Neuroglia and brain energy metabolism

Abstract

The glial control of energy homeostasis is of crucial importance for health and disease. Astrocytes in particular play a major role in controlling the equilibrium between ATP, ADP, AMP, and adenosine. Any energy crisis leads to a drop in ATP and the resulting increase in adenosine is an evolutionary ancient mechanism to suppress energy-consuming activities. The maintenance of brain energy homeostasis in turn requires the availability of energy sources such as glucose and ketones. Astrocytes have assumed an important role in enabling efficient energy utilization by neurons. In addition, neurons are under the metabolic control of astrocytes through regulation of glutamate and GABA levels. The intricate interplay between glial brain energy metabolism and brain function can be best understood once the homeostatic system of energy metabolism is brought out of control. This has best been studied within the context of epilepsy where metabolic treatments provide unprecedented opportunities for thee control of seizures that are refractory to conventional antiseizure medications. This chapter will discuss astroglial energy metabolism in the healthy brain and will use epilepsy as a model condition in which glial brain energy homeostasis is disrupted. We will conclude with an outlook on how those principles can be applied to other conditions such as Alzheimer’s disease.

Introduction

Although the human brain comprises only 2–3% of the total body weight it requires about 25% of glucose and 20% of oxygen relative to the rest of the body and is thus the most energy dependent organ of the human body (Sokoloff, 1977). Neurons in particular have a high demand of energy demand because of costly bioenergetic functions related to action potential firing, synaptic activity, and plasticity changes in addition to their role of maintaining cytoskeletal dynamics and axoplasmic transport (Belanger et al., 2011, Hall, 2012, Engl and Attwell, 2015). In contrast to its high energy demand, the brain is not able to store significant amounts of energy, like muscle tissue for example, and therefore is dependent on a variety of external energy sources to maintain homeostasis. This requires the transport of energy substrates through the blood-brain-barrier (Profaci et al., 2020, Loscher and Friedman, 2020). Although astrocytes can store glycogen as a reserve for glycolytic ATP production (Dienel, 2019), this is not sufficient to meet the immediate energy needs of the brain; thus, astrocytic glycogen storage is considered to be only an emergency energy reserve used to support signaling functions between neurons and glia (Dienel, 2019). While glucose is the obligate energy source for the brain (Dienel, 2012), metabolic flexibility allows the brain to utilize alternate energy sources such as lactate, ketone bodies, and medium-chain fatty acids (Owen et al., 1967, Hasselbalch et al., 1994, Bordone et al., 2019).

Role of glia for brain energy metabolism

Astrocytes can produce ATP directly from internal energy stores through two main mechanisms: anaerobic glycolysis and oxidative metabolism through the Krebs cycle (Boison and Steinhauser, 2018). However, neurons depend largely on glucose supply through the vasculature, which requires glucose transport systems such as GLUT3 (Maher et al., 1994, Dienel, 2012). Oxidative metabolism of exogenous glucose can supply much of the ATP need of neurons, however neurons also depend on metabolic fuel from astrocytes to maintain viability and function. This requires a sophisticated crosstalk of neurometabolic coupling between astrocytes and neurons, whereby astrocytes can supply neurons with lactate as energy source according to the astrocyte-neuron lactate shuttle (ANLS) hypothesis (Pellerin and Magistretti, 1994, Magistretti and Allaman, 2018, Machler et al., 2016). The neurovascular unit emerges as the centerpiece of brain energy homeostasis and involves the interdependent relationships between neurons, astrocytes, and microcirculation. Importantly, in all cells of the neurovascular unit, mitochondria are the key players in the control of bioenergetic capacity (Lin et al., 2010, Hall, 2012). Mitochondrial dysfunction is a key factor that influences intracellular calcium, oxidative stress, and apoptotic cell death.

