Abstract
Glutamate is the main excitatory neurotransmitter in the brain, and its excitatory neurotoxicity is closely related to the occurrence and development of Alzheimer’s disease. However, increasing evidence shows that in the process of Alzheimer’s disease, glutamate is not only limited to its excitotoxicity as a neurotransmitter but also related to the disorder of its metabolic balance. The balance of glutamate metabolism in the brain is an important determinant of central nervous system health, and the maintenance of this balance is closely related to glutamate uptake, glutamate circulation, intracellular mitochondrial transport, and mitochondrial metabolism. In this paper, we intend to elaborate the key role of mitochondrial glutamate metabolism in the pathogenesis of Alzheimer’s disease and review glutamate metabolism in mitochondria as a potential target in the treatment of Alzheimer’s disease.
INTRODUCTION
Alzheimer’s disease (AD) is a slowly progressive neurodegenerative disease characterized by the loss of neurons and synapses in the cerebral cortex, leading to cognitive impairment and profound memory loss. Among the many hypotheses regarding the occurrence and development of AD, the excitatory neurotoxicity induced by the neurotransmitter glutamate has been widely accepted. Fifty years ago, the pioneering study by Olney provided the first evidence of the neurotoxicity of the excitatory neurotransmitter glutamate [1]. Since then, scholars have discovered that glutamate induces signaling cascades and calcium overload by overstimulating postsynaptic neuronal glutamate receptors, leading to mitochondrial dysfunction and ultimately initiating the cell death program [1–5]. As in-depth research is conducted, the key role of glutamate metabolism in mitochondria has been widely studied.
Glutamate not only functions as a neurotransmitter to excite neurons but is also the precursor and intermediate in metabolism. The majority of glutamate uptake is accomplished by astrocytes in the synaptic gap through the sodium-dependent high-affinity transporters, excitatory amino acid transporters (EAATs), which are mainly dependent on the specific glutamate transporters EAAT1 (GLAST) and EAAT2 (GLT-1) [6–10]. After glutamate is transported to astrocytes, it is converted to glutamine by glutamine synthetase. Glutamine is subsequently released and taken up by neurons for resynthesis of glutamate by glutaminase, establishing the glutamate–glutamine cycle (Glu-Gln cycle) (Fig. 1). On the other hand, glutamate is transported from the cytoplasm into mitochondria through glutamate carriers and aspartate-glutamate carriers. The main enzymes involved in glutamate synthesis and metabolism in mitochondria are the mitochondrial enzymes glutamate dehydrogenase, aspartate aminotransferase, alanine aminotransferase, branched chain aminotransferase, and phosphoric activated glutaminase [11] (Fig. 2). With the help of these enzymes, glutamate becomes the precursor and intermediate of energy metabolism and amino acid metabolism. The balance of glutamate metabolism is an important determinant of central nervous system (CNS) health. When disruptions in glutamate metabolism and mitochondrial function homeostasis appear, they will inevitably cause diseases in the CNS.
The uptake, transport, and circulation of glutamate between neurons and astrocytes. (a) The carriers involved in glutamate uptake across the cell membrane are mainly excitatory amino acid transporters (EAATs). (b) The carriers involved in glutamate transport across the mitochondrial membrane are aspartate-glutamate carriers (AGCs) and glutamate carriers (GCs). (c) Tricarboxylic acid (TCA) cycle: Pyruvate is generated from glucose through glycolysis, and then pyruvate enters the mitochondria and undergoes oxidative decarboxylation to generate acetyl-CoA and enter the TCA cycle. After a series of steps to synthesize α-ketoglutarate (α-KG), α-KG is used to produce glutamate under the action of glutamate dehydrogenase (GDH). (d) Glutamate–glutamine (Glu-Gln) cycle: Glutamate is released into the synaptic cleft through synaptic vesicles and then transported to astrocytes via EAATs. In astrocytes, glutamine synthetase (GS) is converted to glutamine, and then glutamine is released into the synaptic cleft and taken up by neurons to synthesize glutamate by glutaminase.
Glutamate metabolism and synthesis in astrocyte mitochondria. (a) Malate-aspartate shuttle (MAS): Aspartate, which is exchanged out by aspartate-glutamate carriers (AGCs), is converted into malate in the intermembrane space, and malate is transported into the mitochondrial matrix by the malate-α-ketoglutarate carrier for resynthesis into aspartate. This shuttling activity carries electrons into the electron transport chain of mitochondria as reducing equivalents to generate ATP for oxidative phosphorylation (OXPHOS). (b) Glutamate metabolism in astrocyte mitochondria: Glutamate is transported to mitochondria through glutamate carriers (GCs) or AGCs and converted to α-ketoglutarate (α-KG) through aspartate aminotransferase (AAT), glutamate dehydrogenase (GDH), branched chain aminotransferase (BCAT), and alanine aminotransferase (ALAT). α-KG is converted to malate. Malate is exported to the cytoplasm and converted to pyruvate by malic enzyme (ME). Pyruvate is converted to lactate or acetyl-CoA by pyruvate dehydrogenase (PD) to achieve complete oxidation. In astrocytes, the astrocyte-specific enzyme pyruvate carboxylase (PC) converts pyruvate into oxaloacetate (OAA), and OAA is converted into α-KG and α-KG is converted into glutamate through AAT or GDH. The glutamate produced through this pathway can enter the tricarboxylic acid (TCA) cycle again. (c) Glutamate synthesis in the mitochondria of astrocytes: Glutamine enters the mitochondria of astrocytes and is catalyzed by phosphate-activated glutaminase (PAG) on the outer surface of the inner membrane of the mitochondria to synthesize glutamate and release ammonia.
Excessive glutamatergic signaling is a key feature of neurodegenerative diseases [12, 13]. Mitochondrial Ca2 + influx [14–16], mitochondrial dysfunction [17–21], and mitochondrial oxidative stress and damage [22, 23] caused by glutamate metabolism and circulation disorders play important roles in neurodegenerative diseases [24]. Glutamate overload leads to increased metabolic stress, which causes mitochondrial dysfunction and eventually leads to the occurrence and development of AD. Therefore, mitochondrial dysfunction and imbalanced glutamate metabolism are jointly involved in the pathological process of AD [25–27]. To date, the use of N-methyl-D-aspartate receptors inhibitors, antioxidants, mitochondrial inner membrane permeability transition pore inhibitors, uncoupling agents, and other treatment methods to treat neurodegenerative diseases has attracted interest based on the glutamatergic theory. Given the close correlation between glutamate metabolism and mitochondrial function, rescuing neuronal mitochondria has become a direction for AD treatment. Therefore, initial treatment with glutamate might be a feasible approach [28].
