Abstract

Mitochondria are involved in many dynamic processes, including their role in the production of the cell’s energy currency, adenosine triphosphate (ATP). This production is carried out through a complex chain of chemical reactions, followed by the transition of electrons through four protein clusters on the mitochondrial membrane called respiratory complexes. This transfer drives protons across the membrane creating a high concentration on one side of the membrane. The difference in concentration, known as the mitochondrial membrane potential (MMP), allows the flow of protons through respiratory Complex V to combine adenosine diphosphate with phosphate to create ATP [1].
Mitochondrial problems were first discovered in Parkinson’s Disease (PD) in the 1980’s when drug users injected contaminated heroin and subsequently developed parkinsonian symptoms [2]. Further investigation in animal models identified that the contaminant was 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP), which was metabolised to 1-methyl-4-phenylpyridinium (MPP+),an inhibitor of mitochondrial respiration [2, 3]. Inhibition of mitochondrial respiration with MPP+results in reduced ATP levels and the production of harmful reactive oxygen species [4–6]. Mitochondrial dysfunction in PD was later confirmed when reduced respiratory complex activity was identified in post-mortem substantia nigra samples from PD patients [7]. Following this discovery mitochondria became the focus of intense research in PD.
As genetic screening became more prevalent, several mutations in mitochondrial related genes were discovered to be associated with increased risk of PD, these genetic variants include PINK1 [8], Parkin and DJ-1 [9]. While other genetic variants have been associated with a functional decline in the mitochondria but don’t have a direct mitochondrial link, for example LRRK2 [10–12], VPS35 [13] and SNCA [14]. The study of these genetic variants has shaped our understanding of mitochondrial involvement in PD. Parkin and PINK1 proteins are both vital to the clearance of defective mitochondria. Work in animal models and patient tissue suggests that Parkin and PINK1 mutants have reduced respiratory complex I activity, low levels of mitochondrial respiration and morphological changes in the mitochondria [15–21]. Parkin models have also shown reduced MMP and ATP levels [15]. A vast amount of research has also assessed mitochondrial dysfunction in LRRK2 mutants. Mutant LRRK2 patient-derived fibroblasts and induced dopaminergic neurons display reduced mitochondrial functioning, ATP levels and respiratory complex activity [10, 22–25]. However, they show a reduced activity in complex III and IV, unlike other forms of PD [10]. Interestingly it is suggested that LRRK2 mutations slow the clearance of defective mitochondria leading to a dysfunctional network [26]. There are conflicting reports about mitochondrial morphology in LRRK2 models, although this may be due to the use of different disease models. Some demonstrate that mitochondria are elongated in both mice models and patient-derived fibroblasts [11, 12], while others report a fragmentation of the network [27, 28]. All this evidence suggests that mitochondrial dysfunction is a crucial component to the pathogenesis of genetic PD.
Research into sporadic PD is far more conflicting leading researchers to theorise that PD is a mechanistically heterogeneous disorder, meaning that different patients have defects in different pathways which are primary drivers of cell death [29]. Evidence suggests that mitochondrial dysfunction is one of these defects and heterogeneity would explain the contradictory reports about mitochondria function in sporadic PD. For example, respiratory complex I activity has been shown to be reduced in patient post-mortem substantia nigra, muscle, and various blood cell types, however, other studies report no deficit in the same tissues [7, 30–32]. Reports of deficits in respiratory complexes II, III and IV are also occasionally observed [30–35]. While studies on other markers of mitochondrial function have reported differences as well, MMP has been observed to be increased [36], decreased [37] or the same [38] as healthy individuals in different studies. Mitochondrial morphology is also altered in sporadic PD patients. Mitochondrial shape and size were found to be variable in post-mortem brain and muscle tissue [39, 40]. Whereas another study showed that mitochondria are swollen and enlarged [37] and a study in 2020 reported a decrease in mitochondrial size and increased fragmentation of the mitochondria [36]. These conflicting results may be driven by the small-scale nature of some of these studies and the complex heterogeneity of sporadic PD.
Several large studies in patient-derived fibroblasts found that there was no difference in mitochondrial activity between controls and patients, this includes respiratory complex activity [41], basal respiration [42] or MMP and ATP levels [38]. Carling et al. (2020) took this one step further and stratified their cohort, identifying a mitochondrial dysfunction subgroup. These patients displayed reduced levels of respiratory complexes as well. This suggests that a proportion of the sporadic PD population are affected by mitochondrial dysfunction and that further classification of PD may be required to personalise patient treatments.