In line with the central role of astrocytes as key regulators of brain energy metabolism it is now well established that astrocytes play a major role in sleep regulation (Ingiosi and Frank, 2023). Specifically, astrocytes release somnogenic substances such as immune factors and cytokines, neurotrophins, prostaglandins, and purines (Tobler et al., 1984, Krueger, 2008, Hayaishi, 2002, Kushikata et al., 1999, Huang et al., 2007, Huang et al., 2011, Faraguna et al., 2008, Blutstein and Haydon, 2012). Specifically, astrocytes play a major role in the release of ATP as a precursor for adenosine (Pascual et al., 2005), which has broad implications for sleep regulation (Lazarus et al., 2017, Porkka-Heiskanen and Kalinchuk, 2011, Bjorness and Greene, 2009). Adenosine promotes sleep, whereas adenosine receptor antagonists such as caffeine promote wakefulness. A contribution of adenosine metabolism to sleep regulation was shown through the transgenic overexpression of adenosine kinase (ADK). Adk-tg mice with increased ADK expression in brain were awake about 1 hour more every day as compared to wild type mice and they spent significantly less time in REM sleep (Palchykova et al., 2010). These findings suggest that astrocyte metabolism controls the important physiological phenomenon of sleep.

Epilepsy, a disorder characterized by spontaneous unprovoked seizures, is a model disease which illustrates the consequences of the disruption of glial energy homeostasis. It is now well established that astrocytes play a key role in the balance between seizure prevention and seizure genesis, via the modulation of synaptic neurotransmission and control of the ion balance between intracellular and extracellular compartments (Boison and Steinhauser, 2018). In the healthy brain astrocytes prevent neuronal hyperexcitability and seizures through K⁺ buffering and the regulation glutamate uptake through the astrocytic glutamate transporter GLT-1 (Boison and Steinhauser, 2018). During seizure activity astrocytes can restore the disrupted homeostasis of ions and neurotransmitters, through the re-uptake of glutamate from the synaptic cleft and through buffering of extracellular K⁺ through glial end-feet which connect to the brain microvasculature (Boison and Steinhauser, 2018, Profaci et al., 2020). In an in vitro stem cell model the re-uptake of glutamate into astrocytes triggered enhanced glycolysis as a source of enhanced pyruvate formation, which in turn was converted to lactate as a substrate for the ANLS and source of neuronal energy supply (Tarczyluk et al., 2013). In addition, astrocytes are a part of large networks of highly inter-connected cells, which enable the bi-directional exchange of metabolites, ions, second messengers, amino acids, peptides, nucleotides, and RNA through gap junction channels (Boison and Steinhauser, 2018). Thus, astrocytes emerge as important network regulators of electrical and metabolic equilibrium. Because astrocytes uniquely can use glycogen as a fuel, they play a crucial role in maintaining energy metabolism, but also permit excessive neuronal firing as seen during epileptic seizures. Growing lines of evidence implicate astrocyte dysfunction in both seizure generation (ictogenesis) and epilepsy development (epileptogenesis) (Figure 1) (Devinsky et al., 2013, Robel and Sontheimer, 2016, Boison and Steinhauser, 2018, Verhoog et al., 2020).

Figure 1: Glial mechanisms of epileptogenesis.

Epileptogenesis is the process that turns a healthy brain into an epileptic brain. Acquired epilepsy can be due to a variety of injurious triggers, such as traumatic brain injury. Epileptogenesis is an ongoing process that extends beyond a latent, seizure-free, period and cab best be explained by a complex derangement of network changes depicted by round circles connected with arrows. Targeting of several critical nodes at the right time provides a rationale for therapeutic strategies aimed at preventing the development and/or progression oof epilepsy. Several epileptogenic mechanisms can be linked to glia. An acute injury to the brain triggers an immediate surge of adenosine (ADO), which promotes glial activation and inflammatory processes, which are key for the epileptogenic process. The acute ADO surge can also reduce the expression of the astroglial glutamate transporter GLT-1, which results in the elevation of glutamate (Glu). Glutamine synthetase is also reduced, which contributes to a rise in glutamate and a decrease in GABA. Finally, astrocyte activation leads to maladaptive overexpression of adenosine kinase (ADK), which drives maladaptive DNA methylation and leads to a deficiency in the brain’s endogenous anticonvulsant ADO.

