The role of exerkines on brain mitochondria: a mini-review

Abstract

Exercise benefits many organ systems, including having a panacea-like effect on the brain. For example, aerobic exercise improves cognition and attention and reduces the risk of brain-related diseases, such as dementia, stress, and depression. Recent advances suggest that endocrine signaling from peripheral systems, such as skeletal muscle, mediates the effects of exercise on the brain. Consequently, it has been proposed that factors secreted by all organs in response to physical exercise should be more broadly termed the “exerkines.” Accumulating findings suggest that exerkines derived from skeletal muscle, liver, and adipose tissues directly impact brain mitochondrial function. Mitochondria play a pivotal role in regulating neuronal energy metabolism, neurotransmission, cell repair, and maintenance in the brain, and therefore exerkines may act via impacting brain mitochondria to improve brain function and disease resistance. Therefore, herein we review studies investigating the impact of muscle-, liver-, and adipose tissue-derived exerkines on brain cognitive and metabolic function via modulating mitochondrial bioenergetics, content, and dynamics under healthy and/or disease conditions.

INTRODUCTION

The discovery in the late 20th century that exercise enhances neurogenesis and neuroplasticity has resulted in a burgeoning area of research investigating the impact of exercise on brain function in health and disease. Given the tremendous impact of exercise on the organ systems of the body, identifying the precise mechanisms by which exercise improves brain function remains challenging. Physical activity reduces cognitive decline associated with Parkinson’s and Alzheimer’s disease, and an individual’s physical activity levels are positively associated with a lower risk of developing several neurocognitive disorders and depression (1). Cellular mechanisms of neurocognitive decline are complicated and multifaceted; however, there is robust literature supporting changes in brain mitochondrial bioenergetics as playing a role in several brain disorder pathophysiologies (2). Herein we investigate the existing literature that physical activity-induced secretory factors (i.e., exerkines) promote brain health by modulating mitochondrial bioenergetics.

Mitochondrial Bioenergetics and Dynamics

Mitochondrial bioenergetics describes the redox systems (e.g., dehydrogenases), electron transport chain (ETC) activities (e.g., oxygen reduction), and mitochondrial membrane potential (ΔΨ_m) regulation that participates in matching ATP resynthesis with energetic demand (for comprehensive review see Ref. 3). In the neuron, for example, mitochondria are spread throughout the dendrites and axons where they produce ATP to support neurotransmission and cell maintenance. Importantly, mitochondria are structurally dynamic organelles that are continuously regulated by biogenesis and dynamics (fusion; mitochondrial elongation and fission; mitochondrial fragmentation) to maintain mitochondrial quality control in response to physiological and/or pathological stimuli (2). In the process of mitochondrial dynamics, dynamin-related protein 1 (Drp1) and mitochondrial fission protein I (Fis1) primarily regulate mitochondrial fission; meanwhile, mitochondrial fusion is mediated by mitofusin 1 and 2 (Mfn1 and Mfn2; outer membrane), and optic atrophy protein 1 (Opa1; inner membrane) (2). Furthermore, mitochondrial autophagy, called mitophagy, is crucial for neuronal health by recycling the impaired mitochondria (4). Since there is a dearth of direct evidence linking myo-, hepato-, and adipose tissue-mediated exerkines to mitophagy in the brain, in this review, we have not addressed the link between exerkines and mitophagy; however, given that exercise has been shown to activate mitophagy and improve mitochondrial function (5), future research is required to explore the direct relationship between exerkines-mediated mitophagy and brain function. Dysfunctional mitochondrial dynamics, content, and bioenergetics are associated with reduced neuronal activity and greater cell death during neurodegeneration (2), and exercise has been shown to improve mitochondrial function, neuronal activity, and neuroprotection. In this review, we discuss the evidence for the beneficial impacts of exerkines modulating mitochondrial bioenergetics, content, and dynamics.