As mitochondria are so pivotal to cellular, and in particular neuronal function, it is clear that any therapeutic which boosts mitochondrial function will be beneficial to PD patients. It should be recognised that mitochondrial dysfunction is unlikely to be the main driving cause of disease in all patients. However, it is theorised that restoring mitochondrial function in patients will slow progression, even if mitochondrial dysfunction is not the direct cause but a secondary mechanism of disease. The enhanced ATP levels and reduced ROS production could reduce cell stress and prolong the life of the cells.
OVERVIEW OF MITOCHONDRIAL MODULATORS IN PD CLINICAL TRIALS
Since the 1980’s mitochondrial dysfunction has been recognised in PD [2], yet there are still very few mitochondrial modulators being utilised as prospective disease modifying therapies in PD. There is one compound, Ganoderma, in phase 3 clinical trials and this compound has multiple targets including mitochondria (Tables 1 and 4). Coenzyme Q10, a mitochondrial antioxidant, has been investigated in several phase 2 and 3 trials with mixed results [43–47]. Although all doses were well tolerated and safe, meta-analysis of 8 trials demonstrated that Coenzyme Q10 did not improve motor symptoms [48].
List of active clinical trials
List of Terazosin clinical trials.
List of Nicotinamide riboside clinical trials
List of Ganoderma clinical trials
Three compounds are currently considered to be the top mitochondrial modulators, two of these are the repurposed compounds UDCA and terazosin, and the third is a supplement called nicotinamide riboside. UDCA, a drug approved to treat primary biliary cholangitis, was tested in a small pilot study at an increased dose of up to 50 mg/kg over six weeks. 7 Tesla 31Phosphorous-Magnetic Resonance Spectroscopy (31P-MRS) indicated a possible beneficial effect in the three patients who had neuroimaging [49]. Payne and co-workers recently completed the UP study assessing UDCA in 30 patients with recent onset PD. UDCA was safe and extremely well tolerated achieving the trials primary outcome. 31P-MRS confirmed midbrain target engagement and gait analysis suggested a possible beneficial effect on motor symptoms. However, there were no significant changes in the clinical assessment based on the UPDRS/part 3 clinical rating scale, but the study was also vastly underpowered to robustly detect a possible difference in UPDRS/part 3 [50]. Terazosin, a drug approved for hypertension and prostate hypertrophy, was also tested in an early proof of concept study in gradually increasing doses up to 5 mg daily. 3 out of 8 patients started on Terazosin dropped out due to orthostatic hypotension or dizziness. However, they reported an interesting correlation between whole blood ATP levels and changes in 31P-MRS in the 5 patients who completed the treatment compared to the 5 patients on placebo [51]. Two ongoing terazosin trials are targeting patients displaying early symptoms of pre-motor PD such as rapid eye movement sleep behaviour disorder (RBD) to try and prevent the development of motor symptoms of PD (Tables 1 and 2). These types of studies may become the future of PD clinical trials as mitochondrial compounds cannot restore neurons but can slow or prevent their degeneration.
Isradipine, a dihydropyridine calcium channel antagonist, that indirectly reduces mitochondrial “burden” has been tested at phase 1, 2 and 3 in PD. Despite early successes the phase 3 clinical trial failed to achieve the primary outcome of improving motor symptom UPDRS scores [52]. This trial may highlight the fundamental issues with PD clinical trials. While isradipine studies were based on solid preclinical data and supported by epidemiological studies, there was no data on target engagement, which is the ability to impact the desired mechanism at the dose tested. Therefore, several additional aspects need to be considered for future trials of isradipine and other compounds. It is suggested that target engagement measures should be included, much like terazosin and UDCA which utilise 31’P-MRS, to confirm doses are adequate to protect neurons. Trials should also aim to add additional outcomes beside the UPDRS score difference, as changes in the UPDRS score are not linear and with each clinical trial the placebo group can change at different rates [53]. Unfortunately, this is difficult without a biomarker of disease, but the use of motion sensors may provide support for upcoming trials.
Multiple supplements have putative mitochondrial beneficial effects. Vitamin D, melatonin and nicotinamide riboside are all commercially available and therefore trials assessing these compounds do not always have a phase allocation. These supplements showed no side effects after administration and have a positive effect on mitochondrial function. However, convincing data on beneficial effect is lacking.
