Mitochondrial biogenesis as a therapeutic target for traumatic and neurodegenerative CNS diseases

Abstract

Central nervous system (CNS) diseases, both traumatic and neurodegenerative, are characterized by impaired mitochondrial bioenergetics and often disturbed mitochondrial dynamics. The dysregulation observed in these pathologies leads to defective respiratory chain function and reduced ATP production, thereby promoting neuronal death. As such, attenuation of mitochondrial dysfunction through induction of mitochondrial biogenesis (MB) is a promising, though still underexplored, therapeutic strategy. MB is a multifaceted process involving the integration of highly regulated transcriptional events, lipid membrane and protein synthesis/assembly and replication of mtDNA. Several nuclear transcription factors promote the expression of genes involved in oxidative phosphorylation, mitochondrial import and export systems, antioxidant defense and mitochondrial gene transcription. Of these, the nuclear-encoded peroxisome proliferator-activated receptor γ coactivator 1α (PGC-1α) is the most commonly studied and is widely accepted as the ‘master regulator’ of MB. Several recent preclinical studies document that reestablishment of mitochondrial homeostasis through increased MB results in inhibited injury progression and increased functional recovery. This perspective will briefly review the role of mitochondrial dysfunction in the propagation of CNS diseases, while also describing current research strategies that mediate mitochondrial dysfunction and compounds that induce MB for the treatment of acute and chronic neuropathologies.

1. Introduction

Mitochondria are vital in maintaining cellular functions, such as amino acid synthesis, fatty acid metabolism, adenosine triphosphate (ATP) production, apoptosis, ion homeostasis, antioxidant defenses and reactive oxygen species (ROS) regulation (Golpich et al., 2017). As such, mitochondria play an extensive role in mediating metabolic pathways and tissue function. Mitochondria are highly dynamic organelles with the ability to routinely modify their size, shape and organization in response to internal and external stimuli (van der Bliek et al., 2013). Dynamic regulation of mitochondrial morphology and content, including fission/fusion, mitophagy and biogenesis, allow mitochondria the plasticity and protection necessary to meet the metabolic needs for cell repair and regeneration while in the presence of cellular stress (Friedman and Nunnari, 2014).

In the presence of toxins or other stressors, mitochondria can become damaged and dysfunctional, leading to a wide breadth of consequences, including hindered oxidative phosphorylation and ATP production, depolarization of the mitochondrial membrane, mitochondrial DNA (mtDNA) fragmentation, oxidative stress, impaired calcium homeostasis, altered mitochondrial dynamics and activation of apoptotic pathways (Funk and Schnellmann, 2012; Gibbs et al., 2016; Gibson et al., 2010; McGill et al., 2012; Morais and De Strooper, 2010; Pisano et al., 2016). Therefore, mitochondrial dysfunction is characteristic of a multitude of acute and chronic diseases, including those of the central nervous system (CNS).

Because disruption of mitochondrial quality control mechanisms has been implicated in various diseases within the CNS, therapeutic strategies aimed at maintaining and/or restoring mitochondrial homeostasis and related cellular processes are becoming increasingly popular. Specifically, pharmacological induction of mitochondrial biogenesis (MB), the generation of new, functional mitochondria, is a promising therapeutic target for a wide range of acute and chronic diseases characterized by mitochondrial dysfunction (Cameron et al., 2016; Scholpa and Schnellmann, 2017; Whitaker et al., 2016). Fortunately, there exist several pharmacological agents known to induce MB that are approved by the U.S. Food and Drug Administration (FDA) and repurposing these drugs for the treatment of various CNS pathologies could be a relatively expeditious process.

This perspective will review the role of mitochondrial dysfunction in the propagation of CNS diseases, as well as describe current research strategies that mediate mitochondrial dysfunction and compounds that induce MB for the treatment of acute and chronic CNS diseases (Scholpa and Schnellmann, 2017).

2. Mitochondrial function/dysfunction

Traumatic and neurodegenerative CNS pathologies are frequently characterized by impaired mitochondrial bioenergetics and disturbed mitochondrial dynamics (Bordone et al., 2019; Reddy and Beal, 2005; Scholpa and Schnellmann, 2017). As expected, the dysregulation observed in many these diseases leads to defective respiratory chain function and reduced ATP production, promoting neuronal death (Burte et al., 2015; Sebastian et al., 2017). A comprehensive review detailing mitochondrial function in the CNS can be found in Smith and Gallo (2018).

2.1. Mitophagy

Mitochondrial dysfunction leads to apoptotic cell death in the presence of increased ROS production, calcium accumulation, opening of mitochondrial permeability transition pore (mPTP) and release of cytochrome c (cyt c) (Rodriguez-Enriquez et al., 2009; Sims and Muyderman, 2010). Furthermore, loss of mitochondrial function can initiate mitochondrial autophagy, or mitophagy, a process of selective mitochondrial degeneration where mitochondrial derivatives are engulfed and transported to the lysosome or peroxisome for degradation (Pickrell and Youle, 2015). A well-studied pathway of mitophagy is the PINK1/Parkin-dependent mitophagy pathway. This pathway is activated in the presence of mitochondrial damage and destabilization of ubiquitin kinase-induce kinase 1 (PINK1). Once released from the outer membrane of the mitochondria, PINK1 will recruit, phosphorylate and activate E3 ubiquitin ligase Parkin. Both PINK and Parkin will then ubiqitinate mitochondrial proteins and promote the formation of autophagosomes. Finally, autophagosomes will fuse with a lysosome, where degradation will take place (Fig. 1) (Fivenson et al., 2017; Narendra et al., 2010; Pickrell and Youle, 2015).

Fig. 1. Mitochondrial dynamics and mitophagy.

Three GTPases mediate the process of mitochondrial fusion: mitofusin (Mfn) 1 and 2, both existing on the outer membrane, and optic atrophy 1 (Opa1) located on the inner membrane. Mfn1 and Mfn2 initiate fusion of the outer membrane of two mitochondria by forming homo-oligomeric and hetero-oligomeric fusion complexes. Activation of the lipid-binding domain within Opa1 then create a pore in the membrane, completing mitochondrial fusion. The process of fission is initiated by recruitment of dynamin-related protein 1 (Drp1) from the cytosol to the mitochondrial membrane, where it can interact with Drp1 receptors on the outer mitochondrial membrane, including mitochondrial fission 1 protein (Fis1). Drp1 oligomerizes to form a ring-like structure around the mitochondria. The mitochondrial membrane can then be split via GTP hydrolysis. Excessive fission may promote mitophagy. The PINK1/Parkin-dependent mitophagy pathway is activated in the presence of mitochondrial damage and destabilization of ubiquitin kinase-induce kinase 1 (PINK1). Once released from the outer membrane of the mitochondria, PINK1 will recruit, phosphorylate (P) and activate E3 ubiquitin ligase Parkin. Both PINK and Parkin will then ubiqitinate (U) mitochondrial proteins and promote the formation of autophagosomes. Finally, autophagosomes will fuse with a lysosome where degradation will take place.

Mitophagy is essential in sustaining mitochondrial homeostasis, biogenesis and total number and quality of mitochondria (Golpich et al., 2017). Mitophagy has been shown to be neuroprotective and to reduce the production of ROS via clearance of dysfunctional mitochondria (Cao et al., 2017). Excessive ROS can exacerbate injuries of the brain, spinal cord and blood-CNS barriers by inducing apoptosis (Qu et al., 2016). Previous studies have shown defective mitophagy in neurodegenerative diseases, leading to the aggregation of autophagosomes, abnormal endosomes and abnormal lysosomes. This dysfunction hinders the mitophagy process, which likely contributes to the pathology of neural degeneration (Franco-Iborra et al., 2018; Ni et al., 2015). As such, insufficient or altered mitophagy can lead to cell death (Murphy, 2009) and may promote the development and propagation of many CNS-related diseases (Golpich et al., 2017).