In this article, we intend to elaborate on the key role of glutamate metabolism in the pathogenesis of AD and review the progress of AD therapeutic research targeting mitochondrial glutamate metabolism.
TRANSPORT AND METABOLISM OF GLUTAMATE IN MITOCHONDRIA
Mitochondrial glutamate transport
The entry of glutamate into mitochondria is accomplished by specific carriers located on the mitochondria. Mitochondrial glutamate carriers and aspartate-glutamate carriers are the main carriers (Fig. 2).
Mitochondrial glutamate carriers
Mitochondrial glutamate carriers (GCs), mainly consisting of glutamate carrier isoform 1 (GC1; SLC25A22) and isoform 2 (GC2; SLC25A18), are transporters that catalyze the cotransport of glutamate with protons (H+) through the inner mitochondrial membrane [29]. Studies have confirmed that GC1 is expressed in all organs, with the highest expression in the pancreas, liver, brain, and testis and lower levels in the heart, kidney, lung, small intestine, and spleen. In contrast, GC2 is mainly expressed in the brain and expressed at low levels in all other tissues [29]. GC1 and GC2 are expressed at the same level in the brain, and GC1 is expressed at a higher level than GC2 in other sites outside the brain. In the brain, GC2 expression is higher in the cortex than in the cerebellum and increases with age [30–32]. GC1 and GC2 are present in oligodendrocytes [33] and neuronal subsets [34, 35] but are expressed at relatively high levels in astrocytes [30, 37]. Thus, GC expression levels differ by brain region, age, and neuronal cell type, which may result from the tissue-specific function of GCs, as GC1 and GC2 meet different metabolic requirements.
Differences in gas chromatography kinetics show that GC1 transports glutamate faster, while GC2 has a higher affinity for glutamate, suggesting that glutamate metabolism is divided into basal and on-demand metabolism [29]. When expressed in the same cell, GC2 appears to conform to the basic metabolic requirements of the cell and is responsible for fulfilling the basic functions of glutamate metabolism; when the cell must complete tasks requiring higher metabolic rates, GC1 is adapted to higher requirements associated with specific metabolic functions (e.g., urine production) and/or specific metabolic conditions (e.g., protein-rich diet) [29]. Thus, when the transport capacity of GC2, which is responsible for basal cellular metabolism, is unable to meet the metabolic demand, GC1, which satisfies on-demand metabolism, may be activated because of the increase in cytoplasmic glutamate concentration. In conclusion, GC1 and GC2 have different Km and Vmax values, which are keys to regulating basal or on-demand metabolism of glutamate in cells and potentially regulate the rate of glutamate transport to mitochondria according to tissue-specific metabolic demands.
Dysfunction of GCs may lead to neurological defects. Due to the key role of GCs in glutamate metabolism, GCs dysfunction is closely related to the pathogenesis of epilepsy [38, 39]. GC1 inactivation in astrocytes leads to reduced levels of NAD+ and ATP, as well as intracellular accumulation of glutamate [17], and therefore may lead to altered neuronal synchrony and epilepsy in GC1-deficient patients [38]. According to clinical studies, GC1 deficiency causes early infantile epileptic encephalopathy type 3 (EIEE3, OMIM 609304). EIEE3 manifests as unstable, refractory seizures, usually myoclonus, and EEG shows an inhibitory seizure pattern [38–41]. GC2 expression appears to correlate with the level of inflammation, as GC2 levels are significantly reduced after spinal cord injury [42], whereas GC2 expression is increased after treatment with IL4/IL13 in macrophage cultures [43]. Clearly, GCs are essential for glutamate metabolism and supplying energy to the cell. The association of GCs with glutamate accumulation and inflammation also suggests a potential role for GCs in neurological health and warrants further study in the context of clinical presentation.
Aspartate-glutamate carriers
In addition to the pivotal role of GCs in the mitochondrial transport of glutamate, aspartate-glutamate carriers (AGCs) are also involved in the mitochondrial transport of glutamate. Aspartate-glutamate carriers include aspartate-glutamate carrier subtype 1 (AGC1, SLC25A12, and Aralar1) and subtype 2 (AGC2, SLC25A13, and Citrin). They are antiporters. AGCs transport glutamate and a proton into mitochondria in exchange for aspartate/cysteine [44]. Aspartate, which is exported by AGCs, is converted into malate in the intermembrane space, and malate is transported into the mitochondrial matrix by the malate-α-ketoglutarate carrier for resynthesis into aspartate. This shuttling activity carries electrons into the electron transport chain of mitochondria as reducing equivalents to generate ATP for oxidative phosphorylation (OXPHOS). This shuttling activity is called the malate-aspartate shuttle (MAS) (Fig. 2). AGCs participate in MAS activities due to reverse transport of aspartic acid. Therefore, AGCs further regulate the balance of glycolysis in the cytoplasm and OXPHOS in the mitochondria [45, 46].
AGC1 is expressed at high levels in the heart, skeletal muscle, and brain [47–50]. AGC2 is mainly expressed in the liver, kidney, and heart [51]. Tissues expressing AGC2 at higher levels than AGC1 usually show high expression of genes related to the urea cycle [51]. In the CNS, AGC1 is expressed at higher levels than AGC2 [50, 53]. The expression of AGC1 is higher in neurons than in oligodendrocytes and astrocytes [36, 55]. Additionally, the expression of AGC1 in astrocytes increases with maturity [56]. The kinetics and calcium sensitivity of the two carriers are also different [57, 58]. In terms of aspartate and glutamate uptake, the Km values of AGC1 and AGC2 are approximately 0.05 mM and 0.2 mM, respectively, while the Vmax of AGC2 (∼200 mmol/min/g protein) is approximately 4 times that of AGC1 [57]. Different subtypes of AGCs are suggested to play different roles in glutamate/aspartate metabolism and regulate the transport rates of glutamate and aspartate according to the metabolic needs of different tissues to meet the needs of basic metabolism and on-demand metabolism and to play a role in glutamate and aspartate homeostasis.
AGCs play an important role in the regulation of glutamate and aspartate homeostasis in mitochondria and cytoplasmic compartments. They are involved in mitochondrial oxidative phosphorylation, lipid metabolism, the urea cycle and glycolysis in the cytoplasm [45, 60]. Because the activity of AGCs is affected by Ca2 +, AGCs become an important target for Ca2 + to regulate urea production, gluconeogenesis, and amino acid catabolism [45]. Mutations in the human AGC1 gene have been reported to lead to a significant decrease in its transport function and AGC1 deficiency (OMIM 612949) [61, 62], potentially resulting in diseases such as infantile epilepsy, congenital hypotonia, and developmental delay, as well as insufficient myelination and decreased N-acetylaspartic acid (NAA) levels in the brain. The AGC1 deletion phenotype has also been reported in a study of AGC1 knockout mice, which exhibited growth retardation, impaired CNS function [63], and impaired proliferation of brain precursor cells [64]. Clinical studies have confirmed that mutations in the gene encoding AGC2 cause liver disease: neonatal intrahepatic cholestasis caused by citrin deficiency (NICCD, OMIM 605814) and type 2 citrullinemia (CTLN2, OMIM 215700), due to lack of cytoplasmic NADH oxidation and impaired urea cycle function [65]. Thus, the function of AGCs is closely related to the MAS shuttle, myelin formation, and urea cycle. The effects of AGCs involve many aspects of the nervous system and may become the target of degenerative neuropathy research in the future.