Mitochondrial modulators have become an extensive area of development across many diseases, which means many compounds are undergoing preclinical testing. Two ways of improving mitochondrial function are to remove dysfunctional mitochondria by a process of recycling called mitophagy or by increasing the production of new mitochondria. For example, The Silverstein Foundation is funding research into modulators of Miro1, a key contributor to the transport of mitochondria for recycling. RNS60 by Revalesio and MSDC-0160 by Metabolic Solutions Development upregulate production of new efficient mitochondria. However, many companies are assessing compounds that have a beneficial effect on mitochondria in models of disease but do not have a known or published mechanism of action, such as those from Pretzel Therapeutics and Lucy Therapeutics. This is not an exhaustive summary of preclinical compounds due to the vast range in targets, companies, stages of development and evolving nature of drug development pipelines.
UDCA
UDCA is a secondary bile acid that is produced by bacterial metabolism in the gut, where it is partially absorbed into the bloodstream. Endogenously UDCA accounts for ∼5% of bile acids in humans.
A drug screen targeting mitochondrial dysfunction in Parkin mutant patient-derived fibroblasts identified that UDCA and similar bile acids rescued ATP and MMP levels [54]. Further studies in LRRK2 G2019 S patient fibroblasts found it rescued ATP levels [10] and in sPD patients it improved Complex I and IV activity as well as MMP and ATP levels [38]. Interestingly, a subsequent study observed lower levels of UDCA in PD patients developing mild cognitive impairment than normal cognition patients, suggesting UDCA may play a role in progression [55]. UDCA was also successful in animal models rescuing ATP and MMP while preventing neuronal death from toxin insult by rotenone and MPP+, known mitochondrial toxins that induce parkinsonism [56, 57]. UDCA, which is already licensed to treat primary biliary cholangitis, has been shown to be well tolerated and safe in patients with motor neuron disease, while also being brain penetrant [58]. Therefore, this repurposing study was able to bypass phase 1 and start a phase 2 trial.
The “UP” study was a proof-of-concept study, meaning it aimed to assess safety, target engagement and efficacy but does not have the participant numbers to reliably assess efficacy. It enrolled 31 patients across two centres. Participants took 30 mg/kg daily and were assessed over 48 weeks with an 8-week washout period following week 48. This trial utilised 31’P-MRS to assess ATP levels in the putamen and striatum in order to understand target engagement. It also assessed participants with OptoGait and Opals systems, which are motion sensors that enable a quantification of movement and walking, to enable the unbiased, quantified assessment of motor symptoms [59].
Secondary outcome measures were assessed as the change from baseline at week 48 in the following:
MDS-UPDRS scores, in the practically defined ‘OFF’ state.
ATP, phosphocreatine and phosphate levels in participant brains, measured using 31P-Magnetic Resonance Spectroscopy.
Supervised, sensor-based gait analysis.
Terazosin
Terazosin is an α-adrenergic blocker used to treat benign prostatic hyperplasia (enlarged prostate) and hypertension. It also binds to phosphoglycerate kinase 1, the first enzyme in glycolysis [59]. Glycolysis is the first stage of respiration and the production of ATP; it produces pyruvate that is then utilised by mitochondria in respiration.
In vitro and in vivo studies found that terazosin increased ATP levels in cells and brain tissue of various PD models, while also preventing neuron loss [60] and protected against cognitive impairment in mouse ventral tegmental area dopamine depleted mice [61]. 4 epidemiological studies have assessed the incidence of PD in populations of patients taking terazosin or tamsulosin, which is also used to treat high blood pressure. These studies observed differing effects of terazosin with some showing it lowered the risk of PD compared to those taking other high blood pressure medication, while one found no difference compared to a control group. However, there is some debate about the selection of an accurate control group and possible effects of undiagnosed PD at initial assessments [60–64].
The two trials by the Cedars Sinai Medical Center are interesting because they are targeting patients that are not displaying PD motor symptoms (Table 1). Their aim is to select participants that are showing early risk signs of PD and treat to prevent/slow the onset of PD by preventing the early loss of neurons.
A newly registered clinical trial by the I.R.C.C.S. Fondazione Santa Lucia is a phase 4 clinical trial assessing terazosin and a supplement, Lisosan-G, in 50 pwp. The main aim of the study is to validate metabolism indicators as possible biomarkers for further studies and to assess the engagement of both compounds in Nrf2 activation. Nrf2 is part of the antioxidant response and therefore has a secondary impact on mitochondrial health when activated [65].
