Biochemical and Chemical Biology Approaches to Investigate and Target the Mitochondrial GTPase OPA1

Mitochondria are double membrane organelles of prokaryotic origin that are historically associated with cellular respiration, being the main site of ATP production through oxidative phosphorylation. Apart from this function, mitochondria are also crucial participants in mediating intrinsic apoptosis as host of numerous apoptogenic factors such as cytochrome c. These are stored within the cristae of mitochondria, which are sac-like structures that are formed by the inner mitochondrial membrane that are connected to the periphery of the membrane by the narrow tubular cristae junctions. Intact, cristae keep apoptogenic within the mitochondria. The disruption of cristae junctions and cristae remodeling allows contained molecules, including cytochrome c, to gain access to cellular cytosol and to initiate a cascade resulting in cell death. Numerous mechanisms control cristae structure, among them being the large GTPase OPA1 (Optic atrophy protein 1). OPA1 forms complexes at the site of cristae junctions to maintain narrow openings and thus prevent efflux of cristae contents. During apoptosis, OPA1 oligomers are disrupted, coinciding with remodeling of the cristae, opening of their junctions and the release of apoptogenic factors. The role of OPA1 in this process was affirmed by previous data showing that the absence of OPA1, and its oligomers facilitate the induction of intrinsic apoptosis. Conversely, increased levels of OPA1 reduce cristae width and raise the threshold for inducing apoptosis. Through its control of cristae morphology, OPA1 also regulates mitochondrial metabolism by stabilizing the respiratory chain supercomplexes that reside on the cristae membranes. Apart from the aforementioned functions, OPA1 mediates mitochondrial fusion together with the outer mitochondrial membrane protein Mitofusin1. Mitochondrial fusion and fission are crucial because they allow these highly dynamic organelles to rapidly adapt to cellular demands and challenges. The machineries driving fusion and fission are located at the inner and outer mitochondrial membrane. The enzymes driving these membrane modulations are members of the family of large GTPases, which comprises dynamin-like proteins that commonly drive membrane fusion or fission in a GTP- and oligomerization- dependent manner. The physiological importance of OPA1 is underlined by the fact that mutant forms of it are associated with the development of autosomal dominant optic atrophy (ADOA) in humans, the leading cause for inherited blindness worldwide originating from the selective death of retinal ganglion cells (RGCs). Interestingly, protein domains that are associated with the OPA1 GTPase domain and GTPase effector domain are hotspots for ADOA-associated mutations. The GTPase effector domain (GED) is thought to function similarly to the classic guanine nucleotide exchange factor of small GTPases to support GTP hydrolysis. The aim of this thesis was to investigate the impact of mutations that are associated with the development of ADOA on the function of OPA1 through genetic, biochemical, and chemical biology approaches. First, we established a yeast-based assay to characterize OPA1 mutations by expressing different disease variants of OPA1 in a yeast strain lacking Mgm1, the yeast homologue of OPA1. Growth on a non-fermentable carbon source (glycerol) forces yeast cells to rely on respiration for energy production, an Mgm1-dependent process. Expression of an Mgm1-OPA1 chimeric protein restores the growth defect observed in the absence of Mgm1. After introducing ADOA-associated OPA1 mutations into the Mgm1- OPA1 fusion protein, we can assess the extent to which growth on a non-fermentable carbon source is rescued and determine their impact on the OPA1 function in this process. To complement this genetic approach, we developed an in vitro system to characterize the enzyme kinetics of wild type and mutant OPA1. To do so, we generated a protocol for the purification of recombinant OPA1 (rOPA1) using an inducible bacterial expression system and affinity based purification. rOPA1 was used to elucidate the enzymatic kinetics of OPA1, which shows positive cooperative hydrolysis behavior. Further biochemical characterization suggests that the cooperative behavior is due to OPA1 oligomerization. We used both systems to analyze the impact of mutations that are associated with ADOA classic and ADOA plus, a more severe multi-systemic form of the disease, in the GTPase domain and GED. Combining the data of both approaches allowed us to correlate mutations associated with ADOA classic with a loss of function phenotype and the ones associated with ADOA plus with a dominant negative effect on the wild type protein in vivo. Based on the kinetics data we obtained for the individual mutant proteins this effect is likely due to changes in the hydrolysis capacity and the cooperative kinetics of OPA1. While ADOA classic mutations lose their cooperative GTPase activity which correlates with a lower affinity for GTP, ADOA plus mutations maintain their cooperativity but have a severely altered catalytic turnover of GTP. Hence we hypothesize that retaining the cooperative behavior allows ADOA plus OPA1 variants to oligomerize with the wild type protein and thus impact negatively on their function. ADOA classic mutants however fail to cooperate with the wild type proteins. Similarly, co-expression of ADOA plus and wild type OPA1 in our yeast system revealed a dominant negative effect in respiration dependent growth and mitochondrial morphology while the ADOA classic mutations have no impact. Together, these data suggests that ADOA-associated mutations can be biochemically grouped into ADOA classic and ADOA plus mutations based on their impact on the cooperative GTPase activity of OPA1. We also sought to characterize the role of OPA1 in apoptosis. Given thatOPA1 has an anti-apoptotic role by controlling the release of cytochrome c from the intracristal space, we hypothesized that it inhibiting this function may yield a pharmacological drug to induce apoptosis in cancer cells. We made use of the rOPA1 to screen a library of small compounds. For that a semi-automated high-throughput screening (HTS) was set up in which the hydrolysis of GTP by rOPA1 served as readout to measure the inhibition of the GTPase activity of OPA1. OPA1 activity was determined using a colorimetric assay that allows the measurement of free organic phosphate that is released during the hydrolysis of GTP to GDP. Given that there is no existing inhibitor of OPA1 nor published model or structure of OPA1 to use in an in silico screen, we selected our HTS library based on diversity of compounds. Our screen revealed 8 compounds that inhibit OPA1 GTPase activity in two repetitions of the screen. The compounds’ effects were further confirmed in a dose-dependent response experiment, and also provided us with maximal inhibition (Imax), half maximal inhibition (IC50), and Hill slope values for each. Based on their potency and biochemical effect on OPA1 we grouped them into potent and moderate inhibitors. Potent compounds inhibit OPA1 activity by more than 50% while moderate ones inhibit up to 50% of the activity. How the inhibitors impact on the different species of OPA1 will be focus of future experiments that may serve as scaffold for the identification of an OPA1 inhibitor in vivo. In conclusion, the data presented in this doctoral thesis contributes to the field of ADOA research by correlating the mutations of ADOA classic and ADOA plus to functionally distinguished alteration on OPA1 GTPase function. Moreover, we developed a robust assay that can be used to characterize OPA1 disease mutation and phenotypes. Lastly, we set up and implemented a high-throughput screening that identified novel inhibitors of OPA1 GTPase activity in vitro.