2.2. Necrosis

Necrosis is an irreversible cell death process characterized by rapid loss of cellular membrane potential thereby leading to swelling of the cell, rupture and subsequent inflammation (Zong and Thompson, 2006). The mPTP is a key effector in the pathway to cell death (Baines et al., 2003; Karch et al., 2013; Karch and Molkentin, 2014; Xu et al., 2019) in that opening of the mPTP further exacerbates mitochondrial dysfunction (Baines et al., 2003; Karch et al., 2013; Karch and Molkentin, 2014). Recent studies suggest inhibition of mPTP opening may regulate programmed necrosis and limit neuronal loss in several CNS injuries (Xu et al., 2019; Ying and Padanilam, 2016). For additional information detailing the regulation of necrotic cell death, see Ying and Padanilam (2016).

2.3. Fission

Fusion and fission are two inverse processes encompassing mitochondrial dynamics that directly contribute to morphological changes of mitochondria (Fig. 1). Fusion mechanisms promote tethering and joining of two mitochondria, whereas fission initiates cleavage and division of mitochondria (Chan, 2006; Okamoto and Shaw, 2005; Westermann, 2008). Coordination of these processes maintain the growth, shape, distribution and structure of mitochondria, which is integral to homeostasis, cell stability and cell survival (Calo et al., 2013; Twig and Shirihai, 2011). Disruption to these processes may result in altered mitochondrial function and cellular bioenergetics, thereby contributing to the onset and propagation of several neuropathologies (Calkins et al., 2011; Filosto et al., 2011). Additionally, disruptions in mitochondrial dynamics may be a requirement for mitophagic mechanisms (Golpich et al., 2017).

Dysfunctional mitochondria contain impaired proteins, damaged membranes and mutated or fragmented mtDNA, all of which promote the division and fragmentation of mitochondria via activation of fission (Frank et al., 2012; Scott and Youle, 2010). Mitochondrial fission is initiated when cells require elimination of damaged mitochondria. The fission process is mediated by a family of dynamin-related proteins (Drps), particularly Drp1 (Boldogh and Pon, 2006; Hollenbeck and Saxton, 2005). This protein, once recruited from the cytosol to the mitochondrial membrane, can interact with Drp1 receptors on the outer mitochondrial membrane, including mitochondrial fission factor (Mff) and mitochondrial fission 1 protein (Fis1). Subsequently, Drp1 oligomerizes with Drp2 to form a ring-like structure around the mitochondria. The mitochondrial membrane is then able to be split via GTP hydrolysis (He et al., 2019; Lee et al., 2016; Loson et al., 2013; Otera et al., 2010; Palmer et al., 2011).

Defects in mitochondrial fission can induce mitochondrial dysfunction and related pathologies (Burte et al., 2015; Golpich et al., 2017). Accumulation of damaged or otherwise compromised mitochondria can have negative effects on electron transport chain (ETC) components and inhibit ATP production, resulting in cell death (Galloway et al., 2012; Twig et al., 2008; Twig and Shirihai, 2011). Furthermore, increased mitochondrial fragmentation has been reported in fibroblast cells of patients with neurodegenerative diseases, resulting in defective oxidative phosphorylation and ATP deficiency (Capaldi et al., 2004). Altered mitochondrial fission has also been linked with behavioral abnormalities and mood disorders. A recent study in mice revealed stress-induced triggers can promote severe mitochondrial fission in peripheral CD4⁺ T cells resulting in inflammation-induced anxiety-like behavior (Fan et al., 2019).

Drp1 expression is upregulated during oxidative stress, disturbing the balance of mitochondrial dynamics, and leading to mitochondrial dysfunction and eventual cell death (Wu et al., 2011). Reports demonstrate a reduction in Drp1 expression and mitochondrial fragmentation with antioxidant treatment, such as vitamin E or the antioxidant CoQ10 (MitoQ) (de Arriba et al., 2013; Ferrari et al., 2011). Reports also show increased oxidative stress and ROS production during multiple injury types, which is reduced with knockdown of Drp1 (Ferrari et al., 2011; Kobashigawa et al., 2011; Peng et al., 2011). Dysregulation of fission can also propagate apoptosis, while its inhibition has been shown to hinder the release of cyt c, a potent mediator of cell death pathways, thereby delaying apoptosis (Herzig and Martinou, 2008). Additionally, mitochondrial dysfunction can activate Fis1, further enhancing mitochondrial fission (Loson et al., 2013). These published data, among others, indicate that maintenance of fission regulates mitochondrial quality and bioenergetic properties (Sebastian et al., 2017; Twig et al., 2008; Westermann, 2012), and that dysregulation can contribute to neurodegenerative diseases.

2.4. Fusion

Fusion is the process by which two mitochondria fuse together, merging their membranes and sharing intracellular contents, such as proteins, lipids, and growth factors. Sharing of these components is a crucial aspect to maintaining proper ETC function (Westermann, 2012). Three GTPases control the conservative process of mitochondrial fusion: mitofusin (Mfn) 1 and 2, which exist on the outer membrane, and optic atrophy 1 (Opa1) located on the inner membrane (Escobar-Henriques and Anton, 2013; Ishihara et al., 2006; Santel and Fuller, 2001). Mfn1 and Mfn2 initiate the fusion of the outer membrane of two mitochondria by forming homo-oligomeric and hetero-oligomeric fusion complexes (Chernomordik and Kozlov, 2005; Hoppins et al., 2007). Then, activation of the lipid-binding domain within Opa1 creates a pore in the membrane, completing mitochondrial fusion (Hoppins et al., 2007; Meglei and McQuibban, 2009) (Fig. 1). Fusion contributes to the balance of matrix metabolites, while also aiding in the equilibrium of mitochondrial membrane contents, particularly that of complex I of the ETC (Busch et al., 2006; Dimmer and Scorrano, 2006; Ono et al., 2001).

Like fission, regulation of this process is critical for cell survival during all stages of life, from embryonic and cortical development, to everyday homeostatic mitochondrial function. With impaired fusion, mitochondrial fragmentation increases, resulting in decreased expression and distribution of mitochondrial-encoded proteins of the ETC and inhibition of ATP synthesis, both of which lead to neuronal death (Westermann, 2012). Reports reveal that enhanced fusion aids in the maintenance of mitochondrial integrity in healthy mitochondria (Benard and Karbowski, 2009); however, hindered activity of fusion may contribute to the development of neurodegenerative disorders (Zuchner et al., 2006; Zuchner et al., 2004). Additional studies demonstrate enhanced mitochondrial fusion in response to stress stimuli (e.g. UV irradiation) or exposure to known stress-inducing drugs. In this hyperfusion state, mitochondria form highly interconnected and elongated networks of mitochondria and have increased ATP synthesis (Mouli et al., 2009; Westermann, 2012). Maintaining a balance between fission and fusion activity is crucial for ensuring mitochondrial homeostasis, while disturbance of mitochondrial dynamics promotes and exacerbates an array of CNS pathologies.

3. Mitochondrial biogenesis (MB)

Given the vital importance of mitochondria to a plethora of cellular functions, it is not surprising that dozens of diseases, including many within the CNS, are characterized by mitochondrial dysfunction (Bordone et al., 2019; Golpich et al., 2017; Sheng et al., 2012). While many mitochondrial genes are encoded by the nuclear genome, mitochondria also contain their own circular genome, which is composed of 13 essential genes required for successful mitochondrial function. As such, induction of MB is dependent on the coordinated activation of both the mitochondrial and nuclear genomes (Scarpulla, 2008).