Glutamate is a precursor and intermediate in metabolism
Glutamate functions as a precursor and intermediate in energy metabolism and amino acid metabolism (Figs. 1 2).
Energy metabolism of glutamate
Glutamate is transported to mitochondria through GCs or AGCs and is converted to α-ketoglutarate (α-KG) by aspartate aminotransferase (AAT) or glutamate dehydrogenase (GDH) [66]. Finally, complete oxidation is achieved. In astrocytes, the astrocyte-specific enzyme pyruvate carboxylase (PC) converts pyruvate to oxaloacetate (OAA), OAA is converted to α-KG, and α-KG is converted to glutamate by AAT or GDH. The glutamate produced in this process can enter the tricarboxylic acid (TCA) cycle again (Fig. 2).
Glutamate enters the TCA cycle through conversion into α-KG in mitochondria and becomes an important substrate for the production of ATP. In addition, the processes mediated by GDH, AAT, and other enzymes are involved in the transmission of the electron respiratory chain through the production of NADH. Therefore, the most important product of glutamate oxidative metabolism is ATP. Notably, the ATP produced by the oxidative metabolism of glutamate in astrocytes exceeds the amount of ATP required for glutamate uptake [67], and these ATP molecules are mainly used to meet the needs of sodium pumps and glutamine synthesis [67–69]. A small amount of ATP produced by oxidative metabolism of glutamate is used to meet the needs of the Glu-Gln cycle [16, 68]. In the case of an energy imbalance, ATP-dependent transporters do not function, thereby affecting the uptake and transport of glutamate. The energy imbalance will affect the metabolic activities of enzymes involved in glutamate metabolism. The complete oxidative metabolism of glutamate is mainly achieved by GDH [70]. GDH is inactive under conditions of sufficient energy, low levels of activities of daily living, high GTP levels, and an adequate glucose supply [71]. Moreover, the relationship between glutamate signaling and the energy balance tends to regulate glutamate signaling through energy activity. Cornell-Bell et al. found that the astrocyte network may constitute a long-range signaling system for neurons in the brain. Therefore, excess ATP can be used to meet the energy needs of other astrocytes mediated by glutamatergic signals through the astrocyte network, such as the ATP-dependent actin cycle and filiform movement [72, 73]. Energy is required for glutamate uptake, transport, metabolism, and signal transduction. Once energy metabolism is imbalanced, it will affect the progress of the glutamate pathway.
In amino acid metabolism, glutamic acid is both a substrate and a product
Regardless of whether glutamic acid is used as a TCA substrate or a metabolic intermediate, all metabolic processes are inseparable from the participation of enzymes. The use of glutamic acid as a substrate or a product is regulated by pH and energy requirements. These processes are introduced below.
Glutamate dehydrogenase
The complete oxidative metabolism of glutamate is mainly achieved by oxidative deamination catalyzed by GDH [70]. Moreover, the process by which GDH catalyzes the conversion of glutamate to α-KG is reversible mutual conversion [74] (Fig. 2). This mutual conversion requires the cofactors NADPH and NADH [75]. The direction of the reaction is determined by the prevailing conditions [76]. At physiological pH (pH 7.2–7.4), the synthesis direction is positive. Therefore, in the brain, the normal direction of the reaction is conducive to the oxidative deamination reaction of glutamate to generate the metabolic substrate α-KG.
The oxidation of glutamate catalyzed by GDH enables the complete oxidation of glutamate [67, 78]. In addition, GDH improves the catalytic efficiency of the cycle by providing additional α-ketoglutarate to the TCA cycle. It is also considered a supplementary pathway for glutamate [66], which plays an important role in the normal neurotransmission of synapses and supplying TCA cycle intermediates. In addition, Bauer et al. found that the inhibition of glutamate uptake by astrocytes is achieved by inhibiting GDH in mitochondria [79]. Therefore, GDH seems to be coupled to glutamate transporters through mitochondria [16, 80]. This coupling may provide the required energy to maintain glutamate transport, and thus it is very important [66, 81]. GDH is inactive under conditions of sufficient energy, low levels of activities of daily living, high GTP levels, and an adequate glucose supply [71]. However, under the condition of an increased energy demand, high levels of ADP and low levels of GTP are beneficial for GDH to metabolize glutamate to α-KG [71], which stimulates the TCA cycle to produce energy. Therefore, GDH plays a role in maintaining energy metabolism and backup glucose metabolism in glutamatergic activities.
Humans express two GDH isoenzymes, namely, hGDH1 and hGDH2 [82]. GDH1 is common to all mammals, while GDH2 is caused by gene duplication in higher primates [82]. In human tissues, GDH2 is expressed at particularly high levels in the brain. Immunohistochemical biomolecular studies of the human cortex showed that hGDH1 is expressed in glial cells (including astrocytes) but not in neuronal cells. Additionally, hGDH2 is expressed in both astrocytes and neurons [82]. GDH1 functions as an energy switch. Because its activity triggers the TCA cycle according to the energy state, it is inhibited by GTP and activated by ADP [83]. Therefore, GDH1 activity is strictly regulated by these metabolic intermediates. However, due to the unique high metabolic requirements of neurotransmission, GDH2 may provide a metabolic advantage. According to Spanaki et al, “The evolution of hGDH2 endowed large human neurons with enhanced glutamate metabolism, thereby enhancing cortical excitatory transmission” [82].
In the brain, fine regulation of glutamate metabolism by GDH is important for both energy homeostasis and excitatory transmission [71, 84]. Alterations in GDH expression or activity have been observed in individuals with many neurological disorders, including AD, schizophrenia, and temporal lobe epilepsy. Dysregulated GDH activity in the CNS is strongly associated with neurological disorders [85–88]. GDH activators ameliorate injury in vivo by increasing α-KG levels. In vitro, increased GDH activity in injured neurons can increase ATP levels. Thus, in an energy crisis, neuronal mitochondrial activity increases and the risk of excitotoxicity decreases. Taken together, modulation of GDH activity under energy-depleted conditions may be a plausible strategy to maintain mitochondrial functions in neurons, thereby preventing metabolic failure. Modulation of GDH activity in the human brain may be a promising therapeutic approach for the treatment of neurodegenerative diseases [84, 89–91].