The pilot study by the University of Iowa was the first study to be completed for terazosin in PD. It tested safety and tolerability in 13 pwp. This study did not have enough participants to adequately assess efficacy but aimed to guide the design of future studies and assess terazosin’s effectiveness and target engagement to support results from animal model studies. One of the main aims was to assess the frequency of falls for the patients as terazosin lowers blood pressure which is already known to dramatically reduce in pwp when they stand up. This study is outlined in more detail below.
The secondary outcomes assessed:
The change in blood pressure at 0, 2, 6 and 12 weeks.
The number of participants with intolerable side effects.
The number of participants not complying with the treatment regime.
Idebenone
More than 100 papers have been published on idebenone, which has been used for treating different types of diseases such as Leber’s hereditary optic neuropathy [66], Duchenne muscular dystrophy [67], Friedreich’s ataxia [68] etc, but at the moment very few (4 papers) have been published regarding the use of idebenone for treating PD.
Idebenone is a short-chain benzoquinone, an analogue of coenzyme Q10. Idebenone has a reduced lipophilic side chain and terminal hydroxyl groups that increase polarity and solubility. Because of this structure it can clear oxygen free radicals acting as an electron carrier and serve as an antioxidant [69].
Due to the nature of its structure, several groups have investigated the effect of idebenone on mitochondrial functions. Giorgio et al., in 2012 showed that idebenone can restore the ATP production in HQB17 and RJ206 cell cybrids after complex I and complex II inhibition. Moreover, idebenone reduces the rotenone-induced lipid peroxidation as well as the level of glutathione and superoxide dismutase (SOD) which play an important role in the antioxidant defence [70]. Idebenone also has a positive effect on mitochondrial stress enhancing mitochondrial clearance via mitophagy in a PD mouse model. In fact, an increased level of PINK1 and Parkin, as well as VDAC1 and BNIP3 have been shown after idebenone treatment, suggesting a potential role of this molecule in the autophagic flux [71].
The Zhejiang University study is investigating the effect of Idebenone after 24 months of treatment. The aim is to evaluate if this treatment may slow the progression of PD.
Secondary outcome is to assess dopamine transporter levels in the striatum after 12 and 24 months, using dopamine transporter positron emission tomography.
Nicotinamide Riboside
NR is a member of the vitamin B3 family. In cells, it is readily converted into nicotinamide adenine dinucleotide (NAD+). NAD+is a critical regulator of NAD+-dependent enzymes that mediate cellular signalling pathways related to metabolism and mitochondrial function. Declining NAD+concentrations in tissue is a pathological factor in various ageing-associated diseases, with the brain being particularly vulnerable due to the high energetic demand of neurons.
NR has shown strong efficacy in animal models and has improved key features of neurodegenerative disorders, including mitochondrial dysfunction. Moreover, NR rescues mitochondrial defects in iPSCs-derived neurons, increasing mitochondrial mass as well as mtDNA, and decreasing the mitochondrial ROS level [76]. The NADPARK study demonstrated that orally NR administration is safe. Furthermore, they have shown that NR is able to increase the cerebral NAD levels (although 3 patients have shown no evidence of NAD increase). Interestingly, patients showing a > 10% increase in cerebral NAD had significantly improved UPDRS scores suggesting improved symptoms. Tissue samples were analysed to further understand the effect NR was having. Peripheral blood mononuclear cells and muscle samples showed elevated metabolites and an upregulation of oxidative phosphorylation related genes [77].
The clinical trial NCT03568968 by the Haukeland University Hospital is interesting because they are testing NR in early PD with the aim to slow the progression of the disease by preventing the NAD-deficiency which is a key-event in the pathogenesis of PD.
In the NOPARK study they tested a double-blinded clinical trial investigating if NR can be used as an oral administration, and if it is safe. Moreover, they evaluated the effect of NR on cerebral metabolism in PD. In this study they have demonstrated that NR administration is safe and leads to an increased level of cerebral NAD. However, the cerebral NAD response is variable and related to the individual person (e.g. 3 patients showed no cerebral NAD increase at all).
The Secondary outcome was to assess clinical changes measured by MDS-UPDRS from using NR.