MB is a multifaceted process involving the integration of highly regulated transcriptional events, lipid membrane and protein synthesis/assembly and replication of mtDNA (Fig 2) (Ventura-Clapier et al., 2008). Furthermore, impaired MB can contribute to mitochondrial and cellular dysfunction (Golpich et al., 2017). Several nuclear transcription factors promote the expression of genes involved in oxidative phosphorylation, mitochondrial import and export systems, antioxidant defense and mitochondrial gene transcription (Fig 2). Of these, the nuclear-encoded peroxisome proliferator-activated receptor γ coactivator 1α (PGC-1α) is the most commonly studied and is widely accepted as the ‘master regulator’ of MB (Kelly and Scarpulla, 2004; Ventura-Clapier et al., 2008).

Fig. 2. Mitochondrial dysfunction and biogenesis.

Mitochondrial dysfunction is characterized by decreased expression of oxidative phosphorylation (OXPHOS) proteins, impaired mitochondrial membrane potential, reduced ATP production as well as enhanced mitochondrial fission, reactive oxygen species (ROS) production and cytochrome c (cyt c) release. Mitochondrial dysfunction is often paired with increased apoptosis and neural degeneration. Conversely, mitochondrial biogenesis (MB) is characterized by mitochondrial fusion, reduced ROS, restoration of mitochondrial membrane potential and increased expression of OXPHOS proteins including ATP synthase β (ATPsyn β), NADH:Ubiquinone oxidoreductase core subunit 1 (NDUFS1) and uncoupling protein 2 (UCP2) as well as superoxide dismutase 2 (SOD2). MB induction often begins with activation of encoded peroxisome proliferator-activated receptor γ coactivator 1α (PGC-1α) that co-activates nuclear respiratory factors (NRFs), which then promotes the transcription of several mitochondrial genes, including mitochondrial transcriptional factor a (TFAM). Once translated, TFAM translocates to the mitochondrial matrix and stimulates mtDNA replication and mitochondrial gene expression. MB is associated with enhanced cellular function and neuronal survival.

PGC-1α serves as a docking platform, recruiting additional transcription factors for the transcription of nuclear-encoded genes necessary for MB. Activation of PGC-1α initiates activation of nuclear respiratory factors (NRFs), which then promote the transcription of several mitochondrial genes, including various subunits of the ETC, such as ATP synthase, cyt c, cytochrome oxidase IV and mitochondrial transcriptional factor a (TFAM) (Cameron et al., 2016; Wu et al., 1999). Once translated, TFAM translocates to the mitochondrial matrix and stimulates mtDNA replication and mitochondrial gene expression (Ventura-Clapier et al., 2008). PGC-1α also has the capacity to interact with other transcription factors, such as peroxisome proliferator-activated receptors (PPARs), thyroid hormone, glucocorticoids, estrogen receptors, mitochondrial transporters, and antioxidant proteins (Ventura-Clapier et al., 2008). The transcription and translation of the aforementioned genes leads to MB and enhanced mitochondrial function. The expression of PGC-1α is highly inducible upon activation of physiological cues signaling for increased cellular metabolic needs, including oxidative stress and cell division (Handschin and Spiegelman, 2006). A variety of both central- and peripheral-related diseases are characterized by mitochondrial dysfunction, with several reports detailing a decrease in PGC-1α expression in both traumatic and neurodegenerative diseases (Scholpa et al., 2018a; Scholpa and Schnellmann, 2017). As such, current research aims to attenuate this dysfunction by targeting the activation of PGC-1α as a therapeutic strategy for inducing MB, as well as for evaluating mitochondrial homeostasis. Furthermore, ischemic injury, such as that which occurs with SCI, is also followed by reduced oxidative phosphorylation proteins as well as decreased PGC-1α and TFAM (Scholpa and Schnellmann, 2017; Whitaker et al., 2016). For more information detailing mechanisms and inducible compounds of MB induction, see Gibbs et al. (2018c) and Cameron et al. (2016). Reports discussed herein support the role of MB in restoring mitochondrial dynamics, increasing oxidative phosphorylation, attenuating expression of mitochondrial proteins and increasing cellular functions and neuronal survival, thereby promoting recovery in CNS diseases (Fig 2). See Table 1 for a partial list of CNS diseases characterized by mitochondrial dysfunction and evidence of pharmacological induction of MB as therapeutic interventions for each.

Table 1.

Partial list of CNS diseases characterized by mitochondrial dysfunction and evidence of pharmacological induction of MB as therapeutic interventions for each.

Disease	Mitochondrial Dysfunction	Tested MB Treatments	References
Acute Diseases
Traumatic Brain Injury	Ischemia: ↓OXPHOS proteins, ↓ATP synthesis, ↓ATP-dependent cellular processes, ↓ion homeostasis Reperfusion: ↑ROS, ↑oxidative stress, ↑Ca2+, mPTP opening, ↑cell death	Quercetin: ↑PGC-1α, ↓cell death DEX: ↑PGC-1α, ↓cell death, ↑behavior 7,8-DHF: ↑PGC-1α, ↑function SS-31: ↑PGC-1α, ↑function, ↑behavior	(Li et al., 2018c) (Li et al., 2018a) (Krishna et al., 2017) (Zhu et al., 2018)
Spinal Cord Injury		Formoterol: ↑PGC-1α, ↓cell death, ↑function LY344864: ↑PGC-1α, ↓cell death, ↑function Mitochondrial transplant: ↑pAKT, ↑function, ↓cell death	(Scholpa et al., 2019a; Scholpa et al., 2019b) (Simmons et al., 2019) (Li et al., 2019)
Stroke		Mitochondrial transplant: ↑pAKT, ↓cell death, ↑function Daidzein: ↓cell death, antioxidant Metformin: ↓cell death, ↑behavior Melatonin: ↑mitophagy, antioxidant	(Chang et al., 2019) (Hayakawa et al., 2016) (Cao et al., 2017) (Andrzejewski et al., 2014; Owen et al., 2000) (Qi et al., 2017)
Neurodegenerative
Alzheimer’s Disease	↓MB, ↑mitochondrial size and structural damage, ↓content, ↑fragmentation, ↑mtDNA mutations	TZDs: ↑behavior Resveratrol: ↑PGC-1α, ↑behavior, ↑pAKT AICAR: ↑behavior, ↑pAKT Melatonin: ↑PGC-1α, ↑function, ↓Aβ	(Cheng et al., 2016; Heneka et al., 2015; Risner et al., 2006; Watson et al., 2005) (Vingtdeux et al., 2010) (Wang et al., 2019)
Parkinson’s Disease	↑mtDNA mutations (complex I), ↑oxidative damage, ↓PGC-1α, ↓OXPHOS, ↓homeostasis	Bezafibrate: ↑PGC-1α, ↑behavior Resveratrol: ↑PGC-1α, ↑function, ↓cell death LY344864: ↑PGC-1α, ↑function, ↓cell death Triterpenoids: ↑Nrf2, antioxidant	(Tufekci et al., 2011) (Corona and Duchen, 2016) (Peng et al., 2016) (Scholpa et al., 2018a)

Adaption from (Gibbs et al., 2018c). Arrows indicated increase (↑) and decrease (↓). AICAR: 5-aminoimidazole-4-carboxamide ribonucleotide, ATP: adenosine triphosphate, Ca²⁺: calcium ion, DEX: dexmedetomidine, ETC: electron transport chain, MB: mitochondrial biogenesis, mPTP: membrane permeability transition pore, mtDNA: mitochondrial DNA, PGC-1α: peroxisome proliferator-activated receptor-γ coactivator 1-α, OXPHOS: oxidative phosphorylation, ROS: reactive oxygen species, TFAM: mitochondrial transcription factor A, TZDs: thiazolidinediones, 7,8-DHF: dihydroxyflavone.

