Mechanics of ESCRT-III catalyzed membrane fission – ESCRTfission

We use biochemical and molecular biology approaches combined with structural biology methods such as X-ray crystallography and SAXS to study the structure of CHMP2A and CHMP3 in its polymeric form in vitro. This will be complemented by cellular electron microscopy studies to understand the structure of ESCRT-III polymers in vivo and by in vitro reconstitution experiments of ESCRT-III polymers on Giant Unilamellar Vesicles (GUVs) and on membrane tubes pulled from GUVs with optical tweezers. This will provide insight into the kinetics of assembly and the forces associated with membrane remodeling.

We have collected a 3.5 Å dataset on the CHMP3 dimer, which represents a potential model for CHMP2A-CHMP3 heterodimers. We further characterized a number of llama VHH and their binding to dimeric CHMP3 and identified a few mutations that affect CHMP2A polymerization. We have extended the data on the analysis of prototype foamy virus (PFV) budding sites to the structural analysis of PFV particles and completed the in situ 3D reconstruction of the glycoprotein, which led to a recent publication in PLoS Pathogens. Regarding GUVs, we are now able to pull tubes to the inside of the GUV, thereby avoiding protein incorporation into GUVs and we have characterized ESCRT-III protein membrane binding using the Quartz Crystal Microbalance setup to determine optimal lipid compositions for ESCRT-III protein GUV interaction studies (manuscript under revision).

The project is ongoing.

Alqabandi M;, Miguet, N., Basserau, P. Bally, M;, Weissenhorn, W, and Mangenot, S. (2016) Interaction between ESCRT-III proteins and supported lipid bilayers, PhysChem ChemPhys (under revision).

Effantin G, Estrozi LF, Aschman N, Renesto P, Stanke N, Lindemann D, Schoehn G, and Weissenhorn W (2016) Cryo-electron Microscopy Structure of the Native Prototype Foamy Virus Glycoprotein and Virus Architecture. PLoS Pathog. 2016 Jul 11;12(7):e1005721

M. Kozlov, W. Weissenhorn, P. Bassereau, Membrane remodeling: theoretical principles, structures of protein scaffoldsand forces involved. In: From molecules to living organisms: an interplay between biology and physics. Editors: E. Pebay-Roula et al., Eds. (Oxford University Press (OUP), 2016).

The Endosomal sorting complex required for transport III (ESCRT-III) is recruited during multivesicular body formation, cytokinesis and budding of some enveloped viruses in order to catalyze the final step of membrane fission, which is common to these seemingly unrelated, but essential cellular and pathological processes. Although we and others have proposed models on ESCRT-III catalyzed membrane fission, no experimental mechanistic insight into this fundamental process is yet available. We propose to combine high and medium resolution structural studies to assemble a model of ESCRT-III with single molecule studies on membrane tubes pulled from giant unilamellar vesicles (GUVs). Our three main objectives are (1) to solve the crystal structure of dimeric CHMP3 that serves as a model for the repeating unit of ESCRT-III CHMP3-CHMP2A polymers, (2) to visualize ESCRT-III polymers in vivo at virus budding sites and (3) to reconstitute ESCRT-III proteins and VPS4 inside membrane tubes pulled from GUVs to understand the mechanics of fission. Objective 1 is advanced in that we have small crystals and co-crystallization tools such as llama VHH nanobodies in hand that will facilitate crystallization. Alternatively, if we fail to obtain crystals we will collaborate with the local NMR group to solve the structure by NMR. This structure will allow us to generate a quasi atomic model of CHMP3-CHMP2A polymers based on our medium resolution electron microscopy map. The final structure will provide important insight on ESCRT-III membrane interaction and polymer formation in vitro, two features which will be tested in vivo by structure-based mutagenesis approaches. The second objective is to visualize ESCRT-III in vivo at ESCRT-dependent enveloped virus budding sites by innovative electron microscopy approaches. We will combine state of the art cryo electron microscopy (CEM) alone and of vitreous sections (CEMOVIS) together with a novel approach to “freeze” ESCRT-III at late budding stages. Although it is challenging to visualize ESCRT-III in vivo, state of the art EM permitting the use of a low defocus (2 to 3 µM) will allow to image ESCRT-III filaments, which are at least 4 nm wide based on the EM analyses of ESCRT-III polymers assembled in vitro. In addition to image the wild type native ESCRT-III complex, our approach will allow us to systematically analyze the bud neck diameters, which will provide indirect information on the path of membrane neck constriction. Our 3rd objective is to develop new and innovative methodologies to reconstitute ESCRT-III proteins and the AAA-type ATPase VPS4 within membrane nanotubes pulled from GUVs. Although challenging, this correct physiological setting is absolutely required to study ESCRT-III function in vitro. After employing many different approaches we have established a protocol to incorporate ESCRT-III inside GUVs and pull membrane tubes. This will allow us now to determine the composition of the minimal fission machinery and study the ESCRT mechanics of membrane constriction and fission. Together our approaches will provide an integrated model of the molecular architecture of ESCRT-III that will be related to its function in membrane fission.