Brain-Derived Neurotrophic Factor

Brain-derived neurotrophic factor (BDNF) plays a dominant role in mediating the beneficial effects of exercise on brain function, especially hippocampal-cognitive function (6). BDNF activates neuronal differentiation from stem cells, augments neurite outgrowth and synaptogenesis, and impedes apoptosis. In neurons, BDNF is expressed throughout development and the adult mammalian nervous system (6). Greater levels of BDNF mRNA and protein are associated with the beneficial effects of exercise in rodents and humans and contribute to better cognitive function (for review see Refs. 6, 7). Although BDNF is not believed to cross the blood-brain barrier (BBB) (8), many of the exerkines reviewed herein act via engaging a BDNF-mediated mechanism in the brain. In addition, BDNF impacts mitochondrial bioenergetics, content, and dynamics in the brain. For example, BDNF enhances the complex I substrate-mediated respiratory control index, which indicates mitochondrial integrity by measuring the coupling of the respiratory chain and improves the efficiency of respiratory coupling (9, 10). BDNF activates peroxisome proliferator-activated receptor γ coactivator 1 α (PGC1α), resulting in an increase in mitochondrial biogenesis, and a corresponding elevation in cellular energy substrates such as ATP and NAD⁺, as well as the maintenance and formation of the synapses (11). Importantly, PGC1α augments BDNF expression levels, supported by evidence that following the knockdown of PGC1α, synapse formation decreases, and the capability of BDNF to stimulate synaptogenesis is impeded (11). These findings suggest a positive feedback loop mechanism whereby BDNF stimulates PGC1α and vice versa. Overall, current findings suggest that BDNF is directly associated with mitochondrial biogenesis and respiratory coupling efficiency. Next, we summarize how the exerkines released from skeletal muscle, liver, and adipose tissues directly impact the parameters of mitochondrial bioenergetics, content, and dynamics in the brain (Table 1).

Table 1.

Role of exerkines on mitochondrial function, content, dynamics, and gene regulation

Exerkine	Impact on Mitochondrial Health
	Function	Content	Dynamics	Gene Regulation
BDNF	↑ complex I – induced respiration (9,10), no Δ complex II – induced respiration (9)	↑ mtDNA (11) ↑ protein content (e.g., cytochrome c) (11)		↑ transcription factors PGC1α, NFR1, TFAM (11)
IL-6	↑ intercellular ATP levels (12)	↑ mtDNA (12), ↑ mitochondrial volume (EM) (12)		↑ transcription factors PGC1α, NFR1, TFAM (12)
β-Hydroxybutyrate	↑ respiration (13–15), ↑ complex II – induced respiration (16), ↑ intracellular ATP levels (14, 17), ↓ ROS production (17,18), ↓ ΔΨ (17), no Δ ΔΨ (16), ↑ NAD:NADH redox (14,15) ↑ complex I-IV activities (18,19), ↑ ROS production (14, 16)	↑ content (via CS activity (19), no Δ mtDNA (14)	↓ Drp1-related fission (17), ↑ Drp1-related fission (15), ↓ Mfn1-related fusion (15)	↑ transcription factors PGC1α (15), no Δ transcription factors PGC1α, NRF1 and 2, TFAM (14)
Lactate	↑ intracellular ATP levels (20)		↓ Drp1-related fission (20) ↑ Mfn1/2-related fusion (20), no Δ Opa1-related fusion (20)	↑ transcription factors PGC1α, NRF2, TFAM (20–22) ↑ mRNA levels of mitochondrial complex genes (20)
Adiponectin	↑ intracellular ATP levels (23,24), ↓ ROS production (23–25), ↓ ΔΨ (24), ↑ ΔΨ (23, 25)		↓ Drp1-related fission (23), ↑ mitochondrial mass (24), ↓ mitochondrial swelling and damaged cristae (25)	↑ mRNA levels of mitochondrial complex genes (23), ↑ transcription factors PGC1α, NRF2, TFAM (24)
Irisin	↓ ROS production (26), ↓ ΔΨ (26), ↑ complex I, II, IV (27), ↑ intracellular ATP levels (27)		↓ mitochondrial swelling and vacuolization (28)	↑ transcription factors PGC1α, NRF2, TFAM (26, 28)
FGF21	↓ ROS production (29), ↓ ΔΨ (29), ↑ respiration (30), ↑ NAD+/NADH ratio (30), ↑ ΔΨ (31)	no Δ mtDNA (30)	no Δ mitochondrial surface area (30)	↑ transcription factors PGC1α (29–31), ↑ mitochondrial respiratory-related genes and proteins (31)

ATP, adenosine triphosphate; BDNF, brain-derived neurotrophic factor; CS, citrate synthase; Drp1, dynamin-related protein 1; EM, electron microscope; IL-6, interleukin-6; Opa1, optic atrophy-1; PGC1α, peroxisome proliferator-activated receptor γ coactivator 1; mfn 1 and 2, mitofusin-1 and 2; NRF1 and 2, nuclear respiratory factor 1 and 2; ROS, reactive oxygen species; TFAM, transcription factor A; ΔΨ, membrane potential.