ATP and adenosine metabolism

Adenosine has early evolutionary roles as a master regulator of energy homeostasis (Boison and Yegutkin, 2019). Any energy challenge broadly leads to a drop in ATP and a resulting increase in adenosine, which acts as a retaliatory metabolite and global inhibitor of metabolic and physiological activity (Newby, 1984). From an energetic standpoint, an epileptic seizure is a state of excessive energy consumption, which prompts the generation of adenosine acting as endogenous anticonvulsant and seizure terminator (During and Spencer, 1992). Adenosine-based inhibitory activity in the brain is largely mediated by activation of G_i protein coupled adenosine A₁ receptors, which can be blocked by the non-selective adenosine receptor antagonists caffeine and theophylline which therefore can have proconvulsant properties (Boison, 2011). Importantly, in the adult brain adenosine metabolism is largely controlled by astrocytes through a variety of mechanisms (Boison et al., 2010). The major adenosine metabolizing enzyme adenosine kinase (ADK), a ribokinase that transforms adenosine into AMP is largely expressed in astrocytes (Studer et al., 2006); due to the existence of an ubiquitous transmembrane transport system for adenosine (equilibrative nucleoside transporter 1) (Baldwin et al., 2004) the intracellular expression of ADK determines extracellular levels of adenosine and turns astrocytes into a metabolic sink for adenosine (Studer et al., 2006, Boison and Yegutkin, 2019, Boison, 2013, Yegutkin and Boison, 2022). Thus, the transgenic or viral overexpression of ADK increases seizure susceptibility and neuronal vulnerability (Pignataro et al., 2007, Shen et al., 2014), whereas a genetic knockdown of ADK or its pharmacological inhibition suppressed seizures and provided neuroprotection (Fedele et al., 2005, Shen et al., 2011, Theofilas et al., 2011).

Maladaptive overexpression of astrocytic ADK has emerged as a pathological hallmark of experimental and human TLE (Aronica et al., 2013, Aronica et al., 2011). In multiple epilepsy models it was shown that reactive astrogliosis was associated with an increased expression of ADK. The resulting adenosine deficiency was not only sufficient to trigger seizures (Li et al., 2008) but also contributed to the epileptogenic process (Williams-Karnesky et al., 2013). Therefore, the therapeutic augmentation of adenosine is therapeutically effective, not only for seizure suppression (Gouder et al., 2003), but also to interfere with the epileptogenic process that leads to the development of epilepsy (Williams-Karnesky et al., 2013). Therefore, a derangement in astrocytic adenosine metabolism affects the pathophysiology of epilepsy broadly and offers new metabolism-based therapeutic options, which directly interfere with the epileptogenic process (Boison and Rho, 2020).

Glutamate, glutamine, and GABA metabolism

An additional astrocyte based metabolic enzyme is glutamine synthetase (GS) which aminates glutamate into glutamine. After transport into inhibitory or excitatory neurons, glutamine can be converted into GABA or re-converted into glutamate, respectively. Therefore, astroglial GS plays an important role in the regulation of the glutamate and GABA balance, which is under metabolic control because GABAergic neurons are more sensitive to an hypoenergetic state (Eid et al., 2013, Sandhu et al., 2021). In line with this mechanism, loss of GS activity in human temporal lobe epilepsy can cause seizures and supports epileptogenesis by promoting (i) an increase in extracellular glutamate, regulated by astrocytes, and (ii) a decrease in GABA production in neurons because of a decrease in glutamine supply (Eid et al., 2013). Specifically, neuroinflammation through oxidative stress inhibits GS activity suggesting that these processes may affect GS in epilepsy (Vezzani et al., 2019). Dysregulation of extracellular glutamate can further be exacerbated by impaired astrocytic glutamate clearance through dysregulation of GLT-1, which is downregulated in the epileptic brain (Bedner et al., 2015, Seifert et al., 2006, Peterson and Binder, 2020). Indeed, the lack of GLT-1 in mice causes epilepsy and increased brain injury (Tanaka et al., 1997). Together, the consequences of this metabolic derangement contribute to both ictogenesis and epileptogenesis (Figure 1).