Ganoderma
Gl is a white-rot fungus largely used in traditional Chinese medicine. It contains several bioactive molecules i.e. polysaccharides, triterpenoids, adenosine and sterols with many beneficial effects. The main action of Gl is as an antioxidant. Indeed, it has been shown that GI increases the activity of antioxidant enzymes including glutathione peroxidase (GPx), catalase (CAT) and manganese-superoxide dismutase in cardiac, hepatic and cerebral mitochondria of aged mice. Moreover, GI also increases the level of mitochondrial α-ketoglutarate dehydrogenase (α-KGDH), pyruvate dehydrogenase (PDH) as well as complex I and II activity [78, 79]. Ren et al., (2019) have shown that GI treatment has a positive effect on the mitochondria, improving the electron transport chain efficiency as well as the ATP production and decreasing the ROS level in mice neuronal cells. In addition, GI rescued the level of protein involved in the mitophagy pathway such as AMPK, ULK1 and PINK1 suggesting a positive modulation of the mitochondrial clearance [80].
The pilot study by the Xuanwu Hospital in Beijing is the first study to be completed for GI. In this study the primary outcome tested the effect of GI on non-motor symptoms of PD and the secondary outcome assessed GI’s effect on PD progression in terms of cognition, mood and quality of daily life.
Isradipine
List of Isradipine clinical trials
Isradipine is an FDA approved dihydropyridine calcium channel antagonist that treats high blood pressure. Ca2+ influx through calcium channels in neurons stimulates mitochondrial metabolism to increase bioenergetics to meet demands. However, prolonged Ca2+ influx can result in extended mitochondria enhancement leading to the production of ROS, which damage components of the cell. Isradipine attenuates this Ca2+ influx.
It was discovered that chronic treatment of isradipine reduces mitophagy and mitochondria oxidative stress and increases mitochondrial mass [82], suggesting a reduction in enhanced mitochondria activity and thus damage. Isradipine also prevented neuron loss and PD symptoms in the 6-OHDA zebrafish model [83] and the MPTP mouse model of PD [84].
The first phase 2 clinical trial by Northwestern University was a small-scale trial to determine the safety of isradipine in pwp. The second phase 2 trial was larger, recruiting 91 patients and assessed dosage tolerability and efficacy of isradipine.
These enabled isradipine to reach phase 3 clinical trials. This trial enrolled 336 patients across 56 centres in the US and Canada with only 16 participants dropping out. Each participant was assessed over 36 months and required to test their blood pressure twice daily as a precaution due to isradipine’s role as a blood pressure medication. 5 mg isradipine tablets were taken twice daily for the course of the trial.
Secondary outcome measures were assessed as the change from baseline to 36 months in the following:
medication, for example levodopa dose, to determine progression of motor symptoms
each section of the UPDRS assessment individually, to measure motor and non-motor symptoms separately
modified Rankin score, ambulatory capacity and Parkinson’s disease questionnaire 39 to measure the level of disability
Montreal cognitive assessment to detect mild cognitive impairment
Beck depression inventory score to measure depression and mood
EPI-589
EPI-589 is an antioxidant molecule with free radical scavenging activity which protects against oxidative stress. Its reduced form can scavenge free radicals. EPI-589 is also able to oxidise itself back to its original form [87]. It has been demonstrated that EPI-589 protects against oxidative stress and prevents the 8-hydroxy-deoxyguanosine increase in urine [88].
This study by Edison Pharmaceuticals Inc is the first study to be completed for EPI-589. In this study they tested the safety of EPI-589 in PD patients as well as EPI-589’s potential to alter the biochemical signature of PD by assessing peripheral blood biomarkers, central nervous system (CNS) biomarkers, and urine biomarker analysis.
Movement Disorder Society Sponsored Revision of the Unified Parkinson’s Disease Rating Scale (MDS-UPDRS)
Non-motor Symptoms Scale (NMSS)
Parkinson’s Disease Questionnaire - 39 (PDQ-39)
EuroQol-5 Dimension (EQ-5D)
Montreal Cognitive Assessment (MoCA) Score
Beck Depression Inventory (BDI) Score
Montgomery and Asberg Depression Rating Scale (MADRS)
Maximum Observed Plasma Concentration (Cmax) of PTC589
Level of Disease-Related Biomarker (Glutathione) in Plasma
Level of Disease-Related Biomarker (Glutathione) in Cerebrospinal Fluid (CSF)
Level of Disease-Related Biomarker (Glutathione) in Urine