Exerkines

Exerkines are defined, broadly, as the secretory factors from any organ in response to physical activity (32). The concept of physical activity-induced secretory factors began to emerge in the early 2000s when Steensberg et al. (33) demonstrated that interleukin-6 (IL-6) is secreted from skeletal muscle into the circulation following acute exercise and acts as an endocrine factor. Later, Pedersen and colleagues (34) introduced the term “myokines” referring to the specific autocrine, paracrine, and endocrine effects of muscle-secreted factors. To date, the biological and physiological functions of a minority of myokines (∼5%) have been defined (7). Furthermore, additional organs are now recognized for their physical activity-induced release of secretory factors, such as the liver (hepatokines) and adipose tissue (adipokines). Presently, there is growing recognition that exerkines impact cognitive function (for review, see Ref. 35).

Herein we review exerkines that 1) pass through the BBB and 2) show evidence of influencing mitochondrial bioenergetics, content, and/or dynamics. The revelations from this review highlight a potential mechanism of physical activity-related brain benefits and also recognize a significant knowledge gap in the field, i.e., that there is a critical need for additional brain region-specific studies.

Myokines

The first discovered and most well-studied myokine is interleukin-6 (IL-6). The role of IL-6 in skeletal muscle adaptations to exercise has been demonstrated extensively. Skeletal muscle contraction promotes IL-6 independent of tumor necrosis factor-α (36), suggesting that muscle-mediated IL-6 increases metabolic function rather than inflammatory function. Several studies have demonstrated that peripherally secreted IL-6 crosses the BBB (37) and regulates food intake, possibly suppressing obesity and obesity-mediated neuronal disorders (for review, see Ref. 7). For example, enhancing muscle-derived IL-6 suppresses food intake and reduces body weight in rodent models by regulating the expression of hypothalamic peptides relevant for energy balance, suggesting that IL-6 may contribute to exercise-induced reductions in body weight (38).

More recently, the discovery of Irisin renewed interest in exercise-induced myokine influences on the brain. Irisin is a membrane molecule that is the cleavage product of fibronectin type III domain containing 5 (FNDC5) and is secreted from muscle in circulation and might cross the BBB (39) and potentially mediates the effects of exercise on BDNF expression in the brain. Exercise upregulates the fndc5 expression via increasing the transcription factor, PGC1α in skeletal muscle, and circulating Irisin also upregulates PGC1α- mediated fndc5 expression and irisin in the brain (40). Specifically, in rodents, greater plasma Irisin is associated with elevated Bdnf gene expression in the hippocampus. In primary cortical neurons, in vitro Irisin treatment promotes Bdnf gene expression, whereas RNA-interference-mediated knockdown of fndc5 results in a reduced Bdnf gene expression, suggesting that in both the hippocampus and cortical cells, the BDNF levels are increased by circulating Irisin. In further support of the idea that Irisin may mediate some of the benefits of exercise to prevent neurodegenerative disorders, Irisin has recently been identified to improve synaptic plasticity and memory function in an Alzheimer’s disease mouse model (39). However, one caveat to consider is that it has been controversial whether exercise elevates plasma levels of irisin in humans (41).

Lactate is also one of the major exerkines derived from the skeletal muscle following exercise (42). There is accumulating evidence that exercise-derived lactate passes through BBB and engages BDNF-mediated signaling in the hippocampus and improves learning and memory (21). Furthermore, lactate directly augments metabolic function in brain regions such as the hypothalamus and hippocampus. For instance, hypothalamic sensing of lactate in circulation maintains glucose production, which is critical for glucose homeostasis (43).

Taken together, the evidence suggests that exercise-mediated myokines such as lactate, Irisin, and IL-6 are released into circulation, cross the BBB, and influence brain function.

