Physics-based oligomeric models of the yeast mitofusin Fzo1 at the molecular scale in the context of membrane docking

Abstract

Tethering and homotypic fusion of mitochondrial outer membranes is mediated by large GTPases of the Dynamin-Related Proteins family called the mitofusins. The yeast mitofusin Fzo1 forms high molecular weight complexes and its assembly during membrane fusion likely involves the formation of high order complexes. Consistent with this possibility, mitofusins form oligomers in both cis (on the same lipid bilayer) and trans to mediate membrane attachment and fusion. Here, we rely on our recent Fzo1 model to investigate and discuss the formation of cis and trans mitofusin oligomers. We have built 3 cis-assembly Fzo1 models that gave rise to 3 distinct transoligomeric models of mitofusin constructs. Each model involves two main components of mitofusin oligomerization: the GTPase and the trunk domains. The oligomeric models proposed in this study were further assessed for stability and dynamics in a membrane environment using a coarse-grained molecular dynamics (MD) simulation approach. A narrow opening 'head-to-head' cisoligomerization (via the GTPase domain) followed by the antiparallel 'back-to-back' transassociations (via the trunk domain) appears to be in agreement with all the available experimental data. More broadly, this study opens new possibilities to start exploring cis and trans conformations for Fzo1 and mitofusins in general but also for other fusion-DRPs.

Mitochondrial Surveillance by Cdc48/p97: MAD vs. Membrane Fusion

“…Interestingly, these regions map to mutations associated with CMT2A [204], suggesting pathogenic relevance. Their importance as a driving force for membrane merging could be confirmed with subsequent functional analysis, allowing further dissection of the fusion mechanism [75,76,194,195,[205][206][207][208][209][210][211][212]. Consistent with the self-assembly properties of DRPs, mitofusins can be found in multiple oligomerization states [184,207] (Figure 4C).…”

“…This raised the hypothesis that Cdc48 action on Fzo1 could be similar to the role of the AAA-ATPase N-ethylmaleimide sensitive fusion protein (NSF) in vesicle fusion, by disassembling SNARE complexes after completed membrane merging [233]. As mentioned above, the mitochondrial tethering complex consists of oligomers of at least three or four subunits, likely an even larger amount of Fzo1 molecules [207,214,216]. The conformational changes in Fzo1, induced by GTP hydrolysis, might allow the tethering complex to expand to the docking ring, leading to increased membrane tension in the narrow mitochondrial fusion side [216].…”

“…Interestingly, these regions map to mutations associated with CMT2A [ 204 ], suggesting pathogenic relevance. Their importance as a driving force for membrane merging could be confirmed with subsequent functional analysis, allowing further dissection of the fusion mechanism [ 75 , 76 , 194 , 195 , 205 , 206 , 207 , 208 , 209 , 210 , 211 , 212 ].…”

Structural dataset from microsecond-long simulations of yeast mitofusin Fzo1 in the context of membrane docking

“…Topologies to run coarse-grained (CG) simulations were generated with the martinize tool choosing the martini v.2.1 force field with an elastic network [6], [7] The initial all atom coordinates were obtained from the published models in the related research paper [1]. The force bond constant was set to 500 kJ mol −1  nm −2 with lower and upper elastic bond cutoffs of 0.5 and 0.9 nm, respectively.…”

“…The colouring was made to highlight the functional domains of the protein following the same scheme as in our original paper [1]. The domains highlighted by colour are: violet, HRN; green, HR1; orange, HR2; red, GTPase and yellow: transmembrane.…”