3.1. Limitations of pharmacological induction of MB

Though a multitude of studies support targeting MB for treatment of various CNS diseases, this therapeutic avenue is not without limitations. One potential limitation is that enhancing MB may increase unhealthy mitochondrial content. In diseases with disruption of mitochondrial homeostasis, such as the ones discussed in this review, it is possible that increasing MB will exacerbate the negative effects of mitochondrial dysfunction. However, defective or mutated mitochondria may not be able to undergo MB due to their dysfunctional state. In addition, MB and quality control are often upregulated in cancer, and mitochondria are reported to play a central and multifunctional role in malignant tumor progression (Zong et al., 2016). Targeting mitochondrial inhibition (i.e. inhibition of oxidative phosphorylation, promotion of ROS production, inhibition of mitophagy) may provide therapeutic opportunities in the treatment of various cancers (Zong et al., 2016). Therefore, while MB induction may prove an effective therapeutic avenue for the treatment of CNS-related diseases, systemic MB induction may also enhance tumor progression.

Fortunately, multiple pharmacological compounds that induce MB are already approved by the U.S. Food and Drug Administration for the treatment of various pathologies. Therefore, though limitations exist, attaining approval for the use of these drugs for the treatment of various CNS disease could be an expeditious process and provide meaningful treatment for populations that currently have limited therapeutic options.

4. Acute neurological diseases

While the canonical role for CNS mitochondria is the generation of energy, they are also involved in the homeostasis and degeneration of neurons (Dubinsky, 2005). In fact, minimal mitochondrial dysfunction can result in the development of neuropathologies (Dubinsky, 2005). Compromised perfusion is common among acute traumatic CNS injuries, including spinal cord injury (SCI), traumatic brain injury (TBI) and stroke, leading to localized ischemia and subsequent mitochondrial dysfunction. Neuronal cells are easily compromised following ischemic events due to their high reliance on ATP-driven processes, in combination with their inadequate energy reserves and limited capacity to buffer oxidative stress (Castro et al., 1997; Scholpa et al., 2018b; Tian et al., 2016). In addition, axons are more susceptible to the damage caused by ionic imbalance due to their high concentration of voltage gated sodium channels in the nodes of Ranvier (Oyinbo, 2011). Taken together, this manifests in the failure to generate and maintain adequate energy production, thereby exacerbating the pathology of acute CNS injuries, resulting in further cell dysfunction and death (Castro et al., 1997; Scholpa and Schnellmann, 2017; Scholpa et al., 2018b).

4.1. Neuroinflammation

During traumatic/ischemic injuries, several factors are released that initiate the immune response (DiSabato et al., 2016; Simon et al., 2017; Viviani et al., 2014). In the CNS, astrocytes and glial cells are activated in response to injury. These cells are involved in the maintenance and support of neurons and comprise a significant component of the blood-CNS barriers (Viviani et al., 2014). Both astrocytes and microglia, when activated, release cytokines and chemokines that modulate neuroinflammation and development. Astrocyte-derived cytokines including, in part, interleukins and tumor necrosis factor-α (TNF-α) promote neurotoxicity (Shabab et al., 2017). TNF-α can suppress expression of the excitatory amino acid transporter (EAAT) responsible for the routine uptake of glutamate at the synapse. This inhibited uptake exacerbates the already heightened levels of glutamate that were released by damaged and necrotic cells following injury. Glutamate excitotoxicity compromises the cellular membrane potential thereby facilitating calcium influx and overload to the cell. Calcium will then bind to and open the mPTP, promoting cyt c release and ROS production, ultimately inducing apoptosis (Viviani et al., 2014) (Fig 3). It should be noted that opening of mPTP can also promote necrosis via separate mechanisms (Baines et al., 2003; Karch et al., 2013; Karch and Molkentin, 2014). For a more comprehensive review of neuroinflammation see Shabab et al. (2017), Viviani et al. (2014) and DiSabato et al. (2016).

Fig. 3. Role of neuroinflammation on mitochondrial dysfunction.

Injury, ischemia, hypoxia, toxins, inflammation and ROS can trigger activation of astrocytes and microglia. Both astrocytes and microglia, when activated, release cytokines and chemokines. Astrocyte-derived cytokines, including interleukins (IL) and tumor necrosis factor-α (TNF-α) can promote neurotoxicity. TNF-α can suppress expression of the excitatory amino acid transporter (EAAT) to exacerbate glutamate levels. Glutamate will bind to glutamate receptors (Glu-R), facilitating calcium influx and overload to the cell. Calcium will then bind to and open the mitochondrial permeability transition pore (mPTP), promoting cytochrome c (cyt c) release and ROS production, ultimately inducing apoptosis.

5. Traumatic brain injury

5.1. Background and pathology

TBI is a devastating injury, often resulting in long-term neuropsychological deficits, including, but not limited to, impaired attention, memory and cognitive functioning. TBI causes significant morbidity and mortality, largely due to minimally effective diagnostics and treatment options (Wang et al., 2017). TBI is initiated by mechanical damage immediately resulting in vascular injury and tissue destruction (Ladak et al., 2019). This primary phase of injury is followed by blood-brain barrier disruption and hypoxia, inducing glial cell activation, astrogliosis, inflammation and further cell death (Ladak et al., 2019; Wanner et al., 2013), all of which encompass a secondary wave of events characterized by a cascade of detrimental biochemical and pathophysiological stressors, referred to as secondary injury. These secondary stressors contribute to ongoing cell death and dysfunction, including glutamate excitotoxicity, mitochondrial dysfunction, free radical-mediated oxidative damage, inflammation and activation of necrotic and apoptotic cell death signaling pathways (Wang et al., 2017).

Numerous studies have documented considerable impairments in mitochondrial function in the injured brain following TBI, such as calcium overload, enhanced ROS production, decreased expression of PGC-1 α and damaged mtDNA (Clark et al., 2000; Gilmer et al., 2009; Raghupathi et al., 2000; Singh et al., 2006; Sullivan et al., 2002). Additionally, the magnitude of mitochondrial dysfunction post-TBI is a critical determinant of cell survival, tissue sparing and functional recovery (Lifshitz et al., 2004; Pandya et al., 2013; Sullivan et al., 1999). Moreover, Liftshitz et al., suggested the presence of a mixed mitochondrial response within the brain, such that structural changes in cortical regions post-TBI result in prominent loss of mitochondrial number, whereas hippocampal mitochondria are more prone to swelling and a transient loss of calcium buffering capacity (Lifshitz et al., 2003). Furthermore, reports document differences between neuronal and non-neuronal mitochondria (Bambrick et al., 2006). One study indicates isolated neuronal mitochondria have differing sensitivities to calcium-induced opening of the mPTP compared to isolated astrocytes, with the astrocytes displaying a fourfold higher calcium uptake capacity (Bambrick et al., 2006). Another study found that isolated neurons and astrocytes many have distinct mechanisms of neuroprotection in response to cyclosporine A treatment (Kahraman et al., 2011). This heterogeneity suggests that mitochondria have highly variable responses to injury across specific brain regions (Wang et al., 2017). With this in mind, combinatorial treatments may be required to address deficits seen in TBI.

