Omaveloxolone (RTA408) is a second-generation synthetic derivative from oleanolic acid, a natural triterpenoid found in medicinal plants. Derivatives of oleane exhibit anti-inflammatory and anticancer activities due to their binding to the adapter protein Kelch-like ECH-associated protein 1 (Keap1) Keap1 and thereby attenuating the transcription factor, erythroid 2-like 2 (Nrf2) degradation by ubiquitination. Subsequently, the accumulation of factor Nrf2 in the nucleus activates the expression of antioxidant genes, decreases the expression of pro-inflammatory genes, and promotes mitochondrial biogenesis (Uittenbogaard and Chiaramello, 2014; Probst et al., 2015) Omaveloxolone was previously evaluated in humans with advanced solid tumors and was shown, in humans, to be tolerable, supporting further investigation of this compound in cancer therapy (Creelan et al., 2017). Arbeti et al. showed that triggering the Nrf2 antioxidant pathway activators was favorable due to the prevention of lipid peroxidation in cellular mouse models of Friedreich’s ataxia (FA). FA is a common inherited recessive ataxia caused by the mutated FXN gene, involving mitochondrial iron accumulation and increased oxidative stress. Subsequently, the same group showed that omavleoxolone had a similar positive effect and also restored mitochondrial respiratory chain complex I (C1) activity, and mitochondrial function, in mouse cells as well as in human FA patient-derived fibroblasts (Abeti et al., 2015; Abeti et al., 2018). Safety, pharmacodynamics, and potential benefit of omaveloxolone were evaluated in two clinical trials in FA patients showing significantly improved neurological function compared to placebo and was generally safe and well-tolerated. It was concluded that omaveloxolone represents a potential therapeutic agent in FA (Lynch et al., 2018; Lynch et al., 2021).

Based on the favorable results, demonstrated by others (Uittenbogaard and Chiaramello, 2014; Abeti et al., 2015; Probst et al., 2015; Creelan et al., 2017; Abeti et al., 2018; Lynch et al., 2018; Lynch et al., 2021), we set out to investigate the effect of omaveloxolone in primary fibroblasts derived from patients with other medical conditions involving mitochondrial dysfunction. Included in this study were fibroblasts derived from patients with two primary, mitochondrial diseases and two patients with genetic forms of Parkinson’s disease (PD) as follows; CI deficiency with fatal encephalomyopathy due to mutated NDUFAF4 (C6ORF66) (Saada et al., 2008), a CI assembly factor; and a relatively mild variant of cytochrome c oxidase (COX, mitochondrial respiratory chain complex IV) deficiency due to a mutated COX4I1, the common isoform of COX subunit 4, which presents with Fanconi anemia-like symptoms (Abu-Libdeh et al., 2017). The two patients with PD were as follows; a juvenile form of PD caused by a biallelic copy-number neutral intragenic inversion in the PRKN gene, encoding the E3 ubiquitin ligase Parkin (Mor-Shaked et al., 2021) and a juvenile PD patient due to biallelic mutations in the PARK7 gene encoding the DJ-1 protein (Namnah et al., 2020) The fibroblasts from these patients were incubated 72 h with or without omaveloxolone and subsequently evaluated for mitochondrial function. The effect on various parameters was variable, but a significant increase in mitochondrial content and mtDNA was noted without a concomitant rise in reactive oxygen species (ROS).

We investigated the effect of omevaloxolone on mitochondrial parameters in order to assess the preliminary therapeutic potential of this compound in diseases involving mitochondrial dysfunction. In this study, we included a variety of primary fibroblasts representing a severe, fatal mitochondrial disease C1 (Saada et al., 2008), a milder mitochondrial disease affecting COX (Abu-Libdeh et al., 2017), and two recessive forms of PD, a common neurodegenerative disease, which has previously been linked to mitochondrial dysfunction and oxidative stress [reviewed in (Zambrano et al., 2022)]. In this preliminary study, we incubated the cells with omaveloxolone in a concentration that was previously reported for the in vitro FA studies (Abeti et al., 2015; Abeti et al., 2018) and examined six different parameters to assess different aspects of its effect on mitochondrial function.