Myokines and Brain Mitochondria

Irisin (FNDC5), IL-6, and lactate each have shown to impact mitochondrial bioenergetics, content, and dynamics (Table 1). In 2013, Wrann et al. (40) demonstrated that the neuronal gene expression of Fndc5 is controlled by PGC1α, and the knockout (KO) of PGC1α in mice induced a decrease in Fndc5 gene expression in the brain, suggesting that PGC1α regulates Fndc5. Conversely, other studies have demonstrated Irisin controls mitochondrial biogenesis, bioenergetics, and structures (26–28). For example, Irisin treatment increases mitochondrial complex I, II, and IV activities, and ATP levels following chronic unpredictable stress (a rodent model of depression) in rats (27). In vivo intracerebroventricular administration of irisin increases mitochondrial biogenesis following subarachnoid hemorrhage in rodents by increasing PGC1α and the mitochondrial transcription factor A (TFAM) protein expression levels (28). Furthermore, the mitochondrial structure observed by transmission electron microscope (TEM) showed that nascent mitochondria increased, with less swelling in mitochondria and fewer vacuoles. Another study has shown that intravenous treatment of Irisin increases nuclear respiratory factor-2 (NRF-2) protein levels and reduces ΔΨ_m and ROS formation following traumatic brain injury (26). Taken together, Irisin treatment attenuates disease-induced impaired mitochondrial biogenesis, bioenergetics, and structure by enhancing biogenesis factors and structure and lowering ROS and ΔΨ_m.

Similar to Irisin, IL-6 improves mitochondrial content, mitochondrial bioenergetics, and structure in astrocytes under the lipopolysaccharide-induced sepsis model (12). In this study, in vitro treatment of IL-6 to astrocytes enhanced PGC1α, NFR1, and TFAM protein levels, accompanied by an increase in mtDNA content and mitochondrial volume density, increased intracellular ATP levels, and improved mitochondrial morphology observed via TEM. Surprisingly, although IL-6 is the first discovered myokine, there is a dearth of studies on the link between IL-6 and brain mitochondria. Considering that IL-6 treatment has been shown to directly control mitochondrial dynamics in skeletal muscle (44), future research is necessary to investigate the exercise-derived IL-6 would directly influence mitochondrial bioenergetics and quality control in the brain.

Muscle-derived lactate is associated with mitochondrial bioenergetics, content, and dynamics in the brain. For instance, intraperitoneal lactate injections activate the histone deacetylase SIRT1, which engages the hippocampal PGC1α/FNDC5 pathway, resulting in elevated BDNF and TrkB protein levels (21). Similarly, intraperitoneal injection of lactate elevates hippocampal PGC1α and TFAM mRNA expression (22). Blood lactate injection increases hippocampal lactate levels similar to that in blood, suggesting that lactate crosses the BBB, and the effects of lactate are suppressed by UK5099, a lactate transporter inhibitor, demonstrating that blood lactate elevates hippocampal mitochondrial biogenesis markers (22). In addition, Hu et al. (20) suggest lactate has a role in regulating mitochondrial bioenergetics and structure. They treated lactate to primary hippocampal neurons and demonstrated that lactate increases hippocampal ATP levels and enhances OxPhos-related genes, such as Ubiquinol-Cytochrome C Reductase Core Protein 1 (Uqcrc1; complex III) and ATP synthase subunit α 1 (Atp5a1; complex V). Corroborating previous studies, lactate also induced an increase in mitochondrial biogenesis factors, including PGC1α, NRF2, TFAM gene expression levels, and mtDNA copy number. Finally, the authors demonstrated that mitochondrial dynamics is regulated by lactate treatment in primary hippocampal neurons (20). Although there was no change in Opa1 protein levels, Mfn1 and 2 protein levels are significantly higher after lactate treatment; conversely, mitochondrial Drp1 and Fis1 protein levels were reduced by lactate (20), shifting to more fused mitochondria. Together, these findings suggest that lactate elicited by exercise can directly influence brain mitochondria, regulating mitochondrial bioenergetics, biogenesis, and dynamics.

Hepatokines

Fibroblast growth factor 21 (FGF21) is an exercise-mediated hepatokine that crosses the BBB (45) and has neuroregulatory actions in the brain (29, 45). Especially, in the hypothalamus, FGF 21 regulates sugar intake (46) and circadian behavior (47), and in the hippocampus, FGF21 is shown to protect against cognitive decline by not only enhancing synaptic plasticity in the obese rat (29) but also reducing inflammation and oxidative stress in the db/db and aged mouse (45).