Mitochondrial glutamate aminotransferase
The partial oxidation metabolism of glutamate is achieved by multiple amino acid transferases that mainly include AAT, alanine aminotransferase (ALAT), and branched chain aminotransferase (BCAT) [92, 93].
Aspartate aminotransferase. AAT catalyzes the reversible mutual conversion of aspartate, α-KG, oxaloacetate, and glutamate [94] (Fig. 2). It is present in all organisms. It has the highest specific activity among aminotransferases in the brain [95]. The equilibrium constant for the enzyme is close to unity [96]. Therefore, the reaction readily occurs in both directions and is generally considered an exchange reaction. AAT is expressed as two isozymes in the brain, mitochondrial (mit) and cytoplasmic (cyt), and their activity is very high [97]. AAT catalyzes the production of aspartate and α-KG in mitochondria, and thus it participates in the MAS shuttle and becomes an important component of this pathway [98, 99]. AAT also works synergistically with GDH in glutamate metabolism. Only GDH adds or removes ammonia, and aminotransferase uses glutamate as a substrate to transfer amino groups from glutamate to keto acid [95, 100].
Alanine aminotransferase. ALAT catalyzes the reversible mutual conversion of alanine, α-KG, pyruvate, and glutamate, and its activity in the brain and cultured brain cells is lower than that of AAT and GDH [66, 101–103] (Fig. 2). The equilibrium constant for the enzyme is close to unity [96]. Therefore, the reaction proceeds easily in both directions. Because pyruvate (one of the substrates of ALAT) is the product of glycolysis, ALAT links the glycolysis pathway with amino acid metabolism. Moreover, as alanine aminotransferase is expressed in astrocytes and neurons, it may play a role in ammonia nitrogen transfer between these cells [104].
Branched chain aminotransferase. BCAT combines three branched chain amino acids (valine, leucine, and isoleucine) with α-KG through the transamination of aminotransferase to form glutamate and three keto acids (α-ketoisovalerate, α-ketoisocaproate, and α-keto-β-methylvalerate) [95] (Fig. 2). The brain exhibits relatively high BCAT activity and has two types of isoenzymes: mitochondrial (BCATm) and cytosolic (BCATc) enzymes [105]. The distribution of these two types of isoenzymes in brain cells differs. BCATm is selectively located in astrocytes, while BCATc is located in neurons [106]. Because branched chain amino acids also mediate the BCAT-dependent ammonia nitrogen shuttle between astrocytes and neurons, this selective localization plays an important functional role [106, 107]. Studies using isolated neurons and astrocytes have shown that branched chain amino acids are metabolized only in the brain at a relatively moderate rate [107]. Branched chain amino acids are the key amino acids required for de novo glutamate synthesis because they contribute approximately 40%of the releasable synaptic glutamate. Dietary supplementation of branched chain amino acids after brain injury may improve hippocampal-dependent cognitive dysfunction and exerts a neuroprotective effect [108]. Therefore, the role of BCAT in the brain should be seriously considered.
Phosphate-activated glutaminase
Phosphate-activated glutaminase (PAG) is an enzyme in the mitochondria that hydrolyzes glutamine to glutamate. Glutamine enters the mitochondria of astrocytes and is catalyzed by PAG on the outer surface of the inner mitochondrial membrane [109, 110] to synthesize glutamate and release ammonia (Fig. 2). PAG is expressed by astrocytes [111], is essential for the synthesis of the transmitter glutamate in glutamatergic neurons, and plays an important role in the synthesis of the neurotransmitter γ-aminobutyric acid (GABA) [112, 113]. Two main types of subsequent metabolism of glutamate synthesized by PAG have been identified. First, glutamate enters the cytoplasm directly from the mitochondria, where it enters the transmitter pool. Second, glutamate released from mitochondria participates in the MAS shuttle and GABA synthesis in the cytoplasm, mechanically linking glucose oxidation with the Glu-Gln and Gln-GABA cycles [112, 115].
Glutamate metabolism is also affected by lipid metabolism
Based on the available evidence, only 0.7%of the healthy adult brain proteome is lipoxidatively modified. Importantly, the degree of lipoxidative damage increases with age, and in all the identified proteins, increased lipoxidation is not due to a higher content of the corresponding protein but rather to increased flux of protein damage. This small but significant pool of lipoxidated protein will probably increase over time. The structural characterization of the modified proteins in the healthy adult human brain with PredictProtein software (https://www.predictprotein.org/) reveals some shared specific traits that may explain this specificity for rendering proteins more susceptible to oxidative damage [116]. One of the main amino acids detected in exposed regions is glutamate, a frequent target of nonenzymatic modification of proteins. Enzymes involved in the TCA cycle are modified by lipid oxidation [117]. Important enzymes involved in the process of glutamate metabolism (such as GDH and glutamine synthetase [GS]) are also affected by lipid oxidation [117, 118]. The enzymes that metabolize glutamate and the uptake of glutamate are also affected by lipid metabolism. Activation of NMDA (N-methyl-D-aspartate) receptors by the neurotransmitter glutamate stimulates phospholipase A2 to release arachidonic acid. This second messenger facilitates long-term potentiation of glutamatergic synapses in the hippocampus by blocking glutamate uptake [119].
GLUTAMATE BALANCE AND MITOCHONDRIAL FUNCTION HOMEOSTASIS PLAY AN IMPORTANT ROLE IN AD
The balance of glutamate is an important determinant of CNS health. The most important metabolic site of glutamate is the mitochondria. Therefore, when glutamate metabolism and mitochondrial function homeostasis appear to be disrupted, diseases in the CNS inevitably occur.
In the pathological process of AD, Aβ and tau disrupt glutamate uptake
Glutamate in the synaptic cleft is taken up by the sodium ion-dependent high-affinity transporter EAATs. More than 95%of glutamate uptake mainly depends on the specific glutamate transporters EAAT1 (GLAST) and EAAT2 (GLT-1) in astrocytes [6–10]. Studies of human tissue samples showed decreased the expression of astrocyte-specific glutamate transporters EAAT1 (GLAST) and EAAT2 (GLT-1) in the cortex and hippocampus of patients with AD, indicating a significant functional impairment in glutamate uptake by astrocytes in the brains of patients with AD [120–123]. Casano et al. also observed decreases in EAAT1 (GLAST), EAAT2 (GLT-1), and EAAT3 (EAAC1) expression in the frontal cortex and hippocampus of AD mice [124]. Moreover, when observing the characteristic pathological manifestations of AD, researchers found that amyloid-β (Aβ) reduced the function and expression of EAAT1 and EAAT2 in the rat hippocampus and cortical astrocytes [125–127]. The accumulation and mislocation of hyperphosphorylated tau in neurons disrupt glutamate transport and neuronal synaptic function [128, 129]. In summary, the accumulated evidence indicates that during the pathological process of AD, glutamate uptake by astrocytes in the hippocampus and cortex is disturbed, and Aβ and Tau are involved in disrupting glutamate uptake.