First, we examined growth in the GLU medium which was only slightly but not significantly affected as measured by methylene blue (MB) or by viable cell count using the MUSE cell analyzer. Only COX showed a mild 15%–20% decrease (Supplementary Figure S1). As there was no significant difference between MB and viable count, we opted to continue with MB as this method is more convenient to perform in microtiter wells. Additionally, growth was not significantly different, and we did not detect any increased cell death in cells grown up to 15 days (medium change every 3 days) in the presence of omevaloxolone relative to 72 h (results not shown). As our previous experience (Golubitzky et al., 2011) showed that 72 h is the optimal time point to examine growth on GAL versus GLU, the following experiment was performed at this time point.

Next, we measured growth in glucose-free medium (GAL), a medium in which cells with impaired oxidative phosphorylation (OXPHOS) are restricted, relative to glucose-containing medium (GLU) where the energy needs are met by glycolysis (Figure 1A). Cell growth on GAL relative to GLU shows markedly impaired growth only in the CI deficient cells, while the growth of PD cells was only slightly but not significantly restricted. Omaveloxolone had no significant effect with the exception of COX where the GAL/GLU ratio was minimally elevated.

We also measured mitochondrial ATP production in permeabilized cells, where CI and PD cells showed decreased baseline ATP production (Figure 1B). Oxamevalone did not significantly alter ATP production, with the exception for PD cells carrying the DJ1 mutations where ATP production was slightly (15%) but significantly increased. In the COX cells, ATP production showed a decreasing trend, but this was not statistically significant.

Unexpectedly, omaveloxolone decreased ROS only in the controls and DJ-1 mutant cells, while other cells were not significantly affected (Figure 1C). Notably, only PRKN mutated cells disclosed an a priori elevated ROS level, which was not affected by omaveloxolone.

Mitochondrial membrane potential (MMP) was estimated by calculating the ratio of TMRE which is a mitochondrial marker, dependent on MMP with MGT staining, a mitochondrial dye, independent of MMP as a marker for mitochondrial mass. Unexpectedly, omaveloxalone significantly decreased MMP in most patient cells (Figure 1D) The decreased TMRE: MTG ratio with the largely unaltered ATP production led us to speculate if our observations could be a result of variable mitochondrial content. Accordingly, we calculated mitochondrial content relative to cell mass (Figure 1E), and we also measured mitochondrial DNA (mtDNA) content by real-time quantitative PCR (Figure 1F). Omaveloxolone significantly elevated mitochondrial mass, up to control values in CI cells which had a priori decreased mitochondrial content. There was also an increasing but not significant trend in the patient cells including PD-P cells which started out with a higher baseline (Figure 1E). mtDNA content was significantly elevated in all cells including the controls (Figure 1F). Taken together, the results show that omaveloxolone had a significant effect on all cells by elevating mitochondrial content or mtDNA content or both. Interestingly, increased mitochondrial biogenesis is predicted to enhance ROS production, but this was not the case with omaveloxolone.

To complete our studies, we also measured OCAR and ECAR in naïve cells and in cells treated with omaveloxalone (Figure 2). The whole procedure of a typical experiment is depicted in Figure 2A, and calculated values depicted in Figure 2B show significantly decreased basal, maximal, and ATP-linked OCR in the patient’s cell. The severely decreased spare capacity in the patient cells indicates that these cells a priori respire near their maximal capacity even under basal conditions. Also, in this system, omaveloxolone showed variable effects. Basal OCR was increased only in controls and in CI cells, while maximal capacity was affected in both CI and PD cells.

Omaveloxolone positively affected ATP-linked OCR in both PD cells, in accordance with previous ATP production experiments (Figure 1B). Spare capacity was slightly but significantly increased only in CI and PD cells with the DJ1 mutations. For reasons we currently are unable to explain, respiratory control (Figure 2C) was decreased in all cells but not affected by omaveloxolone, in discordance with the MMP results (Figure 1D). Nevertheless, the general bioenergetic status of all the cells was improved as depicted in Figure 2D showing elevated basal OCR concomitant with elevated ECAR. This is exemplified by the CI cells; omaveloxolone induced a shift to the right meaning increased acid production as a result of glycolysis and also shifted upwards toward increasing mitochondrial oxygen consumption, meaning that these cells are more energetic.