Role of astrocyte-neuron lactate shuttle

Lactate cannot passively diffuse across the blood brain barrier (Profaci et al., 2020), and since it cannot be directly utilized for energy production, it must be first transported via monocarboxylic transporters (MCTs) into cells (Felmlee et al., 2020), and then be subsequently converted enzymatically to pyruvate by lactate dehydrogenase (LDH) which exists in multiple isoforms – LDH1 being primarily expressed in neurons and LDH5 in astrocytes. However, the ANLS hypothesis is not without some controversy, as there is evidence that oxidative metabolism of lactate in neurons may not always be important for synaptic neurotransmission (Bak et al., 2009, Dienel, 2012, Diaz-Garcia and Yellen, 2019) and the directionalities of lactate transport under different physiological conditions remain unclear. Moreover, when abnormally increased neuronal activity occurs during epileptic seizures, neurons may rely more on their own aerobic glycolysis than astrocyte-derived lactate (Diaz-Garcia and Yellen, 2019). Epileptic neuronal networks have a high demand for energy. Excessive synaptic activity stimulates astrocytic glycolysis causing a rapid drop of glucose and a corresponding rise in lactate, which becomes an essential energy source for neurons. Increased ictal glycogen and glucose utilization leads to the formation of lactate from pyruvate through lactate dehydrogenase (LDH). Through this mechanism, astrocytes can potentially provide a significant supply of energy via the astrocyte-neuron lactate shuttle to sustain the energy demands of hyperactive neuronal networks (Pellerin and Magistretti, 1994). The hypothesis that seizure-induced lactate formation enables epileptic activity is supported by findings that LDH inhibitors such as stiripentol provide robust anti-seizure effects (Sada et al., 2015). In this scenario, LDH inhibition likely interferes with glycolysis by limiting the availability of NAD⁺ and supporting the oxidative mitochondrial metabolism of pyruvate. However, a recent study demonstrated that both astrocytic glycolysis and the astrocyte-neuron lactate shuttle depend on the energy sensor astroglial AMP activated protein kinase (AMPK) (Muraleedharan et al., 2020). The genetic deletion of AMPK in mice led to the depletion of lactate and a reduction in seizure threshold, whereas the deletion of AMPK in astrocytes, but not in neurons, triggered neuronal cell loss in mice and flies (Coras et al., 2010). This apparent contradiction demonstrates the complexity of astrocyte metabolism in the regulation of lactate and brain energy homeostasis, which are of crucial importance for the balance of seizure thresholds.

Seizure-induced impairment of metabolism

Ictogenesis, which is the process that triggers a non-provoked seizure involves complex multi-directional molecular interactions, which include the dysfunction of crucial structural, metabolic and biochemical homeostatic control mechanisms (Rakhade and Jensen, 2009, Pitkanen et al., 2015, Becker, 2018). Glycolytic flux is upregulated immediately after the onset of active seizure activity as the brain uses up glucose in response to the increased energy demand (Ingvar and Siesjo, 1983). Soon, an equilibrium is reached in which the increased ictal energy consumption is off-set by inter-ictal hypo-metabolism (Theodore, 1999, Shultz et al., 2014). Even though the final energy substrates of glycolysis are minimal compared to the oxidative phosphorylation pathway, a phenomenon similar to the Warbug effect in cancer cells exists, in which an up to 30-fold increase in the glycolytic rate under aerobic conditions (aerobic glycolysis) produces enough ATP to sustain seizure activity (During et al., 1994, Boison and Yegutkin, 2019). Increased glucose and glycogen usage through the glycolytic pathway ultimately leads to the increased formation of pyruvate, which is converted to lactate, by lactate dehydrogenase (LDH). When key enzymes involved in oxidative phosphorylation, e.g. pyruvate dehydrogenase or aconitase, which are components of the Kreb’s cycle, are compromised during seizure activity such as during status epilepticus, the equilibrium is driven in favor of the glycolytic pathway with subsequent increased activity of LDH and decreased flux through the Krebs cycle (McDonald and Borges, 2017, Bhandary and Aguan, 2015, McDonald et al., 2017, Liang et al., 2000, Bainbridge et al., 2017). In the same vein, important protein multimers that comprise the electron transport chain are also compromised in temporal lobe epilepsy (TLE) and rodent models of epilepsy (Kunz et al., 1999, Folbergrova et al., 2010, Ryan et al., 2012). Thus, mitochondrial dysfunction, which is associated with widespread impairment in metabolism, as has been found in models of status epilepticus as well as individuals with mitochondrial disorders associated with epilepsy (Rahman, 2012, Rowley and Patel, 2013).