Ketone bodies, similar to lactate, are metabolites that double as signaling molecules and act as exerkines with effects in the brain (48). Ketone bodies are produced in the liver during conditions of limited glucose availability, such as fasting and exercise (49). Under limited glucose conditions, adipocytes secrete fatty acids, which are then converted into the ketone bodies β-hydroxybutyrate (BHB) and acetoacetate by the liver. The metabolic demands of exercise have a potent effect on increasing BHB synthesis and secretion from the liver. BHB passes through the BBB and accumulates in the hippocampus, increasing the histone acetylation in the BDNF promoters, particularly activity-dependent promoter 1, and in turn, increased BDNF expression (50). These findings suggest that BHB as an exerkine provides a link between exercise and BDNF expression in the brain.

Hepatokines and Brain Mitochondria

Hepatokines, such as BHB and FGF21, augment brain function by modulating mitochondrial bioenergetics, content, and dynamics (Table 1). For example, FGF21 treatment to the human dopaminergic neurons increases PGC1α protein levels, oxygen consumption rate, and NAD⁺/NADH ratio (which is a marker of ETC efficiency) under normal conditions, although mtDNA copy number and mitochondrial surface area are not changed following FGF21 treatment (30). Furthermore, Sa-Nguanmoo et al. (29) found that following 6 wk of high-fat diet consumption, intraperitoneal FGF21 injection for 6 wk increased PGC1α protein levels and attenuated ROS production, normalized ΔΨ_m, and decreased mitochondrial swelling in rat brain. These results suggest that in vivo injection of FGF21 exerts a neuroprotective effect in obese rats by enhancing mitochondrial bioenergetics, (specifically i.e., amelioration of ROS and ΔΨ_m, and mitochondrial morphology). Another study showed that in vitro and in vivo treatment of FGF21 improved mitochondrial bioenergetics (ΔΨ_m), biogenesis, and content under the MPTP-induced Parkinson’s disease (PD) model supported by the evidence where FGF21 treatment improved ΔΨ_m (in vitro) and the mitochondrial ETC-related gene and proteins expression (in vivo) in the PD model (31).

Similar to FGF21, BHB directly regulates brain mitochondrial bioenergetics, biogenesis, content, and structure. In 2003, Tieu et al. (16) demonstrated that in the 1-Methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)-treated Parkinson’s disease model, BHB increases mitochondrial oxygen consumption via a complex II-dependent manner. Specifically, suppression of complex II abrogates BHB-induced increases in mitochondrial oxygen respiration and abolishes BHB-mediated neuroprotective effects on substantia nigra pars compacta (SNpc) dopaminergic neurons. However, BHB had no effect on mitochondrial ΔΨ_m, and mitochondrial hydrogen peroxide (H₂O₂) production was increased after BHB treatment, suggesting that BHB may not be related to the amelioration of ROS and ΔΨ_m, perhaps, BHB would increase physiological levels of ROS, and/or cellular antioxidants may increase following BHB treatment. Using high levels of glutamate exposure to mimic metabolically injured neurons, BHB treatment of primary cultured rat cortical neurons increases mitochondrial respiratory capacity, suggesting that neuroprotection by exogenous BHB may partially be regulated by the preservation of mitochondrial respiratory capacity (13). In line with these studies, Marosi et al. (14) found that neurons treated with BHB have an increase in mitochondrial respiration rate, ATP production, and NAD⁺/NADH ratio, although mitochondrial biogenesis markers (e.g., PGC1α, NFR1 and 2, TFAM) and mtDNA copy number were not changed following BHB treatment; in addition, BHB increased mitochondrial ROS production, which is consistent with the previous study (16).

