Molecular basis of the vectorial transport of sterol – SterTrans

We have set up assays to measure sterol and PI(4)P transport between model membranes in real-time with an unprecedented accuracy as well as various steps constituting such a process. For this, we have taken advantage of the fluorescence features of dehydroergosterol, a natural cholesterol analogue. In order to establish that OSBP has the ability to tether two distant membranes, we have used various biophysical approaches. This work has benefited from our expertise in purifying recombinant proteins and from our ability to mimic in a very accurate manner the lipid composition of cellular membrane. Structural analysis by X-ray crystallography of Osh4p has been crucial for our project. Other results, coming from in vitro approaches, have been corroborated by cellular observations using fluorescence and electronic microscopy. This combination of approaches allowed us to propose and validate novel mechanistic schemes and is one of the reasons that explains the success of the SterTrans project.

We have revealed that several ORP/Osh can guarantee an unexpected exchange between sterol and PI(4)P. Our finding is an important breakthrough as it provides completely novel models on lipid transport. PI(4)P is a singular lipid, which is continually synthesized in specific organelles and consumed in the endoplasmic reticulum where the synthesis of various lipids takes place, such as sterol. ORP/Osh proteins might exploit this difference of PI(4)P levels to transfer sterol unidirectionnally from the endoplasmic reticulum towards other compartments by lipid exchange, thereby promoting the accumulation of sterol in these compartments. OSBP, with higher structural complexity, would ensure a similar task in specific subcellular regions in which the endoplasmic reticulum is in close apposition with the trans-Golgi. Our models beautifully explain how proteins can ensure the precise distribution of sterol inside the cells, notably the maintenance of a sterol gradient between the endoplasmic reticulum and the border of the cell.

Discovering the existence of sterol/PI(4)P exchangers has radically changed our vision on sterol transport. Based on recent literature, our new hypothesis is that certain ORP/Osh proteins use PI(4)P gradients not to convey sterol but other types of lipids. Thus we aim to examine in the future to which extent the PI(4)P-driven exchange mechanisms is a general cellular strategy to convey lipids. Interestingly various classes of viruses, once inside cells, trigger strong disturbance in Golgi lipid composition, notably a burst in PI(4)P production, to create platforms for the replication of their genetic material. Likewise, some oncogenesis route are possibly linked to disturbance in PI(4)P metabolism. OSBP seems involved in such processes and constitutes a target for antiviral and anticancer compounds. We are currently developing internal projects, or in collaboration, in order to understand how, thus having opportunities to start translational projects.

At the end of 2011, we published a first article about the finding that Osh4p exchanges sterol for phosphatidylinositol between membranes in the internationally renowned Journal of Cell Biology. Our work has been immediately acknowledged by a commentary paper and almost sixty articles during the 2012-2014 period. Our most recent work on OSBP, published in the journal Cell sheds light on more than twenty years of cellular observation by providing for the first time a relevant mechanism for this protein. A third manuscript demonstrating the ability of Osh4p to create a sterol gradient by Osh4p is currently in revision. We have presented posters and gave talks in about ten international meetings about the work deriving from the SterTrans project and have been invited to write review article about sterol and lipid transport (Annual Review in Biochemistry, CMLS, Bioc. Soc. Trans). Our lab has therefore gained a high visibility in the field of lipid metabolism and transport.

Our project will concern a large family of sterol transporters or sterol sensors for which a description in term of molecular mechanism is lacking. It is known that a proper distribution of sterol in cells is crucial for its viability. A key fact is that sterol is rare in the endoplasmic reticulum whereas it is far more concentrated in trans-Golgi membrane and plasma membrane. Recently, proteins called ORP/OSH appeared as the most serious candidates for ensuring a non-vesicular convey of sterol between organelles. By multiple approaches, we want to unveil information on these intriguing proteins and address a central issue in cell biology: how a protein can ensure a vectorial routing of lipids from one organelle to an other.
We will work on Kes1p/Osh4p, a cytoplasmic ORP/OSH protein whose structure is known. Kes1p is thought to convey sterol and help creating in yeast, a sterol gradient. Likely, Kes1p ensures a single route of transport. Which one? We think that its structure encodes a capacity to guaranty a one-way transport of ergosterol from endoplasmic reticulum to trans-Golgi. First, Kes1p integrates a curvature-sensing ability adequate to target the highly tubulated RE and extract newly synthesized sterol. Once extracted, one sterol is secured inside Kes1p by a lid. We observed that Kes1p releases sterol upon binding to PI(4)P. As this rare lipid pinpoints the trans-Golgi, it likely helps Kes1p to specifically supply this organelle with sterol. Thus, a lid unlocked only by PI(4)P could be a mechanism that prevents sterol from coming back to its original compartment. In other words, we have first with Kes1p a remarkable model to study how a protein is structurally designed to ensure a vectorial transport between two membrane-bound compartments. Secondly, we will recapitulate this vectorial transport in vitro with the most realistic lipid composition. In particular, we aim to observe how sphingolipids in PI(4)P-containing, Golgi-mimicking giant liposomes help Kes1p to work by trapping sterol into membrane domain,
In parallel, we aim to determine for the first time with ORP9L and OSBP the interplay between a membrane tethering activity and a sterol transport/sensing ability. In eukaryotic cells, zone of close appositions exist between the trans-Golgi and the endoplasmic reticulum. Diverse transporters are thought to convey lipids across this narrow space and to participate in controlling tightly membrane composition along the secretory pathway. In this context, the architecture of OSBP and ORP9L is fascinating. In addition to a sterol-binding domain akin to Kes1p, they contain two targeting elements: a N-terminal PH domain recruited on Golgi membrane both by PI(4)P and the small G protein Arf1 and a FFAT motif recognized by the ER-resident receptor VAP. Studies in cell have been providing precious indications on OSBP without revealing its mode of action and even its primary functional role. One hypothesis is that OSBP bridges ER and Golgi membrane in response to a signalling sterol and orchestrate in this transit zone the recruitment of other LTPs with similar dual-targeting determinants. At opposite, ORP9L is likely a sterol transporter. We aim to know if ORP9L sandwiched between two membranes could efficiently convey sterol and if tethering is advantageous for transport. To sum up, by experimentally correlating in vitro a tethering activity to a sterol-binding ability, we aim to address issues emerging in cell biology about these two proteins.
Last, several ORP/OSH proteins seem to be involved in coupling various cellular processes to sterol homeostasis. They may correspond to potential pharmacological targets. Drug design and virtual screening of the known structure of Kes1p in combination with our experimental tools will provide a way to design molecules interfering with this class of proteins.