Mitochondrial dysfunction due to seizures is further exacerbated by increased levels of reactive oxygen species (ROS), as oxygen is oxidized to super-oxide and electron flux is altered by the increased activity of enzymes such as NADPH oxidase 2 and inactivation of mitochondrial superoxide dismutase (Rho and Boison, 2022). Decreased activity of sirtuin 3 and impairments in peroxide detoxification have also been implicated (Rho and Boison, 2022). Ultimately, there is increased production of free radicals which further disrupts metabolic equilibrium during extended seizure activity, propagates oxidate stress, and culminates in free radical damage to cellular macromolecules (Patel, 2004). Expectedly, glutathione, a primary endogenous antioxidant is oxidized to glutathione disulfide and depleted in the process (Smeland et al., 2013, Patsoukis et al., 2005, Mueller et al., 2001, Liang and Patel, 2006). Importantly, calcium flux is also altered during these events which leads to a positive calcium balance promoting the generation of more ROS (Rho and Boison, 2022). ROS in turn propagate the cycle of metabolic impairment, activation of necrotic and apoptotic pathways, and finally impair ATP production (Giorgi et al., 2012, Pathak and Trebak, 2018).

Metabolic and epigenetic changes in epilepsy

Since ictal activity involves a major expenditure of energy, it is important to explore the homeostatic response as attempts are made to restore equilibrium between excessive consumption or depletion of energy. Considering that increased ATP usage is associated with an epileptic seizure, the event is also associated with increased generation of adenosine, the breakdown product of ATP. From a metabolic viewpoint, a rheostat is needed to conserve energy in case of excessive energy consumption or depletion of energy supplies (Newby, 1984). From this perspective, an epileptic seizure represents excessive energy consumption entailing the degradation of ATP into adenosine, with adenosine being used as an innate feedback rheostat to conserve energy. Therefore, seizure-induced adenosine production (During and Spencer, 1992) acts as the brain’s endogenous seizure terminator (Dragunow et al., 1985, Lado and Moshe, 2008). Because adenosine is also part of RNA and a regulator of S-adenosylmethionine-dependent methylation reactions (Boison et al., 2002, Williams-Karnesky et al., 2013), any energy crisis is also expected to affect RNA synthesis by lowering ATP, and DNA methylation by increasing adenosine. Both processes would combine to conserve energy through global reduction of gene transcription.

It is evident that metabolic, biochemical, and epigenetic alterations play a major role in the pathophysiology of acquired epilepsies (Klein et al., 2018). Among those alterations, maladaptive changes in adenosine metabolism constitute a central mechanism, linking metabolism with neuronal excitability and gene expression. Alterations in this metabolic balance play pivotal roles through increased neuronal excitability and gene expression (Fredholm et al., 2005a, Fredholm et al., 2005b). Specifically, adenosine’s effects on its 4 G protein-coupled receptors (A1, A2A, A2B, and A3) which control neuronal excitability, gene expression, and inflammation could potentially be harnessed as a therapeutic adjunct in the treatment of epilepsy. Adenosine is under metabolic control of adenosine kinase (ADK), which exists in a cytoplasmic form ADK-S (regulation of tissue levels of adenosine) and a nuclear form (ADK-L), which controls adenosine metabolism in the cell nucleus (Boison, 2013). ADK-L is a cell cycle enzyme, which drives the flux of methyl groups through the transmethylation pathway, which is needed during the S-phase of the cell cycle. Thus increased ADK-L drives global hypermethylation of DNA. Therapeutic increases of adenosine lead to a global reduction in 5-methylcytosine (5mC), an effect independent of adenosine receptors. In line with this mechanism, cultured cell lines engineered to overexpress ADK-L have increased global 5mC levels as compared to ADK-S cells or ADK-deficient cells (Williams-Karnesky et al., 2013).

