Final Report Summary - CRISTOPA1 (Extending the Optic atrophy 1 dependent cristae remodeling: from models to a rationale for therapy of autosomal dominant optic atrophy)
Mitochondria are key organelles for the life and death of the cells. They represent the metabolic hub where the energy intake from diet is converted into ATP, required to support all endoergonic processes of the cell; they participate in pivotal anabolic and catabolic reactions; they regulate the amplitude and the shape of intracellular Ca2+ signals; and last but not least, they are crucial components of the cell death cascade, where they integrate and amplify intrinsic and extrinsic signals by releasing into the cytoplasm protein cofactors that are normally stored inside the organelle. The momentum to the research into mitochondrial shape changes during apoptosis was greatly stimulated by the discovery of a set of proteins that regulate the fusion-fission equilibrium of the membranes of the organelle. Similarly, it also gained boost from the discovery that at least two important genetic diseases are associated with mutations in the proteins belonging to the core machinery that shapes the organelle, Autosomal Dominant Optic atrophy (ADOA) and the peripheral neuropathy Charcot-Marie-Tooth 2a.
We therefore devised a research program to verify the hypothesis that by engaging in interactions with different partners, OPA1 regulates mitochondrial functions from the cristae. Its mutations increase autophagy and susceptibility to apoptosis, especially in RGCs. These multiple regulatory points offer several potential targets for therapeutic strategies that can interfere with the natural course of the disease. Our project has been structured in the following specific aims (SA), further organized in a sequence of milestones:
SA1: address the relevance of OPA1 interactors in mitochondrial metabolism, autophagy and apoptosis.; SA2: address the molecular mechanism linking mutated/reduced OPA1 to zonal autophagy in RGCs; SA3: counteract OPA1 dysfunctions
We progressed towards each of these aims, in particular:
1. address the relevance of OPA1 interactors in mitochondrial metabolism, autophagy and apoptosis. We addressed the relationship between Opa1, cristae shape and bioenergetics by a combination of biochemistry, genetics and mitochondrial physiology. Within the cristae respiratory chain complexes assemble into functional quaternary structures called supercomplexes (RCS). In our Cogliati et al., Cell 2013 paper, we reported the results of our investigation on the relationship between respiratory function and Opa1-controlled mitochondrial ultrastructure and provided evidence that cristae shape determines the assembly and stability of RCS and hence mitochondrial respiratory efficiency. Genetic and apoptotic manipulations of cristae structure affect assembly and activity of RCS in vitro and in vivo, independently of changes to mitochondrial protein synthesis or apoptotic outer mitochondrial membrane permeabilization. We demonstrate that, accordingly, the efficiency of mitochondria-dependent cell growth depends on cristae shape. In addition, we extended our analysis of Opa1 as a master regulator of bioenergetics by capitalizing on a different model of Opa1 level modulation, achieved in vivo by targeted insertion of a transgenic single copy of Opa1 isoform 1 in a permissive X chromosome locus (Cogliati et al., Cell 2013). This does not interfere with mouse development, but protects from muscular atrophy, from ischemic heart and brain damage, as well as from hepatocellular apoptosis. Mechanistically, Opa1 stabilizes mitochondrial cristae, increasing mitochondrial respiratory efficiency and blunting mitochondrial dysfunction, cytochrome c release and reactive oxygen species production. Thus, our work identified that RCS assembly emerges as a link between membrane morphology and function and assign a role to Opa1 in the modulation of this process (Varanita et al., Cell Metabolism 2015). In order to address this question, we also wished to identify Opa1 interactors by 3D BN-BN-SDS PAGE and by combining 2D BN-BN PAGE approach with stable isotope labeling of proteins in cell culture (SILAC) and to verify the consequence of the modulation of their levels. We approached this issue by comparing Opa1 containing high molecular weight (HMW) complexes isolated before and after mitochondria were treated with the cristae remodeling inducer BID (Cogliati et al., 2013). After several optimization rounds, we improved the complexes separation and analyzed by LC/MS all the spots where Opa1 intensity was reduced. This proteomic analysis provided us with a vast number of proteins whose levels decreased after treatment with cBID. Since the analysis was semiquantitative, we considered relevant only the candidates showing at least a 50% decrease in the number of identified peptides after cBID treatment and subscreened the results using Mitocarta as database. We observed a decrease in metabolic enzymes, proteins involved in nucleoids formation and function, scaffold proteins, phosphatases and kinases, and unknown proteins, as well as in proteins with a role in oxidative phosphorylation. We further analyzed the datasets by filtering out proteins that were absent in control (untreated mitochondria), proteins not of mitochondrial, ER and Golgi localization, and well-characterized mitochondrial soluble enzymes. In order to refine the screening of candidate proteins we performed a quantitative analysis by metabolic labelling of proteins with medium containing Light aminoacids (L-Arginine and L-Lysine) or Heavy aminoacids ( [13C6, 15N4]-L-Arginine and [[13C6, 15N4 ]-L-Lysine). We could identify a set of proteins residing in Opa1-containing HMW destabilized during cristae remodeling (the results were filtered to select the proteins that change after cBID but not after treatment with the cristae remodeling incompetent cBIDKKAA mutant compared with untreated mitochondria) whose levels decrease during cristae remodeling. It shall be noted that these are only a handful of proteins, whereas levels of many other proteins belonging to the same complex are unaltered during remodeling. Among the hits we found proteins implicated in maintenance of mitochondrial ultrastructure and network, regulators of Opa1 cleavage, and other proteins of unknown function. We went further in the characterization of the functional and physical interaction between two identified hits, as a proof of principle for the analysis of the modulators of Opa1 function and dysfunction and published this in Glytsou et al, Cell Reports 2016.