Previous studies aimed at targeting mitochondrial dysfunction after TBI have proven successful in preclinical models. Mitochondrial uncoupler 2,4-dinitrophenol (DNP) was reported to have a neuroprotective role resulting in greater tissue and neuronal sparing, as well as improved behavioral outcomes following TBI (Korde et al., 2005; Maragos et al., 2003; Pandya et al., 2007; Pandya et al., 2009). Though preclinical data is promising, observed benefits have not yet been validated in the clinic. (Hubbard et al., 2018). Another mitochondrial uncoupler and prodrug of DNP, MP201, has shown therapeutic potential when delivered acutely after TBI in vivo. This study documented improved mitochondrial function, enhanced oxidative phosphorylation and restoration of ROS levels, paired with improved cognitive function and cortical sparing (Hubbard et al., 2018). Finally, upregulation of mitochondrial uncoupling protein 2 (UCP2) was shown to decrease ROS production and cell death, while reducing UCP2 levels elevated ROS (Sullivan et al., 2003; Sullivan et al., 2004). After TBI, overexpression of UCP2 also reduced ROS production and increased cortical sparing (Mattiasson et al., 2003). While the aforementioned preclinical studies have yielded promising data, currently no drugs have been FDA-approved for the treatment of TBI (Hubbard et al., 2018).

5.2. Current MB strategies

Given the detrimental ongoing effects of secondary injury post-TBI, mitigating this spread of injury via MB induction has been a focus for TBI research. Reports suggest that quercetin, a dietary flavonoid used as a food supplement, can induce MB via PGC-1α activation, thereby attenuating brain injury in a TBI mouse model (Li et al., 2018c). Quercetin reduced TBI-induced neuronal apoptosis in these studies and ameliorated mitochondrial lesions. Treatment with quercetin also restored the level of cyt c, and superoxide dismutase in mitochondria, suggesting that mitochondrial dysfunction was attenuated (Li et al., 2016; Li et al., 2018c).

Dexmedetomidine (DEX) is a selective α₂-adrenergic receptor agonist associated with sedative and analgesic effects (Keating, 2015). Following a weight-drop model of TBI in rats, DEX treatment upregulated PGC-1α expression, while also attenuating encephala edema and apoptosis, resulting in increased behavioral function (Li et al., 2018a).

Tropomyosin receptor kinase B (TrkB) has recently emerged as a regulator of hippocampal long-term potentiation and learning (Seese et al., 2019). Treatment with the TrkB agonist, 7,8-dihydroxyflavone (7,8-DHF) following moderate fluid percussion injury in rats showed restored levels of PGC-1α and AMPK, while also augmenting synaptic plasticity and enhancing hippocampal functional connectivity (Krishna et al., 2017).

Finally, the antioxidant SS-31, a mitochondria-targeted peptide known to reduce ROS levels, provided neuroprotection in a variety of neurological diseases. In mouse models of TBI, SS-31 treatment beginning 30 min after injury restored expression of PGC-1α, increased activity of superoxide dismutase (SOD) and decreased cyt c release, all while attenuating neurological deficits, DNA damage and neural apoptosis (Zhu et al., 2019; Zhu et al., 2018). Taken together, these data demonstrate induction of PGC-1α and MB may be a viable treatment strategy to improve mitochondrial function, neuroprotection and behavioral recovery after TBI (Zhu et al., 2019; Zhu et al., 2018).

6. Spinal cord injury

6.1. Background and pathology

Spinal cord injury (SCI) is a traumatic event that can generate an array of impairments ranging from loss of function to complete paralysis below the injury site (Scholpa and Schnellmann, 2017). The clinical outcomes of SCI are diverse and may include loss of sensory and/or motor function, paraplegia or tetraplegia (Alizadeh et al., 2019). Similar to TBI, acute SCI is comprised of a primary injury with subsequent secondary injury resulting from a progressive local cascade of events promoting dysfunction and damage (Witiw and Fehlings, 2015). Primary injury results from a traumatic insult leading to mechanical damage of the spinal cord. As is seen following TBI, disruption of the vasculature causes hemorrhage and edema within seconds, impairing perfusion and ultimately facilitating a localized ischemic event (Witiw and Fehlings, 2015). Ischemia is a key mechanism of secondary injury post-SCI, with the degree of functional loss being proportional to the degree of ischemia (Tator and Fehlings, 1991). This sensitivity leads to compromised oxidative phosphorylation among other mitochondrial function in the presence of ischemia. Additional consequences of secondary injury include neuronal cell death (Anwar et al., 2016; Beattie et al., 2002), progressive axon demyelination (Totoiu and Keirstead, 2005), inflammation (Qiao et al., 2010; Qiao et al., 2015) and mitochondrial dysfunction (Scholpa and Schnellmann, 2017). Loss of mitochondrial homeostasis results in decreased ATP production and inactivation of ATP-dependent ion pumps that are essential for regulation of ion concentrations and reuptake of the excitatory neurotransmitter glutamate. This facilitates excitotoxicity, calcium overload, and the eventual initiation of cell death cascades, all of which are hallmarks of SCI that further exacerbate injury (Choi and Rothman, 1990; Oyinbo, 2011; Rowland et al., 2008).

Based on temporal data conducted by Sullivan et al. (2007), restoring mitochondrial function acutely after injury may be an advantageous strategy for the treatment of SCI (McEwen et al., 2011; Rabchevsky et al., 2011; Sullivan et al., 2007). Many pharmaceuticals that have shown that beneficial outcomes for the treatment of SCI in vivo affect mitochondria or mitochondrial function. For example, treatment with the antibiotic minocycline after SCI stimulated mitochondrial stabilization, inhibited release of cyt c, increased antioxidant activity and impeded mitochondrial-dependent cell death (Aras et al., 2015b; Casha et al., 2012; Wells et al., 2003). Minocycline also displayed neuroprotective effects along with behavioral and cellular recovery when administered post-SCI in rats (Ahmad et al., 2016; Aras et al., 2015b; Casha et al., 2012; Sonmez et al., 2013; Teng et al., 2004; Wells et al., 2003).

Studies also support beneficial outcomes following SCI after restoring mitochondrial dynamics equilibrium. The mechanistic target of rapamycin (mTOR) pathway plays a role in modulating mitochondrial functions (Morita et al., 2015). Rapamycin, an inhibitor of mTOR, may reduce neuronal death by enhancing mitophagy, activating autophagy pathways and attenuating apoptosis in the injured spinal cord (Li et al., 2018b; Sekiguchi et al., 2012; Song et al., 2015). Mitochondrial division inhibitor-1 (Mdivi-1), a selective Drp1 inhibitor, has proven beneficial in models of various CNS pathologies (Luo et al., 2013; Wu et al., 2016). In rat studies, Mdivi-1 treatment prior to SCI increased ATP and mitochondrial membrane potential, decreased caspase-3 release and the number of apoptotic cells, all while increasing locomotor activity (Liu et al., 2015). Additional studies revealed Mdivi-1 treatment increased endogenous antioxidant activity, decreased ROS and decreased cyt c release in cultured spinal cord neurons with glutamate-induce injury (Liu et al., 2015; Scholpa and Schnellmann, 2017). Cyclosporine A (CsA) is an immunosuppressant that inhibits mPTP opening. Treatment with a derivative of CsA, NIM811, reduced oxidative damage and attenuated mitochondrial function and tissue sparring following SCI, while also significantly improving locomotion, tissue sparing and bladder control in rodents (Springer et al., 2018).