Metabolic therapies

As outlined above, astrocytes play a key role as metabolic master regulator of brain activity. Because maladaptive changes in metabolism and basic biochemistry are linked to the generation of seizures and the development off epilepsy, astrocytes are attractive targets for metabolic therapies (Boison and Steinhauser, 2018). The most widely used metabolic therapies are the high-fat, low-carbohydrate ketogenic diet, its variants (e.g., the medium-chain-triglyceride diet, or the Atkins diet), and the low glycemic index treatment (LGIT) (Diaz-Garcia and Yellen, 2019, Freeman et al., 2006, Huttenlocher, 1976, Kossoff et al., 2006). Those metabolic interventions force the brain to use ketone bodies rather than glucose as the main energy source. Ketogenic diets and related therapies enhance ketone body production in the liver, but astrocytes constitute a major site for fatty acid oxidation in the brain and they are the only cell types in the brain capable of producing ketone bodies. The diet-induced increase in fatty acid oxidation leads to a subsequent increase in blood ketone substrates such as beta-hydroxybutyrate and acetoacetate (Hasselbalch et al., 1994, Owen et al., 1967). Variants of the diet such as the Atkins diet are being developed to reduce side effects and to increase palatability (Kossoff et al., 2006). Altogether, metabolic therapies including the ketogenic diet have been associated with greater than 50% seizure reduction in up to 60% of individuals with treatment-resistant epilepsy (Neal et al., 2008, Kim et al., 2019, Lyons et al., 2020, Martin-McGill et al., 2020). The mechanism of action is multi-factorial and includes the following: modulation of the aerobic oxidative phosphorylation pathway to restore energy balance and correct mitochondrial dysfunction (DeVivo et al., 1978, Bough et al., 2006, Kim et al., 2010, Patel, 2004, Liang and Patel, 2006, Jarrett et al., 2008, Milder et al., 2010); GABAergic / glutamatergic pre-synaptic and post-synaptic regulation of excitatory and inhibitory neurotransmitters through receptors, K_ATP channels, metabolic substrates and enzymes (Juge et al., 2010, Kawamura et al., 2010, Sada et al., 2015, Ma et al., 2007); reduction in glycolytic flux and regulation of glycolysis apoptotic factors, e.g., via the glycolytic inhibitor 2-deoxy-D-glucose and fructose −1,6-bisphosphate, which augments the pentose phosphate pathway (Shao et al., 2018, Gimenez-Cassina et al., 2012, Lian et al., 2007); enhancement of the mTOR pathway (McDaniel et al., 2011, Warren et al., 2020); bi-directional anti-inflammatory effects on neurons and blood-brain barrier permeability (Shimazu et al., 2013, Youm et al., 2015, Koh et al., 2021, Loscher, 2020); augmentation of adenosine (Masino et al., 2011); and disease-modifying effects through epigenetic changes on DNA and histones (Kobow et al., 2013, Shimazu et al., 2013, Lusardi et al., 2015). Elucidating the multiple and synergistic mechanisms of metabolic therapies is currently of active research interest. The mechanism of action of fatty acids in the therapeutic efficacy of the ketogenic diet involves the direct inhibition of voltage-gated sodium and calcium channels, as well as activation of proliferator-activated receptor-α (PPARα), a gene which play a pivotal role in the regulation of β-oxidation of fatty acids (Bordoni et al., 2006, Elinder and Liin, 2017). Fatty acids also have an anti-inflammatory and antioxidant effected through increased expression of PPARγ. Finally, the LGIT diet is a glucose-restrictive diet which imparts relative hypoglycemia (Muzykewicz et al., 2009, Pfeifer and Thiele, 2005). A comparison of the clinical effectiveness of these metabolic therapies suggests that the traditional ketogenic diet and the medium-chain triglyceride diet are comparable (Neal et al., 2008), while the results are still equivocal for the modified Atkins diet and LGIT due to the absence of high quality comparison studies (Martin-McGill et al., 2020, Sondhi and Gulati, 2021, Sourbron et al., 2020).

