Structural and functional characterization of the Metallospherae sedula ESCRT-III-like proteins – ESCRT-ARC

Concerning CdvB, we tried to optimize the purification protocols in order to obtain higher amounts of soluble protein. Different CdvB paralogs as well as C-terminus deleted constructs were tested without success. Now, the next step will be to generate one-point-mutations in order to inhibit protein polymerization and to obtain a monodisperse sample. The design of new mutations is in progress.
Concerning CdvC, the same strategy was applied. This AAA-type ATPase expression of soluble proteins was optimized for crystallographic purpose. Among the strategies, a N-terminus deleted construct was cloned and expressed. The hexameric complexes gave crystals diffracting at 3.5 A resolution. However, under these conditions we cristallised an hexamer, i.e. one ring , and not the dodecamer. We will now generate new constructs to check if the linker will restore the association of the 12 sub-units.

Several constructs of CdvB were cloned and purified using various protocols. However, we are still not able to produce a soluble form of the recombinant protein in amount required for crystallization purpose.
The main result of these first 6 months is that we were able to solve the structure of the archaeal AAA-ATPase CdvC under its hexameric conformation (see figure below). Refinement is still in progress, but at this stage we can say that this structure corresponds to a dimer of trimers. This is the first X-ray crystallography data set obtained on Vps4-like proteins. We also determined nucleic acid concentrations required to promote the AAA-ATPase complex stabilization. Finally, we evidenced a relationship between the association of the two hexameric rings and salt concentration. On the overall, these data would help us to crystallise the dodecamer.

As mentionned above, for both proteins of interest, we will design and clone new mutants in order to improve the solubility of CdvB and to crystallise the dodecameric form of CdvC. We will also perform preliminary SAXS analysis of CdvB, an approach that requires much less protein than X-ray crystallography.

none

The Archaea constitutes one of three domains of life on Earth, along with Eukarya and Bacteria. Archaea comprise three major phyla Euryarchaeota, Crenarcheaota and Thaumarchaeota, that exhibit some critical differences in major molecular mechanisms, such as DNA replication and cell division. Thus, Euryarchaeota and Thaumarchaeota division employs a bacterial-like FtsZ, apparatus whereas for Crenarchaeota, where FtsZ is lacking, cytokinesis relies on a newly discovered Cdv (for Cell division) machinery. Importantly, two of the three Cdv proteins, CdvB and CdvC are homologous of eukaryotic proteins belonging to the “endosomal sorting complex required for transport” (ESCRT) machinery, which catalyzes vesicle budding during endosomal protein sorting, budding of some enveloped viruses and cytokinesis. CdvB is a homologue of ESCRT-III, implicated in membrane fission and CdvC is a homologue of the AAA-type ATPase Vps4, required for ESCRT-III disassembly. The third member of the Cdv cluster, namely CdvA, is only present in Archaea. Our own experiments support the hypothesis that CdvA could constitute an ancient cytoskeleton protein involved in cell division.
Several data reinforce the concept that archaeal Cdv proteins are closely related to the eukaryotic ESCRT-III counterparts including the presence of a MIT domain interacting motif (MIM) in the C-terminus of CdvB which is employed to interact with CdvC. This region of CdvB also contains a domain responsible for its interaction with CdvA, the CdvA/CdvB and CdvC/CdvB interactions being not mutually exclusive. Although interactions of Cdv proteins encoded by the Cdv gene cluster were established, nothing is known about the CdvB paralogs present in Crenarcheal genomes. Moreover, we lack functional, mechanistic and structural details showing how these proteins catalyze cytokinesis. These findings raise fascinating questions regarding the evolutionary history of cell division in Archaea and, on a more global extend, the mechanisms responsible for membrane remodeling.
In order to understand the structural basis of membrane remodeling by archaeal ESCRT-III we will pursue five main objectives. First, we will solve the crystal structure of full length CdvB. This will serve as a basis to dissect the structure and provide insight into its activation and polymerization properties. Secondly, we will dissect the polymerization propensity of CdvB and determine the structure of the repeating subunit of the polymer, which will provide important insight into membrane interaction and remodeling. Aims 1 and 2 will be extended to the CdvB homologues and characterized with regard to their interactions with CdvB and/or CdvC. The third aim is to study the mechanism of ESCRT-III disassembly by CdvC. Our main goal is to determine the crystal structure of the CdvC dodecamer in the presence or absence of nucleotides or analogues. This will ultimately allow to determine the conformational state of each protomer within the dodecamer and will thus provide essential new insight into the mechanism of AAA-type ATPases and ESCRT-III disassembly. A fourth goal is to study the effect of CdvB on membrane structure in collaboration with the Bassereau’s lab (Institut Curie, Paris) in order to directly visualize changes in membrane structure employing GUV-membrane tubes. Finally, insights from the structural studies will be complemented and validated by mutational studies in vivo to obtain further insight into the mechanism of archaeal cytokinesis (coll R. Bernander, Uppsala, Sweden).
This project will provide a comprehensive view of the structural basis of the archaeal membrane remodeling machinery leading to cytokinesis. Because of the importance of ESCRT-III during diverse eukaryotic processes, the archaeal work is performed in parallel to studies on the eukaryotic ESCRT machinery in the Weissenhorn’s lab and will thus add important complementary insight into ESCRT-III driven membrane remodeling processes.