During the development of AD, excessive glutamatergic signals are directly activated by Aβ oligomers
Glutamate is the main excitatory neurotransmitter in the CNS. The dysregulation of glutamate uptake in the synaptic cleft is followed by sustained overactivation of NMDA receptors (NMDARs) at the postsynaptic terminal. In addition to reducing glutamate uptake, Aβ oligomers (AβOs) trigger the abnormal release and accumulation of glutamate in astrocytes in the extrasynaptic space, thereby overactivating NMDARs [130, 131]. Changes in glutamatergic signals are involved in synaptic plasticity, learning, and memory. Excess glutamatergic signaling is a key feature of neurodegenerative diseases [12, 13] and plays an important role in the Ca2 + imbalance and synaptic dysfunction in AD [132, 133]. Excess glutamate overstimulates CA1 neurons and induces calcium influx and signaling cascades, leading to excitotoxicity. Gene expression studies in AD-like rats have shown that NMDA receptor subunits (NR1 and NR2B) are differentially expressed in the CA1 and CA3 regions, and NR2B is overexpressed in the CA1 subregion [134]. In addition, AβOs directly activate NMDARs [135], especially those containing NR2B subunits [136, 137]. Thus, during AD development, NMDARs are overexpressed and are directly activated by AβOs. Therefore, dysregulated glutamate uptake may promote the development of AD.
The excitotoxicity of glutamate is involved in the pathological process of AD
Excess glutamate or excess glutamatergic signaling lead to calcium overload, activation of related signaling cascades and excitotoxicity to adjacent neurons in postsynaptic neurons, ultimately initiating the cell death process [1–4]. Excitotoxicity is considered the main factor contributing to the occurrence and development of AD [130, 131].
Excessive activation of glutamatergic neurons leads to calcium overload. Calcium overload leads to an increase in reactive oxygen species (ROS) production, further aggravating neurotoxicity, cell homeostasis disorders, and neuronal death [138–141], leading to the pathological progression of AD. Glutamate changes Ca2 + influx in mitochondria through NMDARs, metabotropic glutamate (mGlu) receptors, GLT-1 and GLAST transporters, and mitochondrial acidification [14–16]. When cytoplasmic calcium levels increase, mitochondrial carriers are activated, and Ca2 + is absorbed by mitochondria [18, 19], causing Ca2 + influx in mitochondria. At the same time as Ca2 + influx, mitochondrial glutamate uptake is affected, which in turn affects OXPHOS and ATP production [20]. Excessive Ca2 + accumulation in the mitochondria will deplete the mitochondrial membrane potential [142], leading to ATP depletion and the opening of the mitochondrial inner membrane permeability transition pore (mtPTP). The opening of mtPTP leads to the release of ROS [22]. ROS damage large molecules in the brain and many different types of cells [143, 144], eventually leading to mitochondrial dysfunction and cell death [145]. This phenomenon of glutamatergic activation leading to increased mitochondrial Ca2 + signaling has been identified in early AD [146], and significant changes in intracellular Ca2 + signaling precede neuronal death and cognitive deterioration in individuals with AD [147]. At the stage when pathological alterations appear in individuals with AD, AβOs also promote the toxic accumulation of ROS caused by Ca2 + dysregulation [148], leading to excessive mitochondrial translocation and fission [149, 150].
Calcium overload, resulting from overactivation of glutamatergic neurons, not only induces an increase in ROS production but also affects mitochondrial movement. In neurons and astrocytes, mitochondrial movement is inhibited by glutamate and Ca2 +, but mitochondrial movement in oligodendrocytes is enhanced by the application of glutamate (approximately 76%) [151]. In contrast, the removal of Ca2 + eliminates the mitochondrial movement in oligodendrocytes [14, 153]. In subjects with AD, Aβ and/or oxidative stress induce Ca2 + signaling, leading to increased DLP1 activation. In addition, the application of glutamate also reduces mitochondrial fusion in organotypic cultured astrocytes [153]. In addition to reducing fusion, increasing glutamate also reduces the length of mitochondria from ∼3μm to 1μm [153]. The aforementioned abnormalities in mitochondrial dynamics, together with increased free radical production and oxidative damage, decreased ATP/ADP ratio and impaired bioenergetics, are major manifestations of mitochondrial dysfunction. They are induced by calcium overload resulting from overactivation of glutamatergic neurons, which is a recognized feature of AD [154–157].
In the process of glutamate excitotoxicity, some other scholars found that the activation of NMDA glutamate receptors also results in increased intracellular levels of zinc ions. Increased Zn2 + levels potentially lead to mitochondrial dysfunction, Ca2 + dysregulation, and ultimately neuronal death [158–162]. Zn2 + chelation exerts a strong neuroprotective effect: it prevents mitochondrial failure, irreversible Ca2 + homeostasis and neuronal death [163]. In neurons, blockade of excitotoxicity-driven Zn2 + elevation reduces mitochondrial dysfunction, improves intracellular calcium circulation [164], and provides neuroprotection in the excitotoxic environment of glutamate [165]. The early stage of AD is characterized by abnormal glutamatergic activation [166], and Zn2 + disorders have been observed in this process [146]. In this case, Ca2 + and Zn2 + cooperate to promote the initial steps of the pathogenic cascade, causing neuron loss associated with AD [65, 167–169]. In the later stages of AD, Zn2 + affects the neurotrophic axis and leads to structural synaptic remodeling, two key processes related to learning and memory performance [170]. Zn2 + affects Aβ metabolism and the resulting oxidative stress and promotes tau pathology [171].
Although accumulating evidence proves that glutamate excitotoxicity is related to AD, the development of drugs for clinical has not been completely successful (see “Current Status of Drug Development”, below). Therefore, our understanding of glutamate excitotoxicity is still limited, and more experiments are required to determine its effect on AD.
