated resistance to some effects of diet-induced obesity including weight gain and the expression of certain SREBP target genes, suggesting that GSK3 in the liver plays a role in the development of these phenotypes.
Membrane contact sites between the endoplasmic reticulum (ER) and other organelles are present in all eukaryotic cells. Their roles in calcium signaling and transport between the ER and the plasma membrane (PM) or the ER and mitochondria are quite well understood, but the molecular mechanisms underlying their roles in lipid synthesis and transport remains unknown. In order to identify the importance of organelle-ER contact sites, I used Saccharomyces cerevisiae - a model organism that has proven to be a particularly informative for studying lipid-related cellular processes. Previously, we found a role for an ER anchor protein, Scs2, being important for PM-ER contact sites. Further, SCS2 interacts genetically with ICE2, an ER gene with unknown function. In Chapter 2, I investigated a role for PM–ER contact sites in regulating phosphatidylcholine (PC) synthesis and I found that Δscs2Δice2 cells are choline auxotrophs and PM–ER contacts are required for PC synthesis. Osh2 and Osh3, the oxysterol-binding protein homologues in yeast, rescued the choline auxotrophy phenotype of Δscs2Δice2 cells but did not restore pmaER, indicating that they may function with Opi3 in PC synthesis. In search for regulators of pmaER, we identified the phosphatidic acid phosphohydrolase Pah1 that seems to be involved in establishing pmaER, independent of its enzymatic activity. Finally, we proposed that PE to PC synthesis by Opi3 happens “in trans” at PM-ER contacts. In Chapter 3, I aimed to discover novel genes involved in PE synthesis/traffic from ER to mitochondria. By doing a genome-wide screen for CHO2, we identified genetic interactions between CHO2 and Emc proteins indicating that Emc proteins are important for PE metabolism and we proposed that Emc facilitates PS transfer from the ER to mitochondria for PE synthesis. In Chapter 4, I investigated for roles of SCS2 in polarized growth. I found a physiologically important function of the ER diffusion barrier, which is to restrict diffusion of the spindle from mother to bud until M phase. Scs2 interacts directly with the spindle capture protein Num1 and it prevents Num1 from diffusing from the mother into the bud during S and G2 phases.
Recognition of membrane lipids by soluble proteins is important for the recruitment of these proteins onto membranes. Hence, changes in the concentration of these lipids affect the activity of these proteins, which alters downstream signal transduction pathways. Therefore, these lipids play significant roles as signaling molecules. Phosphatidic acid (PA) and phosphoinositides are signaling lipids that are present in all eukaryotes and are involved in the regulation of numerous critical cellular processes. The objective of this thesis was to identify new regulators and mechanisms of PA and phosphoinositide signaling through the utilization of the model eukaryote Saccharomyces cerevisiae. Based on a genome-wide screen to identify novel factors affecting PA signaling, the binding of proteins to PA was found to be dependent on intracellular pH (pHi) and the protonation state of its phosphomonoester headgroup. In yeast, a rapid decrease in pHi in response to glucose starvation regulated binding of PA to a transcription factor, Opi1, that coordinately repressed phospholipid metabolic genes. Hence, PA is a pH biosensor that enabled coupling of membrane biogenesis to nutrient availability (Chapter 2). Many phosphoinositides also possess phosphomonoesters on their headgroup that are sensitive to protonation within the physiological pHi range; including phosphatidylinositol 4-phosphate (PI(4)P) that is enriched in the trans-Golgi. Binding of Osh1, a member of the oxysterol-binding protein (OSBP)-related protein family (ORP), to late-Golgi PI(4)P was also found to be dependent on pHi and the protonation state of its phosphomonoester. Osh1 binding to Golgi PI(4)P regulated TORC1 (target of rapamycin complex 1) signaling and facilitated the expression of downstream genes involved in amino acid metabolism, which was inhibited by the release of Osh1 from Golgi PI(4)P due to pHi acidification. Hence, PI(4)P is a pH biosensor that regulates amino acid metabolism (Chapter 3). Together, these findings indicate that pHi is a signal that utilizes pH-sensing by lipids to regulate anabolism in yeast. A number of other potent signaling lipids also contain headgroups with phosphomonoesters, implying that pH sensing by lipids may be widespread in biology.
Polarization of cellular membranes into domains is an important mechanism tocompartmentalize cellular activities within the membrane and establish cell polarity.Recent studies have uncovered that the endoplasmic reticulum (ER) is polarized bydiffusion barriers, which in neurons controls glutamate signaling in dendritic spines, butthe molecular identity of these diffusion barriers is unknown. In Chapter 2 we show thata direct interaction between integral ER protein Scs2 and septin Shs1 creates the ERdiffusion barrier in yeast. We uncovered a new ER-associated polarisome subunit,Epo1, which is required for the tethering of ER to septins. The human homologue ofScs2, VAP-B, also interacts with Shs1 in yeast indicating that the tether may beconserved. As mutations in VAP-B cause amyotrophic lateral sclerosis, loss of ERpolarization in dendritic spines is a potential mechanism underlying motorneurondisease.Synthesis of phospholipids, sterols and sphingolipids is thought to occur atcontact sites between the ER and other organelles because many lipid synthesizingenzymes are enriched at contact sites. In only a few cases have the enzymes beenlocalized to contacts in vivo and in no instances have the contacts been demonstratedto be required for enzyme function. In Chapter 3 we show that plasma membrane (PM) -endoplasmic reticulum (ER) contact sites in yeast are required for phosphatidylcholinesynthesis and regulate the activity of a key enzyme, Opi3, whose activity requires a lipidbinding protein, Osh3. Thus, membrane contact sites provide a structural mechanism toregulate lipid synthesis.
Phosphomonoester-containing phospholipids are an important group of signaling lipids with pKa values in the physiological range. The protonation state of phosphomonoester headgroups can respond linearly to the surrounding pH, allowing for pH-sensing. Previous work has established phosphatidic acid (PA) as a pH biosensor. A similar function exists for phosphatidylinositol 4-phosphate (PI4P) in its pH-dependent binding to the yeast oxysterol-binding protein (OSBP) homolog Osh1, an interaction which regulates the localization of the high affinity tryptophan permease Tat2. To solidify PI4P as a pH biosensor, an understanding of its pH-sensing in a physiological context is necessary. Using glucose as a physiological signal, we demonstrate that the interaction between PI4P and Osh1 is regulated by glucose in a pH-dependent manner. We provide evidence that Tat2 is regulated by glucose through pH, and that Osh1 regulation of Tat2 localization is likely through the lipid counter-transport activity of Osh1 at ER-trans-Golgi network (TGN) contact sites. Thus, we propose a model where pH-biosensing by PI4P in response to glucose availability regulates Tat2 sorting through lipid counter-transport by Osh1.