Therapeutic manipulation of astrocytic adenosine metabolism

In the adult brain astrocytes are the major cell type regulating the basal tissue tone of adenosine, which acts as the brain’s endogenous seizure terminator and anticonvulsant (Studer et al., 2006, Kiese et al., 2016, Lado and Moshe, 2008, During and Spencer, 1992, Dragunow, 1991). Astrocytes regulate extracellular adenosine through an equilibrative transport system for adenosine that depends on intracellular ADK a ribokinase that removes adenosine by adding a phosphate group, transforming adenosine into AMP (Boison, 2013). Through the activity of ADK, astrocytes become a metabolic sink for adenosine (Boison et al., 2010). Astrocyte activation in epilepsy leads to maladaptive increases in ADK, resulting in brain wide adenosine deficiency, which can be a direct cause for the generation of spontaneous recurrent seizures (Gouder et al., 2004, Li et al., 2007a, Li et al., 2008). Therefore, the therapeutic targeting of ADK in astrocytes is of therapeutic interest for the treatment of epilepsy. In line with this goal, mouse embryonic stem cells were engineered to lack both alleles of Adk. Using an in vitro differentiation protocol, the Adk-null ES cells were transformed in glial cells and shown to release adenosine as a direct consequence of this therapeutic manipulation (Fedele et al., 2004). Transplantation of these adenosine releasing cells into the infrahippocampal fissure in rats robustly suppressed the development off kindling (Li et al., 2007b). These findings show that alteration of glial metabolism is of therapeutic interest.

Adenosine for epilepsy prevention

An additional, ketone-independent mechanism of ketogenic diet therapy is an increase in adenosine, a multifunctional metabolite controlled by astrocytic metabolic clearance, in conjunction with downregulation of ADK, which is a likely explanation for the disease modifying properties of metabolic therapies (Lusardi et al., 2015, Masino et al., 2011). Because maladaptive increases of ADK expression in astrocytes drive the epileptogenic process through increased DNA methylation, therapeutic adenosine augmentation is a rational approach for epilepsy prevention (Figure 1). Thus, adenosine delivered via cell-based brain implants suppressed both kindling epileptogenesis and the development of chronic seizures in kainate induced epilepsy models (Li et al., 2009, Li et al., 2007b, Li et al., 2008). Adenosine delivered locally and transiently to the hippocampus of rats after the onset of epilepsy via silk-based implants prevented epilepsy progression (Williams-Karnesky et al., 2013), and a transient systemic dose of the small molecule ADK inhibitor 5-iodotubercidin attenuated the epileptogenic process in mice after an intrahippocampal injection of kainic acid (Sandau et al., 2019). Because one isoform of ADK is specifically expressed in the cell nucleus (ADK-L) the opportunity exists to directly target nuclear adenosine metabolism to capitalize on the epigenetic mechanisms of adenosine, while minimizing excessive rise of adenosine in the extracellular space. Importantly, a transient treatment with adenosine for only ten days after the onset of epilepsy in rats or an adenosine elevating drug given transiently during the latent period of epileptogenesis for only five days beginning with a delay of three days after an epilepsy triggering SE provided lasting suppression of epilepsy progression and development, respectively (Sandau et al., 2019, Williams-Karnesky et al., 2013). An early therapeutic intervention would also reduce additional risk factors associated with progression of epilepsy, such as sudden unexpected death in epilepsy (SUDEP) and the development of comorbidities and ASM resistance. The epilepsy preventing properties of adenosine are in line with clinical findings showing that a single nucleotide polymorphism in the Adk gene constitutes a biomarker for increased risk of posttraumatic epileptogenesis after a traumatic brain injury (Diamond et al., 2015). Together, those studies demonstrate that transient metabolic treatments and strategies, which decrease astrocytic adenosine metabolism and thereby increase adenosine availability have the unique potential to disrupt the epileptogenic process via an epigenetic mechanism (Boison and Rho, 2020, Boison and Steinhauser, 2017). The transient use of preventative treatments would mitigate adverse effects associated with chronic treatment regimens in established epilepsy.