Glutamate transport in mitochondria may become the direction of AD research
The entry of glutamate into mitochondria is accomplished by specific carriers on the mitochondria. GCs and AGCs are the main carriers that regulate the transport of glutamate from the cytosol across the mitochondrial membrane. Goubert et al. found that GC1 dysfunction in astrocytes leads to decreased mitochondrial ATP levels and intracellular glutamate accumulation [17], which may cause reverse glutamate transport into the extracellular space [172] and promote excitotoxicity. The study of mitochondrial glutamate carriers may become the key to regulating glutamate mitochondrial metabolism and excitotoxicity. However, at present, few studies have examined the mitochondrial glutamate carriers in the nervous system, and the focus is mainly on AGC1. The transport of aspartate by AGCs to generate acetyl-CoA [173] is essential for myelination. The primary function of oligodendrocytes is to encircle axons in the CNS, forming insulating myelin structures. Therefore, the myelination defect caused by mitochondrial AGC1 abnormalities has become the starting point for research. Undifferentiated mouse neuroblastoma nerve 2A cells downregulate AGC1, show significant proliferation defects associated with decreased mitochondrial respiration, and do not properly synthesize N-acetylaspartate (NAA) [52] (NAA is the precursor for myelin synthesis from aspartate and acetyl-CoA under the action of aspartate-n-acetyltransferase [174]). Consistent with these findings, Petralla et al. reported that a lack of AGC1 led to a decrease in the proliferation of oligodendrocyte precursor cells [64]; Wibom et al. found that AGC1 deficiency caused myelin sheath defects [61]. Therefore, based on the experimental results, the abnormal expression of AGC1 in mitochondria not only affects the proliferation of specific cells but also affects the formation of the myelin sheath, which in turn affects the CNS. As mentioned above, AGCs participate in the MAS shuttle due to the transport of aspartate, and thus the expression of AGC1 is related to increased glycolysis [45] and increased glutamate oxidation [175]. This process is activated by calcium [176]. As shown in the study by Hertz et al, an increase in extramitochondrial Ca2 + activates AGC1 [57], leading to an increased transfer of NADH to mitochondria and an increase in mitochondrial glutamate-dependent respiration by increasing the glutamate supply [20, 177–180]. Moreover, Raini et al. found that astrocytes from patients with AxD (Alexander disease) and VWM (leukoencephalopathy with vanilling white matter) both show dysregulated OXPHOS, increased glycolysis, and altered AGC1 expression [181]. These findings reveal a connection between nerve cells and mitochondrial activity mediated by AGC1. Given the important effect of mitochondrial homeostasis on neurodegenerative diseases, we speculate that the role of mitochondrial glutamate carriers (represented by AGC1) in neurodegenerative diseases (represented by AD) will certainly become a future research direction.
Imbalances in glutamate metabolism together with mitochondrial dysfunction contribute to brain pathology
In astrocytes, mitochondria, AGC1, GC1, and GLT-1 are co-compartmented [14, 183]. Mitochondria accumulate near the glutamate transporter and are presumed to contribute to glutamate metabolism and ATP production to meet the increase in energetics while buffering the ion changes mediated by glutamate uptake [68]. Glutamate exists as an excitatory neurotransmitter and participates in two cycles: “Glu-Gln cycle” and “Gln-GABA cycle” (Fig. 1). In several studies conducted with rodent models of AD, mitochondrial-driven glucose metabolism abnormalities were recorded using magnetic resonance spectroscopy or nuclear magnetic resonance spectroscopy. As a result, glutamate and GABA levels were decreased [184, 185]. Thus, glucose oxidation and neurotransmitter circulation in glutamatergic and GABAergic neurons are impaired in AD models. Impaired glutamine synthase and decreased glutamate flux through the GABA pathway may be caused by mitochondrial dysfunction [186]. NADH produced by OXPHOS in the mitochondrial TCA cycle is used together with glutamate and GABA to synthesize ATP. This process also contributes to the maintenance of synaptic plasticity [187]. Therefore, researchers have speculated that the inhibition of astrocyte mitochondrial respiration may impair the synthesis of neurotransmitter precursors and further reduce energy production by oxidation, thus reducing glutamate intake and possibly further enhancing glutamate-induced damage in the brain [188].
Mitochondrial dysfunction in the brain of subjects with AD may also be induced by an imbalance in glutamate metabolism. In the cortical pyramidal neuron subsets of patients with AD, the astrocyte-specific glutamate transporters GLAST and GS are abnormally expressed, indicating that glutamate metabolism in astrocytes is obviously dysfunctional [120, 121]. Research on mouse astrocytes provides evidence that glutamate promotes glycolysis, impairs mitochondrial respiration, and leads to a decrease in ATP levels [189–193]. In addition, the content of glutamate is upregulated during brain injury, including cerebral ischemia, tumors, and traumatic brain injury [194, 195], indicating that an increase in glutamate levels may lead to changes in the glycolysis rate and participate in pathological changes in the brain. Based on these results, an imbalance in glutamate metabolism damages the mitochondrial function of nerve cells and further enhance damages to the brain [196].
Taken together, glutamate in the brain functions as a “double-edged sword”, which is both an essential neurotransmitter and a neurotoxin that may lead to neurological diseases. Accumulating evidence has revealed the role of glutamate in the progression of AD, not only limited to its role as a neurotransmitter causing excitotoxicity but also related to the disturbance of its metabolic balance. The maintenance of this balance is closely related to intercellular glutamate uptake, intracellular recycling, mitochondrial transport, and mitochondrial function. The metabolic balance of glutamate and the fate of functional mitochondrial homeostasis are closely linked and mutually influence each other, playing an important role in neurodegeneration, as represented by AD [24].
The role of peripheral glutamate metabolism in AD
The receptors, transporters, and enzymes involved in glutamate metabolism are also expressed in peripheral nerve tissues, and thus the peripheral glutamate metabolism process is similar to that in the CNS. Interestingly, glutamate plays different roles in various diseases, all of which are related to the glutamate concentration. Nerve damage caused by ischemia, trauma, and other pathological processes will cause an immediate increase in the extracellular glutamate level. When glutamate accumulates in endothelial cells to a concentration exceeding the plasma level, it enters the bloodstream through diffusion and uptake by EAATs in brain endothelial cells. In addition, the metabolism of α-ketoglutarate by aminotransferase or GDH occurs in endothelial cells and astrocytes, and metabolites are also exported into the blood. In most animal models of traumatic brain injury, glutamate overload is present and glutamate homeostasis is disrupted. This disorder may be caused by a combination of many factors, including excessive release of damaged cells, cytokine stimulation, glutamate clearance obstacles caused by the loss of functional glutamate transporters, and glutamate receptor transport disorders [197–200]. In the latter case, a blood-based scavenger system was shown to decrease the infarct size and transiently abrogate the inhibition of LTP induced by glutamate dyshomeostasis in rat models [201]. This effect has also been reported in animal models of stroke, AD, and other neurological deficits. Recently, drugs that reduce the free extracellular glutamate concentration by increasing the number of functional transporters have been reported to reduce cognitive impairment in transgenic mice overexpressing the human disease-causing mutant Aβ precursor protein [202]. An increase in extracellular glutamate levels is involved in the pathological process of brain injury and AD, and the clearance of peripheral glutamate can protect the hippocampus from Aβ-mediated damage to glutamate homeostasis and synaptic destruction [203].