The increases in ROS production were shown to underlie BHB-mediated Bdnf expression, supported by the evidence that reduced mitochondrial ROS production impeded BHB-mediated Bdnf promoter activity (14). In this study, paradoxically, antioxidant enzymes (e.g., superoxide dismutase 2 and heme oxygenase 1) were concurrently increased in neurons treated with BHB, suggesting that although mitochondrial ROS production was increased to promote Bdnf expression, simultaneously, enhanced antioxidant defenses after BHB treatment defends against the increased mitochondrial ROS production. Furthermore, in the oxygen-glucose deprivation/reoxygenation (OGD/R) treatment to SH-SY-5Y cells to mimic the ischemic injury model, BHB treatment confers against OGD/R-mediated reductions in ATP generation, increased ROS, and decreased ΔΨm (17). Furthermore, OGD/R-induced Drp1 mitochondrial recruitment promotes mitochondrial fission; however, mitochondrial Drp1 translocation is suppressed by BHB treatment, suggesting that BHB reduces mitochondrial fragmentation.

More recently, Dabke et al. (19) showed that BHB treatment to HT22 hippocampal murine neurons augments mitochondrial complexes (i.e., complex I + III) and citrate synthase enzyme activities. In the same cell line, BHB treatment increased mitochondrial complex I and IV enzyme activities (18). The findings suggest that BHB increases mitochondrial respiratory enzyme activity. Furthermore, the authors found that hippocampal neurons treated with BHB exhibited lower ROS production and improved antioxidants, such as glutathione and catalase, which is inconsistent with previous studies (14, 16) showing an increase in ROS production following BHB treatment. These divergent findings may be attributed to timing, dose-dependent manner, and experimental type (in vitro vs. in vivo). Finally, using both an in vivo model using a mutated uracil-DNA glycosylase 1 that induces mitochondrial toxicity and an in vitro model where cultured hippocampal neurons and human fibroblasts were treated with H₂O₂ to cause oxidative stress, BHB treatment improves mitochondrial biogenesis, structure, and function markers by upregulating PGC1α protein and gene expressions (in vivo) and increasing oxygen consumption rate and NAD⁺/NADH ratio (in vitro) (15). These findings suggest that BHB improves mitochondrial bioenergetics, biogenesis, and dynamics under normal and disease conditions.

Taken together, mitochondrial biogenesis and respiration markers are consistently improved by the hepatokines (BHB and FGF21). However, although FGF21 ameliorated ROS production, ROS production following BHB treatment is more nuanced, with conflicting results. Future studies into these opposing findings are needed to establish the impact of hepatokines on ROS production in the brain and the relationship with mitochondrial function.

Adipokines

Adiponectin (ADN) is a protein that is secreted into the circulation from adipocytes and acts as a signaling molecule (adipokine). ADN has antidiabetic, insulin-sensitizing, antiatherogenic, anti-inflammatory, and neuroprotective effects (51) and is increased following exercise (52).

ADN can pass through the BBB and improve cell proliferation and reduce depression-like behavior (51). Specifically, exercise-mediated mitigation of depression-like behavior was offset in ADN-deficient mice via an impaired AMP-activated protein kinase signaling in the hippocampus. Interestingly, running increased BDNF protein levels in the isolated dentate gyrus of both wild-type (WT) and ADN KO mice, suggesting that ADN does not regulate BDNF in the dentate gyrus (51). Even if the role of ADN indicates a direct interplay between adipose and brain tissues after exercise, the underlying mechanisms of how exercise-induced adipokines influence brain function need to be elucidated.

Adipokines and Brain Mitochondria

Similar to myokines and hepatokines, ADN augments brain mitochondrial bioenergetics, content, and dynamics in several brain disorders (Table 1). For example, in 2018, a study found that ADN protects HT22 hippocampal neuronal cells from oxygen-glucose deprivation (OGD)-induced mitochondrial structural dysfunction and oxidative stress (25). In detail, using the OGD-induced neuronal injured in vitro model, ADN treatment ameliorates mitochondrial structural damages, including vanished cristae and mitochondrial swelling (25). Moreover, ADN treatment attenuated higher ROS and decreased ΔΨ_m in OGD-treated cells (25). More recently, Wu et al. (23) demonstrated that in transcriptomic analysis, the expression of mitochondrial respiratory-related genes, such as mt-Nd1, Nd4, mt-Atp8, and mt-cytb, are increased after ADN peptide (ADNp) treatment in an intracerebral hemorrhage (ICH)-injured rodent model. In line with these findings, in vivo ADNp intraperitoneal injection increases ATP levels following the ICH injury (23), suggesting adiponectin improved mitochondrial respiratory factors. Using an in vitro ICH model, mitochondrial ROS and ΔΨ_m are normalized following ADNp incubation in the primary astrocytes (23). In addition, ADNp increased Drp1 (serine 637) phosphorylation, which suppresses the translocation of Drp1 to the mitochondria, and decreases mitochondrial fragmentation in the astrocytes, suggesting that ADNp regulates mitochondrial dynamics by suppressing mitochondrial fragmentation. In an in vivo model of ICH-induced injury, Yu et al. (24) activated ADN receptor 1 by intraperitoneally injecting AdipoRon, the agonist of the ADN receptor and found that activation of the ADN receptor increased ATP levels and reduced ROS production in the brain, and further elevated mitochondrial biogenesis factors by enhancing the protein levels of PGC1α, NFR1, and TFAM. Furthermore, in the ICH model, activation of the ADN receptor increased overall mitochondrial mass and attenuated the hyperpolarized mitochondrial ΔΨ_m. These findings suggest that ADN improves mitochondrial bioenergetics, content, and dynamics in brain tissues under brain disease models.