Acetyl-l-carnitine (ALC), a metabolite transported through the inner mitochondrial membrane capable of acetyl-CoA synthesis (Pettegrew and McClure, 2002), has shown beneficial effects in several pathologies, including Parkinson’s disease, Alzheimer’s disease, and multiple sclerosis (Pettegrew and McClure, 2002; Puca et al., 1990; Tomassini et al., 2004). Studies report ALC treatment after SCI in rats reduced neuronal degeneration and neuroinflammation (Karalija et al., 2012, 2014), maintained mitochondrial function, improved functional recovery, protected both white and gray matter (Cohen et al., 2009; Patel et al., 2012; Patel et al., 2010), and reduced the number of damaged mitochondria. Treatment also improved mitochondrial membrane potential, and decreased SCI-induced apoptosis (Zhang et al., 2015). These data, though preclinical, provide evidence for ALC as a potential therapeutic treatment of SCI.

6.2. Current MB strategies

Given the aforementioned studies as well as others that have targeted mitochondrial function post-SCI, pharmacological activation of MB represents a promising approach. In support of this, PGC-1α expression is decreased in the spinal cord after contusive SCI in rats (Hu et al., 2015; Hu et al., 2016) and mice (Scholpa et al., 2018b; Simmons et al., 2020). In addition, spinal lentiviral overexpression of PGC-1α immediately after injury attenuated neuronal cell death and promoted functional recovery (Hu et al., 2015; Hu et al., 2016). Evidence also indicates that pharmacological-induced restoration of mitochondrial homeostasis promptly following injury may improve neuronal survival and promote functional recovery (Rabchevsky et al., 2011; Scholpa and Schnellmann, 2017; Scholpa et al., 2019a; Sullivan et al., 2007), all of which is suggestive of the potential benefit of pharmacologically increasing PGC-1α and MB following injury.

Studies have also demonstrated a positive correlation between PGC-1α and angiogenesis (Arany et al., 2008; Saint-Geniez et al., 2013; Thom et al., 2014). As compromised vasculature is a delirious consequence of SCI, particularly blood-spinal cord barrier dysfunction, enhanced angiogenesis could serve as an effective treatment for SCI. Therefore, therapeutics targeting reestablishment of mitochondrial homeostasis through MB activation may unveil a therapeutic avenue for mediating several facets of secondary injury progression, improving functional and vascular recovery and neuronal survival following SCI (Scholpa and Schnellmann, 2017).

A recent report demonstrated transplantation of mitochondria from bone marrow mesenchymal stem cells to injured neurons may be neuroprotective in the injured spinal cord of rats. Transplantation resulted in improved bioenergetics, improved activation/phosphorylation of an intermediate signaling molecule of MB, protein kinase B (pAKT), and mitochondrial respiration, decreased apoptosis, enhanced motor neuron survival and improved locomotor recovery (Li et al., 2019). This study supports the theory that restored mitochondrial number and function promote recovery following SCI. However, pharmacological activation of MB may serve as a more efficient approach for promoting mitochondrial function and locomotor recovery.

Pharmacological activation of MB can be initiated via agonism of the G protein-coupled 5-hydroxytryptamine 1F receptor (5-HT_1FR) and β₂-adrenergic receptor (β₂R). Formoterol is a well-studied agonist of the β₂R and potent mediator of MB, (Gibbs et al., 2018a; Gibbs et al., 2018b; Wills et al., 2012) and is currently FDA-approved for the treatment of asthma and related disorders. Reports show that formoterol treatment post-SCI has prominent effects in the injured spinal cord, including enhanced PGC-1α, increased mitochondrial number, mediation of mitochondrial dysfunction and improved locomotor capabilities in mice (Scholpa et al., 2019a; Scholpa et al., 2019b). In these studies, formoterol treatment also increased body weight and skeletal muscle mass. Similar mitochondrial and locomotor recovery was seen with treatment of the selective 5-HT_1FR agonist LY344864. In addition, treatment with this 5-HT_1FR agonist improved integrity of the blood-spinal cord barrier after injury (Simmons et al., 2020). Though LY344864 failed phase II clinical trials, the more selective 5-HT_1FR agonist lasmiditan was recently approved for the treatment of migraines (Brandes et al., 2019; Shapiro et al., 2019). With both of these drugs already FDA-approved for the treatment of other pathologies, repurposing these drugs for the treatment of SCI could accelerate approval for treatment of SCI.

7. Stroke

7.1. Background and pathology

Stroke is the leading cause of adult disability and mortality in most developing and developed countries (Guzik and Bushnell, 2017). Ischemic events occur when blood flow to brain tissue is decreased or blocked. In an ischemic stroke patient, a significant decline in the focal cerebral blood flow leads to deprivation of glucose and oxygen, which rapidly compromises biochemical functions, leading to cell death and ultimately brain damage. (Yang et al., 2018). As is common with acute/traumatic injuries, the downstream signaling pathways of stroke induce glutamate excitotoxicity and excessive calcium influx leading to mitochondrial dysfunction and ROS production (Galluzzi et al., 2012). Such events initiate various pathological processes such as mtDNA damage, mitochondrial fission and fusion changes, mitophagy and apoptosis (Dharmasaroja, 2016; Hofmeijer and van Putten, 2012; Lee et al., 2000).

The current therapy for treatment of ischemia is to quickly restore blood flow to the compromised tissue. As such, the standard of care for ischemic stroke is to remove the blockage from the cerebral artery via thrombolysis (Adeoye et al., 2011) or angiographic revascularization (Zaidat et al., 2013). Unfortunately, a major generator of ischemic injuries is the restoration of blood flow to the ischemic tissue, known as ischemia and reperfusion (IR) injury (Chouchani et al., 2016; Dawson and Dawson, 2017; Lesnefsky et al., 2017; Sanderson et al., 2013). Mitochondria and other systems have the potential to generate a considerable quantity of ROS during the reperfusion process, further compromising mitochondrial and cellular processes and functions. IR is a key driver of the pathology of stroke (Chouchani et al., 2016; Sanderson et al., 2013); therefore, therapeutic research now aims to minimize the effects of IR-induced injury following ischemia (He et al., 2019; Lesnefsky et al., 2017).

Inflammation is another pivotal mechanism in the pathology of ischemic stroke. Recent studies have recognized the roles of mitochondria in the regulation of the inflammatory response (DiSabato et al., 2016). MitoQ is capable of hindering production of cytokines in peripheral blood monocytes (Boldogh and Pon, 2006). Studies using MitoQ treatment for ischemic stroke, while limited, suggest a neuroprotective role for MitoQ in adult ischemic studies and restoration of mitochondrial function may play a role in promoting these observed effects (Calo et al., 2013; Loson et al., 2013; Thornton et al., 2018).

Since mitochondrial dysfunction is a known hallmark of ischemic stroke, restoring function is essential to allow for cell survival and neurological improvement after ischemia. Several studies report a neuroprotective effect of ALC, an acetylated derivative of L-carnitine, in ischemic stroke models. In an animal model of global cerebral ischemia and reperfusion, reports show treatment with ALC normalized brain energy metabolites and improved neurological outcomes (Rosenthal et al., 1992). ALC also protected against early changes in brain metabolites and mitochondrial function following a rat model of perinatal hypoxia ischemia of both male and female pups (Tang et al., 2016). In primary cultures of rat cortical neurons, ALC may inhibit both acute and delayed cell death following excitotoxic injury, potentially via inhibition of mPTP opening (Zanelli et al., 2005). Following calcium exposure in primary rat cortical neurons and astrocytes, PTP inhibitors such as CsA showed reduced calcium sensitivity of the mPTP and membrane depolarization, though the mechanism explaining this protective effect may not be through mPTP inhibition (Kahraman et al., 2011).