Glutamate metabolism not only changes with aging but is also affected by genetic risk factors for AD
Glutamate metabolism changes with aging, the content of glutamate in the brain decreases with age, and the function of glutamate decreases with age [204], but the changes in different areas of the brain are not the same. Stephens et al. confirmed significantly reduced glutamate release induced by stimulation in the cornu ammonis 3 (CA3) area of aged rats and an increased glutamate uptake rate of the hippocampal dentate gyrus [205]. The subregions of the mammalian hippocampus exhibit changes in glutamate metabolism during ageing, and these subregions are essential for learning and memory. Compared with young subjects, neuronal mitochondrial metabolism and Glu-Gln cycle flux are approximately 30%lower in elderly subjects. The reductions in individual subjects correlated strongly with reductions in N-acetylaspartate and glutamate concentrations, consistent with chronic reductions in brain mitochondrial function [206]. Burbaeva documented a significant increase in the contents of glutamate-metabolizing enzymes, including GDH, GS, and PAG, in the prefrontal cortex of patients with AD [85]. Glutamate metabolism not only changes with aging but is also affected by genetic risk factors for AD, such as Apolipoprotein E (APOE). The APOE genotype is a powerful genetic modifier of AD. The APOE4 subtype significantly reduces the average age of onset of dementia through an unknown mechanism. Chen et al. found that APOE4 reduces the expression of NMDA and AMPA receptors on the surface of neurons, and by reducing the function of APOE, NMDA, and AMPA receptors, it selectively impairs glutamatergic neurotransmission, thereby promoting Aβ-mediated synaptic suppression [207]. In addition to APOE, early genetic studies on many individuals with early-onset AD identified autosomal dominant mutations in the amyloid protein precursor (APP), presenilin 1 (PSEN1), and presenilin 2 (PSEN2) genes. Among these genetic risk factors associated with early-onset AD, Tambini et al. found that AβPP is involved in promoting the release of glutamatergic synaptic vesicles [208]. The data reported by Zoltowska et al. showed that presenilin 1 interacts with GLT-1 in astrocytes and neurons in vivo and in vitro [209]. This interesting finding may reveal molecular crosstalk between proteins related to the maintenance of glutamate homeostasis and Aβ pathology.
CURRENT STATUS OF DRUG DEVELOPMENT
Glutamate excitotoxicity is a contributing factor to many neurodegenerative diseases, and thus small-molecule compounds that block or reduce the excitotoxicity of extracellular glutamate have become the focus of research.
NMDARs are believed to play a leading role in the pathophysiology of AD. As an NMDA receptor antagonist, memantine is believed to reduce the excitatory neurotoxicity of glutamate [210–212] and has shown moderate efficacy and safety in the treatment of moderate to severe AD [213]. Applications have been submitted to extend the indication of memantine to include the treatment of mild AD, but neither the FDA nor EMEA have granted approval for this indication. Memantine has not been approved for the treatment of other forms of dementia. Therefore, the direction of drug development has shifted to memantine derivatives. Among the many derivatives of memantine, nitromemantine has attracted attention. Various studies have revealed a nitrosylation site on the extracellular domain of NMDARs, and S-nitrosylation of this site reduces the excess activity of the receptor, blocks apoptotic cell death, increases the neuronal survival rate, and thus provides neuroprotection [214]. A nitric oxide–releasing moiety (nitrooxy moiety) from nitroglycerine was attached to memantine to generate the new bifunctional drug nitromemantine that might eliminate these systemic side effects [214]. AβOs cause synaptic inhibition through extrasynaptic NMDARs. Nitromemantine inhibits this synaptic suppression, and the effect of nitromemantine on extrasynaptic NMDARs is greater than that of memantine [130]. Wu et al. found that memantine nitrate MN-08 inhibits Aβ production in transgenic mice. In vitro, MN-08 binds to and antagonizes NMDARs, inhibits calcium influx, and prevents glutamate-induced neuronal loss [215]. In addition to the development of memantine derivatives, many studies have focused on combination therapy with memantine and other drugs, such as heterodimers of tacrine and carbazole, ARN14140 (heterodimers of galantamine and memantine), heterodimers of memantine and 6-chlorocrine. Their common feature is that they not only block NMDARs but also inhibit AChE activity, potentially exhibiting better neuroprotective activity.
Other noncompetitive NMDAR antagonist compounds are under development, such RL-208, (3,4,8,9-tetramethyltetracyclo [4.4.0.03,9.04,8, 4.4.0.03,9.04,8] dec-1-yl) methylamine hydrochloride, a polycyclic amine compound with potentially promising therapeutic effects on age-related cognitive problems and AD [216]. Additionally, pharmacological activators of GLT-1 have been explored for decades and are currently emerging as promising tools for the treatment of multiple neurodegenerative diseases [6, 218].
Therefore, some researchers have begun to consider an enzyme-based alternative approach to minimize glutamate excitotoxicity rather than using traditional pharmacological interventions targeting glutamate transporters/ion channels. For example, Pérez-Mato et al. reported that the administration of human recombinant glutamate oxaloacetate transaminase 1 (GOT1) in a rat model of ischemic stroke (middle cerebral artery occlusion) reduces brain and serum glutamate concentrations [219]. This treatment reduces the stroke-induced infarct volume and sensorimotor impairment, and this reduction is most pronounced when herbacetin is coadministered with the enzyme [219]. Khanna supports the hypothesis that upregulating the expression of adenylate-activating proteins in the CNS is a useful treatment for diseases associated with glutamate excitotoxicity [220].
The mitochondrial effects of Ca2 + appear to be generated following entry through the mitochondrial Ca2 + transporter (MCU). However, Ca2 +-targeted therapy, which mainly blocks Ca2 + from entering the cell, has shown little efficacy [221, 222]. In addition, studies using MCU blockers (blocking mitochondrial Ca2 + uptake) not only produced mixed results but also exacerbated the damage in some studies, possibly partially by preventing mitochondria from buffering Ca2 + loads and leading to faster increases in cytoplasmic Ca2 + loads [223].
Therefore, Zn2 + may be an attractive target because the increase in its levels occurs before the damage becomes irreversible. Although substantial evidence supports the concept that Zn2 + dysregulation plays an important role in glutamate-driven neuronal death that occurs in stroke, brain trauma, and AD, less evidence has been accumulated for excitotoxic diseases, such as amyotrophic lateral sclerosis, Huntington’s disease, or Parkinson’s disease. Additionally, a decrease in brain Zn2 + levels may also exert a negative effect on glutamate neurotransmission [224, 225]. Therefore, preclinical and clinical studies are needed to evaluate when, where, and to what extent Zn2 + disorders should be treated.