CONCLUSIONS

In this mini-review, we describe how exerkines influence brain health by directly modulating mitochondrial bioenergetics, content, and dynamics (Fig. 1). Since the initial findings that exercise increases BDNF levels in the brain, promoting neural plasticity and neurogenesis and improving cognitive function, researchers have been searching to identify factors mediating these beneficial actions of exercise on the brain. In the past two decades, several exerkines derived from muscle, liver, and adipose tissues have been shown to directly affect brain mitochondria, thereby showing the prevention or treatment of brain disorders. Apart from the exerkines covered in this mini-review, other peripherally derived exerkines have been shown to improve brain function (for review, see Refs. 7, 35, 53). However, whether these exerkines directly affect brain mitochondria is currently unknown. Thus, further studies are needed to shed light on the effects of these exerkines on brain health and/or disease via modulating mitochondrial bioenergetics, content, and dynamics under various disease conditions.

Figure 1.

Effects of exerkines on brain mitochondrial bioenergetics, contents, dynamics, and transcriptional factors. Exercise-mediated secretion of peripheral factors (exerkines) derived from skeletal muscle (FNDC5/irisin, IL-6, and lactate), adipose tissue (adiponectin), and liver (BHB and FGF21) are elevated in the bloodstream. These exerkines pass through the BBB and directly promote aspects of mitochondrial bioenergetics, content, dynamics, and transcriptional factors or indirectly enhance these parameters via elevating BDNF expression in the brain. Improved mitochondrial function results in increased neurogenesis, neuroplasticity, and cognitive function. BBB, blood-brain barrier; BDNF, brain-derived neurotrophic factor; BHB, β-hydroxybutyrate; FGF21, fibroblast growth factor 21; FNDC5, fibronectin type III domain containing 5; IL-6, interleukin-6.

Recently, Schmidt et al. (3) proposed that to better appreciate mitochondrial bioenergetics and draw the conclusions that are physiologically relevant in the mitochondrial environment, recent advances in comprehensive mitochondrial bioenergetics analyses (e.g., creatine-kinase clamp test, membrane potential, NADH/NAD⁺ redox states) would be a proper approach. Current bioenergetics studies have been limited because they rely on the supraphysiological environment in the mitochondria and fail to recapitulate the free energies balance (e.g., ΔG_redox, ΔG_ΔΨ, and ΔG_ATP) and demand-driven environment. Therefore, future research exploring the link between exerkines and brain mitochondria needs to utilize comprehensive mitochondrial bioenergetics testing to better understand the mechanisms of how exerkines derived from peripheral tissues directly influence mitochondrial bioenergetics in a dynamic range of physiologically relevant conditions in the brain.

GRANTS

We acknowledge support by the Department of Defense, through the Clinical & Rehabilitative Medicine Research Program: W81XWH-20-1-0885 (to J.A.C.).

DISCLAIMERS

Opinions, interpretations, conclusions, and recommendations are those of the authors and are not necessarily endorsed by the Department of Defense and/or the National Institutes of Health.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

J.H., E.E.N., and J.A.C. conceived and designed research; J.H. prepared figures; J.H. drafted manuscript; J.H., E.E.N., and J.A.C. edited and revised manuscript; J.H., E.E.N., and J.A.C. approved final version of manuscript.