2. address the molecular mechanism linking mutated/reduced OPA1 to zonal autophagy in RGCs. We wished to address whether and how changes in Opa1 levels affected autophagy. To this end, we employed a multipronged approach to address if mutated Opa1 altered mitochondrial function in soma and neurites of retinal ganglion cells; if it affected autophagic flux and apoptosis; if mitochondria expressing mutated Opa1 were targeted by mitophagy.
3. counteract OPA1 dysfunctions. During this aim we wished to (i) generate a mouse model of ADOA, by capitalizing on the Opa1flx/flx mouse generated in our lab (Cogliati et al., 2013). To this end, we crossed the Opa1flx/flx with Grik4-Cre that expresses Cre in ON-OFF directional selective ganglion cells (DSGCs) that respond to movement of objects in the 4 directions (Ivanova et al., 2010a; Demb, 2007). In afoveate animals like mice, these DSGCs are important in motion detection that can be measured experimentally by using an Optokinetic test, a surrogate measurement for vision (Mansergh et al., 2014). Of note, ablation of Opa1 in these RGCs results in visual dysfunction at 4 months of age. (ii) modulate autophagy in vitro and in vivo. We modulated autophagy in RGCs expressing mutated Opa1 by chemical inhibitors such as 3-MA and obtained a remarkable protection that constitutes the basis for this application and will be extensively discussed in the Preliminary Results section; moreover, we generated double mutants mice lacking Opa1 and the key autophagy gene Atg7 in the same RGCs (by crossing the double Opa1flx/flx, Atg7flx/flx with the Grik4-Cre) and obtained encouraging preliminary results showing a blockage of the visual loss caused by Opa1 ablation. Again, these results will be presented in detail in the Preliminary Results section. (iii) modulate mitochondrial morphology. In a set of preliminary experiment, long term treatment of RGCs (1 week) with the Drp1 inhibitor Mdivi-1 was toxic. Notably, inducible genetic ablation of Drp1 in cortical neurons was also toxic, leading to neurodegeneration and death (Oettinghaus et al., Cell Death and Differentiation 2015; Savoia et al., in preparation). Altogether, these results argued against the possibility of manipulating mitochondrial fission to compensate for mutated Opa1 in RGCs and prompted us to drop this action line.
We therefore devised a research program to verify the hypothesis that by engaging in interactions with different partners, OPA1 regulates mitochondrial functions from the cristae. Its mutations increase autophagy and susceptibility to apoptosis, especially in RGCs. These multiple regulatory points offer several potential targets for therapeutic strategies that can interfere with the natural course of the disease. Our project has been structured in the following specific aims (SA), further organized in a sequence of milestones:
SA1: address the relevance of OPA1 interactors in mitochondrial metabolism, autophagy and apoptosis.; SA2: address the molecular mechanism linking mutated/reduced OPA1 to zonal autophagy in RGCs; SA3: counteract OPA1 dysfunctions
We progressed towards each of these aims, in particular:
1. address the relevance of OPA1 interactors in mitochondrial metabolism, autophagy and apoptosis. We addressed the relationship between Opa1, cristae shape and bioenergetics by a combination of biochemistry, genetics and mitochondrial physiology. Within the cristae respiratory chain complexes assemble into functional quaternary structures called supercomplexes (RCS). In our Cogliati et al., Cell 2013 paper, we reported the results of our investigation on the relationship between respiratory function and Opa1-controlled mitochondrial ultrastructure and provided evidence that cristae shape determines the assembly and stability of RCS and hence mitochondrial respiratory efficiency. Genetic and apoptotic manipulations of cristae structure affect assembly and activity of RCS in vitro and in vivo, independently of changes to mitochondrial protein synthesis or apoptotic outer mitochondrial membrane permeabilization. We demonstrate that, accordingly, the efficiency of mitochondria-dependent cell growth depends on cristae shape. In addition, we extended our analysis of Opa1 as a master regulator of bioenergetics by capitalizing on a different model of Opa1 level modulation, achieved in vivo by targeted insertion of a transgenic single copy of Opa1 isoform 1 in a permissive X chromosome locus (Cogliati et al., Cell 2013). This does not interfere with mouse development, but protects from muscular atrophy, from ischemic heart and brain damage, as well as from hepatocellular apoptosis. Mechanistically, Opa1 stabilizes mitochondrial cristae, increasing mitochondrial respiratory efficiency and blunting mitochondrial dysfunction, cytochrome c release and reactive oxygen species production. Thus, our work identified that RCS assembly emerges as a link between membrane morphology and function and assign a role to Opa1 in the modulation of this