To summarize, our results show variable effects of omaveloxolone, still with some positive aspects, namely, the effect on the bioenergetic status and the maintained or decreased ROS production despite increased mitochondrial content. PD cells with the DJ1 mutations clearly benefitted the most from omaveloxolone treatment with increased ATP production and ATP-linked OCR with decreased ROS production.

Despite advances in the diagnosis and understanding of the pathomechanisms of diseases involving mitochondrial dysfunction, only a few are responsive to treatment. This is especially true for primary mitochondrial disorders, a common group of inborn errors of metabolism where treatments are mostly supportive and with relatively few ongoing clinical trials. Promisingly, over the past decade, some advances in therapy development mainly with many small molecules are now transitioning from preclinical to early phase human interventional studies, as is omaveloxolone for FA. However, the conclusion of these studies is not yet finalized (Lynch et al., 2018; Pitceathly et al., 2021).

Our results in the CI-deficient cells with severe mitochondrial dysfunction show slightly beneficial effects of omaveloxolone in terms of oxygen consumption and elevated mitochondrial content. Still, there was no significant increase observed neither in ATP production nor in ATP-dependent oxygen consumption. The same is also true for cells derived from the COX patient, which were a priori less affected. Thus, the question remains open if elevating mitochondrial content alone could have some beneficial metabolic effect other than increasing ATP production. Indeed, increasing mitochondrial biogenesis has previously been suggested as a therapeutic target in neurodevelopmental disorders and neurodegenerative diseases (Uittenbogaard and Chiaramello, 2014). We preferred DCF for measuring ROS production rather than mitochondria- specific dyes, in order to estimate overall cellular oxidative stress. Notably the two cells with primary mitochondrial diseases did not express elevated ROS. This is in accordance with our previous experience, showing that not all mitochondrial disease fibroblasts express elevated ROS, depending on the nature of the mutation and the antioxidant capacity of the individual cell (Douiev et al., 2016; Yu-Wai-Man et al., 2017). Interestingly, in this study, ROS production levels remained stable or decreased even though mitochondrial content was increased by omaveloxolone.

With respect to the PD cells, there was a notable difference between the two. It seems clear that the PD-D cells harboring DJ1 mutations gained significant benefit from omaveloxolone with increased ATP production and ATP-dependent OCR concomitantly with decreased ROS. These cells were a priori less affected than the PD cells harboring the PRKN inversion, which disclosed elevated ROS which was not ameliorated with omaveloxolone. On the contrary, the mitochondrial content and mtDNA were markedly increased in PD-P, Parkin cells. Apparently, this elevation could be responsible for the slight, but not statistically significant, increase in ROS. In these cells, we also noted a discrepancy between mitochondrial ATP production in microtiter wells, which was not affected by omaveloxolone and ATP-linked OCD which showed a slight but significant increase.

Taken together, the only parameter which was common to all cells was the increase in mitochondrial mass and/or mtDNA, which is consistent with the promotion of mitochondrial biogenesis by NRF2 activators (Uittenbogaard and Chiaramello, 2014). Other parameters varied between the cells, and as of now, we are not sure whether these differences stem from the nature of the mutation or from the individual genetic background. This will be addressed in the future, depending on the availability of fibroblasts with the same mutation but with a different genetic background. Our previous study examining fibroblasts with the same homoplasmic mtDNA mutation from different, unrelated individuals suggests that the genetic background does play a significant role (Yu-Wai-Man et al., 2017). Another limiting factor, in studying primary fibroblasts, is that the phenotype might alter with passages and the number of replications, and thus the number of cells is limited. For these reasons, we performed our study with cells from passages 4–6. This also limited our ability to perform more in-depth investigation, including nuclear extraction for examining Nrf2, biochemical, and omics studies. Still from our previous experience, primary fibroblasts maintain the phenotype more consistently than immortalized cells (Saada, 2014). Despite the limitation of this study, it is in line with the suggested Nrf2 signaling pathway as one of the targetable, potential therapeutic pathways in mitochondrial and neurodegenerative diseases (Millichap et al., 2021; Saha et al., 2022).

We conclude that omaveloxolone seems to have a therapeutic potential in certain diseases involving mitochondrial dysfunction and including PD; however, the investigation should be individually evaluated and further studies are required to determine the precise mode of action.