In addition, several reports demonstrated that treatment with Mdivi-1 improves outcomes after oxygen/glucose deprivation in the brain (Ma et al., 2016; Wang et al., 2014; Wappler et al., 2013; Zhao et al., 2014) and also attenuates brain injury in several models of stroke and TBI (Chuang et al., 2016; Thornton et al., 2018; Wu et al., 2017; Wu et al., 2016). Although there exists extensive promising data in preclinical models of stroke, no drug nor neuroprotective compound has been elucidated for the treatment of stroke (He et al., 2019).

7.2. Current MB strategies

Studies employing mitochondrial transplantation attenuated stroke-induced neuronal death (Chang et al., 2019). Furthermore, injection of astrocyte-derived mitochondria in an rat model of stroke increased levels of pAKT (Gibbs et al., 2018a), as well as B-cell lymphoma-extra large (BCL-XL), a transmembrane molecule in the mitochondria (Valentin et al., 2018). Enhanced expression of both pAKT and BCL-XL also promoted cell survival in mice (Hayakawa et al., 2016). Finally, injection of neuron-, astrocyte- and microglia-derived mitochondria facilitated enhanced mitochondrial function and locomotor recovery (Huang et al., 2016).

Recent studies revealed that oxidative stress in ischemic neurons may potentially involve PGC-1α and its pathways (Ham 3rd and Raju, 2017). Upon activation of the PGC-1α signaling pathway following global ischemia, expression of UCP2 and SOD2 increased and MB was induced (Chen et al., 2010; Su et al., 2017). Additionally, several studies show a reduction in ROS production with treatment with MB-inducible compounds. In a rat model of middle cerebral artery occlusion (MCAO), melatonin increased mitophagy, reduced ROS and inhibited inflammasome activation, thereby suppressing the inflammatory response (Cao et al., 2017). Daidzein is an MB agent and antioxidant (Rasbach and Schnellmann, 2008). Daidzein suppressed ROS production and mitochondrial swelling, while increasing antioxidant activities (Aras et al., 2015a; Liu et al., 2017; Stout et al., 2013; Thornton et al., 2018; Yuan et al., 2017). Metformin, a known inhibitor of interleukin 1β (IL1β), which is also a modulator of MB (Markowicz-Piasecka et al., 2017), reduced ROS production, hindering mitochondria-mediated apoptosis (Andrzejewski et al., 2014; Owen et al., 2000). Improved behavior assessments, promotion of oligodendrocyte survival and remyelination were also reported following metformin treatment in mice (Qi et al., 2017).

Restoration of impaired mitochondrial dynamics has also been investigated in in vivo stroke models. Decreased mitochondrial fission via downregulation of Drp1 resulted in a diminished degree of infarct size in ischemic stroke (Barsoum et al., 2006; Grohm et al., 2012; He et al., 2019; Zhao et al., 2014). Conversely, enhancing mitochondrial fusion with overexpression of Opa1 was shown to attenuate cerebral edema following cerebral ischemia in rats (Zhang et al., 2014). These data taken together support the concept that restoration of mitochondrial homeostasis and balance may generate neuroprotection against stroke (Thornton et al., 2018). Such effects may be possible with pharmacological activation of MB (He et al., 2019); however, cerebral ischemia is an intricate and complex pathology and effective and meaningful treatment may require a combinational approach.

8. Chronic neurological disease

A hallmark of several chronic neurological diseases is neurodegeneration characterized by an array of disruptions in neuronal systems (Golpich et al., 2017). Neurodegeneration is an umbrella term for the progressive death of nerve cells and loss of brain tissue (Golpich et al., 2017). Several lines of pathological and physiological evidence reveal that impaired mitochondrial function and dynamics play crucial roles in both aging and pathogenesis of neurodegenerative diseases. The transcriptional decline in genes that control MB frequently occurs with advancing age (Beckervordersandforth et al., 2017) and likely serves a key role in the pathogenesis of neurodegenerative disease. As mitochondria are the major intracellular organelles that regulate both cell survival and death, they are considered potential targets for pharmacological-based therapies in chronic neurological diseases (Golpich et al., 2017).

Mitochondrial dysfunction is implicated in multiple neurodegenerative diseases including Alzheimer’s disease (AD) and Parkinson’s disease (PD), Huntington’s disease (HD) and Multiple Sclerosis (MS), among others. This review, however, will focus on AD and PD. For comprehensive reviews detailing MB in neurodegenerative diseases, see Golpich et al. (2017) and Sebastian et al. (2017).

9. Alzheimer’s disease

9.1. Background and pathology

AD is the most common neurodegenerative disorder and is characterized by an accumulation of amyloid plaques (Aβ) and intraneurofibrillary tangles (NFT) made from hyperphosphorylated tau protein (pTau) in the brain (Golpich et al., 2017). Progressive loss of synapses and cholinergic fibers, as well as the proliferation of reactive astrocytes and microglia, also contribute to the pathology of AD (Lin and Beal, 2006). In addition, mitochondrial dysfunction is a hallmark of AD, even in its early stages (Golpich et al., 2017). Altered mitochondrial respiration in postmortem brains of AD patients has been reported in several studies (Chaturvedi and Flint Beal, 2013; Golpich et al., 2015), specifically diminished ATP production paired with increased oxidative stress. There also exist reports of damaged and mutated mtDNA, as well as compromised oxidative phosphorylation and loss of glucose metabolism (Lin and Beal, 2006; McFarland et al., 2010).

Aβ plaques and pTau aggregate in synaptic clefts, highly metabolic regions, thereby hindering synaptic mitochondria and neurotransmission. This hindrance leads to dysfunctional neurons and diminished cognition in AD patients (Lin and Beal, 2006). pTau selectively impairs complex I of the ETC, further hindering mitochondrial respiration (Golpich et al., 2017; Oliver and Reddy, 2019). Aβ aggregates also promote ROS production, further damaging mtDNA, while also propagating additional Aβ formation (Oliver and Reddy, 2019).

An imbalance in mitochondrial dynamics also aids in the pathogenesis of AD, namely upregulation of mitochondrial fission, downregulation of mitochondrial fusion, as well as a reduction in mitophagy (Kerr et al., 2017). This altered balance of mitochondrial morphology leads to considerable neuronal loss, brain volume shrinkage and cognitive decline (Oliver and Reddy, 2019). Evidence suggests that interaction between Aβ and Drp1 is a critical component of mitochondrial dysfunction and impaired dynamics in AD (Reddy et al., 2018). It has been suggested that pTau can interact with and increase expression of Drp1, effectively enhancing mitochondrial fission, compromising mtDNA and damaging synaptic activity in AD neurons(Calkins et al., 2011; Manczak and Reddy, 2012), all leading to cognitive impairments (Reddy et al., 2018; Reiss et al., 2018). The gradual loss of mtDNA may also contribute to increased fission and decreased mitophagic activity seen in AD (Oliver and Reddy, 2019). These reports implicate mitochondrial dysfunction in the pathogenesis of AD, while also revealing a potential target for therapeutic strategies via MB induction.