process (Varanita et al., Cell Metabolism 2015). In order to address this question, we also wished to identify Opa1 interactors by 3D BN-BN-SDS PAGE and by combining 2D BN-BN PAGE approach with stable isotope labeling of proteins in cell culture (SILAC) and to verify the consequence of the modulation of their levels. We approached this issue by comparing Opa1 containing high molecular weight (HMW) complexes isolated before and after mitochondria were treated with the cristae remodeling inducer BID (Cogliati et al., 2013). After several optimization rounds, we improved the complexes separation and analyzed by LC/MS all the spots where Opa1 intensity was reduced. This proteomic analysis provided us with a vast number of proteins whose levels decreased after treatment with cBID. Since the analysis was semiquantitative, we considered relevant only the candidates showing at least a 50% decrease in the number of identified peptides after cBID treatment and subscreened the results using Mitocarta as database. We observed a decrease in metabolic enzymes, proteins involved in nucleoids formation and function, scaffold proteins, phosphatases and kinases, and unknown proteins, as well as in proteins with a role in oxidative phosphorylation. We further analyzed the datasets by filtering out proteins that were absent in control (untreated mitochondria), proteins not of mitochondrial, ER and Golgi localization, and well-characterized mitochondrial soluble enzymes. In order to refine the screening of candidate proteins we performed a quantitative analysis by metabolic labelling of proteins with medium containing Light aminoacids (L-Arginine and L-Lysine) or Heavy aminoacids ( [13C6, 15N4]-L-Arginine and [[13C6, 15N4 ]-L-Lysine). We could identify a set of proteins residing in Opa1-containing HMW destabilized during cristae remodeling (the results were filtered to select the proteins that change after cBID but not after treatment with the cristae remodeling incompetent cBIDKKAA mutant compared with untreated mitochondria) whose levels decrease during cristae remodeling. It shall be noted that these are only a handful of proteins, whereas levels of many other proteins belonging to the same complex are unaltered during remodeling. Among the hits we found proteins implicated in maintenance of mitochondrial ultrastructure and network, regulators of Opa1 cleavage, and other proteins of unknown function. We went further in the characterization of the functional and physical interaction between two identified hits, as a proof of principle for the analysis of the modulators of Opa1 function and dysfunction and published this in Glytsou et al, Cell Reports 2016.
2. address the molecular mechanism linking mutated/reduced OPA1 to zonal autophagy in RGCs. We wished to address whether and how changes in Opa1 levels affected autophagy. To this end, we employed a multipronged approach to address if mutated Opa1 altered mitochondrial function in soma and neurites of retinal ganglion cells; if it affected autophagic flux and apoptosis; if mitochondria expressing mutated Opa1 were targeted by mitophagy.
3. counteract OPA1 dysfunctions. During this aim we wished to (i) generate a mouse model of ADOA, by capitalizing on the Opa1flx/flx mouse generated in our lab (Cogliati et al., 2013). To this end, we crossed the Opa1flx/flx with Grik4-Cre that expresses Cre in ON-OFF directional selective ganglion cells (DSGCs) that respond to movement of objects in the 4 directions (Ivanova et al., 2010a; Demb, 2007). In afoveate animals like mice, these DSGCs are important in motion detection that can be measured experimentally by using an Optokinetic test, a surrogate measurement for vision (Mansergh et al., 2014). Of note, ablation of Opa1 in these RGCs results in visual dysfunction at 4 months of age. (ii) modulate autophagy in vitro and in vivo. We modulated autophagy in RGCs expressing mutated Opa1 by chemical inhibitors such as 3-MA and obtained a remarkable protection that constitutes the basis for this application and will be extensively discussed in the Preliminary Results section; moreover, we generated double mutants mice lacking Opa1 and the key autophagy gene Atg7 in the same RGCs (by crossing the double Opa1flx/flx, Atg7flx/flx with the Grik4-Cre) and obtained encouraging preliminary results showing a blockage of the visual loss caused by Opa1 ablation. Again, these results will be presented in detail in the Preliminary Results section. (iii) modulate mitochondrial morphology. In a set of preliminary experiment, long term treatment of RGCs (1 week) with the Drp1 inhibitor Mdivi-1 was toxic. Notably, inducible genetic ablation of Drp1 in cortical neurons was also toxic, leading to neurodegeneration and death (Oettinghaus et al., Cell Death and Differentiation 2015; Savoia et al., in preparation). Altogether, these results argued against the possibility of manipulating mitochondrial fission to compensate for mutated Opa1 in RGCs and prompted us to drop this action line.