In addition, Mg2 + affects the enzyme activity of the TCA cycle in mitochondria, and a decrease in the mitochondrial Mg2 + concentration will lead to decreases in the mitochondrial membrane potential and ATP synthesis [226–228]. Mg2 + outside the mitochondria inhibits mitochondrial Ca2 + uptake and depolarizes the mitochondrial membrane potential [229]. Therefore, Mg2 + plays an important role in maintaining cell survival [230]. Some researchers have reported a lower than normal concentration of Mg2 + in the brains of patients with neurodegenerative diseases [231, 232]. Mg2 + supplementation or overexpression of Mg2 + channels exerts neuroprotective effects on cell and animal models of AD [233]. Shindo et al. found that quinidine and amiloride inhibit the transient decrease in Mg2 + levels induced by glutamate and alleviate the decrease in energy metabolism induced by glutamate [234]. Therefore, intracellular Mg2 + homeostasis and the Mg2 + transport system have potential in the treatment and prevention of AD.
In the pathological process of AD, AβOs and tau (insoluble, abnormal conformation, and hyperphosphorylated tau) disrupt glutamate uptake, triggering the abnormal release and accumulation of glutamate in the extrasynaptic space of astrocytes to subsequently overactivate NMDARs and promote the toxic accumulation of ROS. Therefore, the treatment of AβOs and tau has become the direction of drug research and development. Aducanumab (NCT01677572) is one of these drugs. It is a human IgG1 monoclonal antibody. The donor tends to bind to Aβ and exhibits a dose- and time-dependent effect on the process of removing amyloid from the patient’s brain, slowing the rate of cognitive decline [235]. On Mar 21, 2019, Biogen announced that the anti-amyloid antibody aducanumab failed futility analyses in two identically designed Phase III AD trials, and discontinued its development. However, on June 7, 2021, the U.S. Food and Drug Administration (FDA) approved aducanumab as a new treatment for AD, taking advantage of an Accelerated Approval program [236]. The central controversy of this type of drug is whether the amyloid clearance protects patients from cognitive and functional decline. This should have been answered by two identically designed (pre-approval) Phase III trials, but it was not. Reanalysis of data up to March 2019 confirmed the drug’s ineffectiveness in one study, but the other suggested cognitive benefit. Attempting reassurance, the FDA committed Biogen to a nine-year post-approval confirmatory study. So, we may not know until at least 2030 whether aducanumab slows cognitive decline. Solanezumab (NCT01900665) is another similar drug. However, in a recent double-blind, placebo-controlled Phase III trial, its effectiveness at preventing cognitive decline in patients with mild AD was not significant [237]. Another key target in this category is β-secretase 1, also known as BACE1, which controls the rate-limiting step in the cleavage of AβPP at the β-site to generate Aβ [238]. Despite numerous clinical trials, none of the strategies proved successful. The administration of oleocanthal (OLC) in AβOs-poisoned rat brain astrocyte cultures significantly reduces astrocyte activation and the levels of the IL-6 and GFAP genes and increases GLT-1 expression to promote glutamate clearance, increases GLUT-1 levels to increase glucose uptake, and restores the energy and synaptic function of astrocytes. In patients with AD, when the disease progresses, OLC treatment effectively improves synaptic function [239, 240]. The only anti-Tau agent entering Phase III clinical trials is TRx0237 (NCT01689246), which is a reduced form of methyl sulfoxide chloride designed to prevent or dissolve tau aggregation to reduce tau pathology [241]. However, Phase III trials using different doses of TRx0237 showed no beneficial effects on the cognitive performance of patients with mild-to-moderate AD [242].
The exploration of existing drugs as treatments for AD has continued. For example, donepezil reduces the NR1 level on the cell surface and glutamate-mediated Ca2 + entry, which may contribute to its neuroprotective effect [243]. Another acetylcholinesterase inhibitor, tacrine, reverses the decreases in the mitochondrial membrane potential, ATP production, and glutamate-induced neuronal cell death [244]. Curcumin treatment protects cultured neurons from glutamate-induced excitotoxicity through a mechanism requiring tumor necrosis factor α receptor 2 (TNFR2) activation, suggesting that treatments for cognitive decline designed to selectively enhance TNFR2 signaling may be more beneficial than the use of anti-inflammatory drugs themselves [245]. Riluzole, a glutamate receptor antagonist, reduces glutamate-mediated excitotoxicity. Currently, the drug has entered Phase II clinical trials. BHV-4157 (troriluzole), a prodrug of riluzole, a glutamate regulator, increases synaptic glutamate absorption and reduces synaptic glutamate levels by increasing the expression of glutamate transporters. Phase II and Phase III trials were launched in July 2018 to evaluate the efficacy of BHV-4157 in patients with mild to moderate AD. AXS-05 is a mixture of dextromethorphan (DMP) and bupropion. DMP is an NMDA receptor antagonist, glutamate receptor modulator, sigma-1 receptor agonist, and serotonin and NE transporter inhibitor. Bupropion is a dopamine-NE reuptake inhibitor and CYP2D6 inhibitor that increases the pharmacodynamics of DMP. Phase III trials are ongoing to evaluate the efficacy of AXS-05 on anxiety in patients with AD. Rhynchophylline, a key oxindole alkaloid in the Chinese medicinal herb Uncaria rhynchophylla, suppresses the Aβ-induced activation of extrasynaptic NMDARs and restores impaired LTP and spatial memory in AD model mice [246]. In addition, anemodise A3 is a compound isolated from the Chinese medicinal herb Pulsatilla chinensis. It potentiates hippocampal LTP and spatial memory as a noncompetitive NMDA receptor modulator and facilitates GluA1 phosphorylation and trafficking to synapses [247]. However, its role in alleviating neurodegeneration in AD awaits further investigation. Moreover, the target specificity of herb-derived compounds requires careful consideration.
CONCLUSIONS
Glutamate homeostasis is an important determinant of CNS health. Glutamate transport, circulation, synthesis, and degradation are key functions required for neurotransmission. In addition, as an energy source, glutamate oxidation can buffer the use of glucose in astrocytes and increase glucose availability to neurons. Glutamate forms a buffer or barrier to resist changes in brain amine and ammonia nitrogen levels as a precursor of other important metabolites, including glutathione, glutamine, and GABA. Mitochondria play a vital role in glutamate metabolism. The metabolism of glutamate is inseparable from mitochondrial function. These processes may interact with each other and concomitantly promote the progression of AD. The treatment of various brain injuries and neurodegenerative diseases with approaches targeting the mitochondria has always been a topic of interest. If the preservation of neuronal mitochondria is the direction of AD treatment, glutamate may be a feasible starting point.