9.2. Current MB strategies

In mouse models of AD, overexpression of PGC-1α mediated mitochondrial dysfunction (Sheng et al., 2012) and improved cognition (Dumont et al., 2012). Treatment with thiazolidinediones (TZDs), a PPAR activator, was shown to induce MB and improve cognitive capabilities in early mild to moderate cases of AD (Cheng et al., 2016; Heneka et al., 2015; Risner et al., 2006; Watson et al., 2005). Similar protective effects were observed with resveratrol and AICAR (5-aminoimidazole-4-carboxamide ribonucleotide), both activators of the MB signaling molecule, AMP-activated protein kinase (AMPK) (Vingtdeux et al., 2010). Treatment with resveratrol also activates PGC-1α and has been investigated in animal models of PD, HD and ALS, leading to marked decreases in neuronal degeneration (Blanchet et al., 2008; Ho et al., 2010; Kim et al., 2007; Long et al., 2009; Whitaker et al., 2016). Additionally, melatonin increased PGC-1α, Nrf2 and mtDNA, among other mitochondrial genes, while also restoring mitochondrial structure and function in in vitro models of AD (Wang et al., 2019). The generation of Aβ was also mediated with melatonin treatment, all of which suggest a neuroprotective effect via MB induction (Wang et al., 2019). These observations further propose MB induction as a therapeutic avenue for the treatment of mild to moderate cases of AD.

10. Parkinson’s disease

10.1. Background and pathology

Precise etiology has yet to be elucidated for PD, but it is suggested that genetic and environmental factors contribute (Golpich et al., 2017). Cellular mechanisms resulting in nigrostriatal cell death are also unclear (Golpich et al., 2017); however, mitochondrial dysfunction and oxidative stress are known hallmarks of PD (Yan et al., 2013). Like with many diseases characterized by mitochondrial dysfunction, brains from PD patients show increased oxidative stress, chronic inflammation, aberrant protein folding and abnormal protein aggregation (Golpich et al., 2017; Harischandra et al., 2019; Pickrell and Youle, 2015; Truban et al., 2017). Postmortem studies observed decreased activity in complex I, antioxidant coenzyme Q10 and complex IV in the substantia nigra (Hyman et al., 2012; Kilbride et al., 2011). This decreased activity, particularly in complex I, may encourage the accumulation of protein inclusions or lewy bodies (LB) containing alpha-synuclein (α-Syn), both of which are known to be key drivers for the progression of PD (Bengoa-Vergniory et al., 2017; Golpich et al., 2017).

α-Syn was implicated in the maintenance of mitochondrial dynamics in neurons, with aggregation present in mitochondria from the substantia nigra (Devi and Anandatheerthavarada, 2010; Devi et al., 2008; Rani and Mondal, 2019). Vacuolar protein sorting associated protein 35 (VPS35) is another modulator of mitochondrial dynamic activity (Wang et al., 2016a) that encodes α-Syn. In PD, mutated VPS35 depicts increased interaction with Drp1, thereby enhancing mitochondrial fission. This imbalance in mitochondrial dynamics then facilitates mitochondrial fragmentation and neuronal cell death (Wang et al., 2016b). Additionally, PD-associated α-Syn mutations have also been shown to compromise mitophagy. (Rani and Mondal, 2019).

As stated in other sections, mutated and fragmented mtDNA lead to reduced oxidative phosphorylation. Such impairments are known to occur in early stages of PD (Coskun et al., 2012). Compared to other neuronal cells, dopaminergic neurons of the substantia nigra are particularly vulnerable to mtDNA mutations and oxidative damage (Bender et al., 2006; Juarez Olguin et al., 2016). These neurons are charged with creating dopamine (DA). If not properly stored, DA will be metabolized into 3,4-Dihydroxyphenylacetaldehyde (DOPAL), which is highly toxic to dopaminergic neurons (Gandhi et al., 2012). DOPAL can easily modify proteins in such a way as to promote aggregation, resulting in enhanced ROS production and opening of the mPTP, all of which leads to mitochondrial dysfunction and neurodegeneration in PD (Chiu et al., 2015; Coelho-Cerqueira et al., 2019; Kristal et al., 2004; Rani and Mondal, 2019; Zhou and Lim, 2009).

A variety of drugs targeting mitochondrial quality control have been studied in both in vivo and in vitro models of PD. Oral administration of CoQ10 in mice attenuated the loss of DA neurons in PD. However, CoQ10 has yielded conflicting evidence in clinical trials for PD (Raizner, 2019; Rani and Mondal, 2019). Another mitochondrial-targeted antioxidant, MitoQ, has shown positive outcomes in numerous cell and animal models of mitochondrial dysfunction (Ramis et al., 2015), though it did not slow progression of PD in clinical trials (Snow et al., 2010). Finally, P110 is a peptide derived from the interaction site of Drp1 and Fis1. P110 was able to effectively block the localization of Drp1 to mitochondria (Qi et al., 2013), thereby hindering fusion and providing neuroprotection in several rodent models of neurodegeneration, including PD (Disatnik et al., 2013; Disatnik et al., 2016; Guo et al., 2013; Qi et al., 2013; Thornton et al., 2018). Treatment with neuroprotective peptides has also been explored. Small molecule antioxidant peptides SS31 and SS20 showed neuroprotection in PD models by targeting various mitochondrial aberrations and oxidative stress (Yang et al., 2009). Finally, isolated mitochondria were delivered systemically in a mouse model of Parkinson’s disease and found to increase behavioral outcomes (Shi et al., 2017), thereby supporting the theory that restored mitochondrial number and function promote recovery following SCI.

The most potent treatment available for PD consists of a DA precursor levodopa (Tambasco et al., 2018). Though currently serving as the standard treatment for locomotor deficits in PD, several adverse side effects exist, including motor fluctuations and dyskinesia (Virmani et al., 2016). To date, there exist no efficient drugs for repressing the loss of dopaminergic neurons in PD (Rani and Mondal, 2019).

10.2. Current MB strategies

Several studies employing pharmacological-induction of MB reveal beneficial outcomes in animal models of PD. Triterpenoids have been shown to activate Nrf2 pathways in mouse models of PD, thereby promoting expression of genes involved in MB and antioxidative defense (Tufekci et al., 2011). Treatment with benzafibrate, a PPAR agonist and known activator of PGC-1α and inducer MB, has yielded beneficial outcomes in rodents (Corona and Duchen, 2016). Reports in both in vitro and in vivo models of PD showed that PGC-1α overexpression rescued the loss of the DA-ergic neurons and mitochondrial homeostasis, while also enhancing expression of antioxidants (Di Giacomo et al., 2017). Rotenone, a commonly used pesticide, is considered an environmental risk factor for the pathogenesis of non-familial PD. Ongoing exposure to low doses of rotenone results in compromised ETC function and oxidative stress. Following rotenone exposure, treatment with resveratrol improved mitochondrial homeostasis in both cell and animal models, correlating to increased TFAM and PGC-1α expression, indicating improved MB (Peng et al., 2016). Treatment with the mitochondrially biogenic agent LY344864 enhanced expression of mtDNA and PGC-1α, and induced MB in various brain regions in a PD mouse model. LY344864 also attenuated tyrosine hydroxylase immune-reactivity, while improving locomotor activity (Scholpa et al., 2018a). These data indicate that restoring PGC-1α and/or inducing MB may be a promising approach for the development of effective drugs for the treatment of PD (Golpich et al., 2017; Rani and Mondal, 2019).

11. Conclusion

Mitochondrial dysfunction serves as a hallmark in many diseases and injuries of the CNS. Such dysfunction is often characterized by decreased PGC-1α and suppressed MB (Chen et al., 2011; Kim et al., 2010; Li et al., 2018c; Scholpa et al., 2019b). Studies discussed in this review demonstrated that activation of MB restores mitochondrial function and homeostasis and promotes meaningful recovery across multiples CNS related pathologies. As such, inducing MB and attenuating mitochondrial dysfunction may be an effective therapeutic strategy for a multitude of CNS diseases.