Astrocyte energy (dys)regulation in neuro-degenerative diseases

In addition to the previously mentioned role of astrocytes in ensuring brain energy homeostasis, they also possess immunocompetency manifested through the expression of a variety of pattern recognition and cytokine receptors which help trigger an inflammatory response in the presence of brain damage (Farina et al., 2007, John et al., 2005). This is more pronounced in the aging brain where disruptions in homeostatic mechanisms and dysregulated immune signaling could give rise to neuronal death with subsequent changes in cognitive function, a hallmark of many neuro-degenerative diseases such as Alzheimer’s disease (AD). Astrocytes possess functional plasticity which is useful in metabolic regulation; however, reactive astrogliosis is associated with alterations in bio-energetic pathways which can lead to the development and worsening of neuro-degenerative diseases (Zheng et al., 2021).

When astrocytes are exposed to inflammatory stimuli, they have been found to exhibit alterations in metabolism viz: increased glycolytic flux, decreased rate of oxidative phosphorylation, and an increased rate of production of reactive oxygen species (ROS) (Motori et al., 2013). Regarding the latter, a phenomenon known as mitochondrial fragmentation has been described in which astrocytes undergo significant changes in the mitochondrial network leading to fragmentation and impaired energy metabolism/respiration rate when exposed to pro-inflammatory stimuli (Motori et al., 2013). In the same vein, astrocytes exposed to pro-inflammatory cytokines such as TNF-α were found to have impaired mitochondrial respiration, increased glycolytic rate, and cumulative decrease in ATP production (Pamies et al., 2021, Russ et al., 2021). Taken together, these findings suggest that since neuro-degenerative diseases, such as Alzheimer’s are associated with increased neuro-inflammatory signaling, there is a change in astrocyte bioenergetics which could lead to a preponderance of fissuring of networks and subsequently impaired function due to increased ATP consumption with impaired regeneration.

In Alzheimer’s disease, astrocytes exhibit increased cytokine signaling, dysregulated calcium homeostasis, and increased hypertrophy and hyperplasia in response to accumulated Aβ-plaques (Oksanen et al., 2017). This implies an upregulation of glycolytic flux as a compensatory mechanism for alteration in bioenergetics. However, as the disease progresses, dysregulated metabolism which have been reported in both animal and human transcriptomic AD analysis become prevalent, inducing a hypometabolic and hypo-energetic state that further compounds the problem (Yoo et al., 2020, Zheng et al., 2021). Many studies have shown that in response to amyloid plaque accumulation, the ensuing astrogliosis culminates in increased ROS production, reduced neuronal cell viability, and depletion of ATP levels which contribute to the progression of pathology (Allaman et al., 2010, Joshi et al., 2019).

In the same vein, there are reports that the ApoE4 allele in AD pathology hinders the ability of astrocytes to metabolize fatty acids for the use by neurons. This is specifically accomplished by decreasing the capability of lipid droplet uptake by astrocytes as well as a decreased rate of fatty acid oxidation (Farmer et al., 2019, Qi et al., 2021). This propagates AD pathology by increasing the rate of Aβ aggregation and decreased clearance due to cell-specific alterations in cellular metabolism (Qi et al., 2021). Similarly, variations in transcriptional pathways have also been shown to affect lipid metabolism and subsequently, the astrocytic response to neuro-degenerative states such as those found in AD. Precisely, decreased expression of forkhead box O transcription factor 3 (Fox O3) has been linked to a decrease in Aβ clearance as a result of decreased fatty acid oxidation (Du et al., 2021).

Conclusions and perspectives

Glia and in particular astrocytes play a preeminent role in the regulation of brain energy metabolism, which constitutes a new frontier for therapy development. The therapeutic reprogramming of glial metabolism offers unique strategies for disease modification and preventative therapeutic strategies. These opportunities have extensively been explored in the field of epilepsy. Thus, the therapeutic reprogramming of adenosine metabolism in astrocytes, via manipulation of the astrocyte-based enzyme ADK was shown to prevent the development of epilepsy and its progression. Because similar glial based pathogenic mechanisms have also been characterized in neurodegenerative conditions such as Alzheimer’s, Parkinson’s, and amyotrophic lateral sclerosis (Boison and Aronica, 2015), disease modifying therapies based on reprogramming of astroglial metabolism hold promise for a wide range of preventative therapeutic interventions. In line with this perspective, metabolic therapies including the ketogenic diet are now widely used in a broad range of brain pathologies.