Microglia represent the immune cells resident in the central nervous system (CNS) and are, therefore, the primary immune defense of this central area of the human body. Several studies suggest that microglia mainly originate from the embryonic yolk sac, and this evidence was confirmed in rodent models and humans
[1][2][3][4][5]. In mice, microglia precursors enter the CNS around day 9.5 of embryonic life (E9.5) through extravascular pathways, as the first cerebral capillaries appear only on day E10 and are soon closed off from the periphery by E13.5, due to the blood–brain barrier’s (BBB) formation. In humans, microglia penetrate the cerebral cortex by gestational week 4.5 (GW4.5); then, a second wave of microglia infiltration penetrates the embryonic brain via the vasculature at GW12–13
[6][7][8][9][10][11][12]. After entering the brain, microglial cells rapidly proliferate
[11][13][14][15] and, by GW22, take on a ramified morphology, becoming fully mature by GW35. Early entry and subsequent brain colonization represent actual events in the initial brain development.
Microglia regulate the number of neural precursors, promote neuron survival, and are involved in the phagocytosis of damaged neurons, synaptic pruning, angiogenesis, synaptogenesis, and the maturation of neural circuits
[6][11][16][17][18][19][20][21][22][23]. Microglia have a unique genetic signature with respect to the perivascular, meningeal, and choroidal macrophages in the CNS, probably due to macrophages’ presumed hematopoietic origin and developmental processes
[24]. Similarly, pre-natal and post-natal microglia differ from adult microglia
[25][26]. Microglia play critical physiological roles during CNS development. Resting microglial cells exhibit numerous ramifications that constantly monitor the surrounding microenvironment, thus maintaining CNS homeostasis by phagocytosing cellular debris
[27][28]. Microglia play an essential role in synaptic remodeling through synaptic pruning to optimize neurotransmission processes and are directly involved in the formation and reorganization of neural networks and in providing trophic support to mature neurons
[12][20][29].
Microglia can sense the CNS environment by monitoring the signaling pathway molecules that guarantee the physiological crosstalk between microglia and neurons, which is fundamental for the maintenance of cerebral homeostasis
[12][30][31][32][33]. A minimal variation in the extracellular milieu composition allows microglia to respond by modulating neuronal activity. The mature homeostatic microglia phenotype progresses in a multiple-step process during CNS development and requires continuous instructions from the adult brain
[34][35][36][37][38].
Furthermore, microglia are promptly activated in response to CNS stimuli or pathological conditions and undergo massive morphological and functional changes accompanied by rapid clonal proliferation
[39][40][41], a process identified as microglia activation. Activated microglia change their morphology from branched to amoeboid, resembling macrophages circulating in the bloodstream, and migrate toward the lesion site
[28][42]. Convergence at the injury site occurs in response to signal molecules released by damaged neurons
[42]. Activated microglia-induced neurotoxic effects can occur due to the release of cytotoxic molecules, including pro-inflammatory mediators, such as tumor necrosis factor α (TNF-α) or interferon γ (INFγ), and free radicals, superoxide anion, or nitric oxide (NO), and constituents triggering oxidative stress
[43]. Under specific conditions, activated microglia acquire a defined anti-inflammatory phenotype, thus releasing neuroprotective factors
[44].
Microglia express membrane receptors for several neurotransmitters
[45], allowing them to respond to various external stimuli that determine the cell status. Indeed, some neurotransmitters can influence the activation state of microglia, producing changes in membrane potential as well as in the intracellular calcium concentration, causing the release of cytokines and generating cell motility
[45][46]. In homeostatic or resting conditions, microglia exhibit the so-called “surveillant” phenotype, characterized by small cell bodies, limited cellular mobility, and extensive and highly mobile branches to control the surrounding environment
[28][42]. This quiescent state is commonly identified as neutral, or “M0”
[28][47], and characterized by the low expression of surface markers typical of circulating macrophages, i.e., the common lymphocyte antigen (CD45) and major histocompatibility complex class II (MHCII).
As discussed above, microglia are extremely sensitive to changes in the environment, thus being rapidly activated after exposure to specific signals, such as growth factors, neurotransmitters, or cytokines, that indicate the presence of infection, trauma, neuronal damage, or inflammation
[48][49]. Moreover, microglia classify pathogens by recognizing damage-associated molecular patterns (DAMPs), receptor patterns of exogenous microorganisms, or endogenous cells involved in the immune response. As for peripheral macrophages, according to the first proposed nomenclature, the microglia activation state includes at least two distinct phenotypes: M1, described as a pro-inflammatory and neurotoxic phenotype, and M2, also known as the “alternative activation phenotype”
[50][51][52], with anti-inflammatory and neuroprotective properties. These two phenotypes differently respond to distinct signals from the microenvironment and, in turn, are involved in producing many effector molecules
[53], promoting the transcription of genes that activate cellular defense mechanisms, including the release of inflammatory cytokines and chemokines
[50][54].
However, this unequivocal classification is inconsistent with the vast repertoire of microglial phenotypes and core functions in different situations, i.e., development, plasticity, ageing, and diseases. As recently reported, considering the coexistence of multiple states, microglia phenotypes occurring in a specific condition should be characterized by more potent analysis tools than those presently applied, such as proteomic, metabolomic, transcriptomic, morphological, and epigenetic ones [55]. Accordingly, a new nomenclature is needed to define the microglia phenotype in each specific physio-pathological environment. This scenario is even more complex since microglia cells can acquire specific activation cellular patterns depending on the pathological environmental conditions in which microglia participate, thus leading to different and peculiar “disease-associated microglia” (DAM) phenotypes [56].
2. Glutamate Receptors Expressed in Microglia
The specific activities of microglia likely result from the cell state, regulated by different environmental stimuli, which in turn activate cellular structures, mainly receptors, that act as sensors of external messengers and trigger several intracellular signals with distinct biological functions
[46]. The excellent recent literature reports the presence and the role of many microglia-expressed receptor families. Purinergic, serotoninergic, histaminergic, and cannabinoid receptors are the most relevant in tuning the microglia state by regulating their phenotypic characteristics and functions, including proliferation, branch motility, cytokine release, cell migration, and phagocytosis, in physiological
[57][58][59][60][61] and pathological conditions, including neurodegenerative diseases
[62][63][64][65][66][67][68]. Although supported by limited evidence in the literature, it is worth mentioning that microglia also express GABAergic, cholinergic, adrenergic, and dopaminergic receptors
[45][69].
Glial cells, including microglia, widely express glutamatergic receptors, whose activation exerts numerous crucial effects on the glia themselves and glia–neuron interactions in physiological and pathological conditions
[70][71]. Glutamate (GLU) is the primary excitatory amino acid neurotransmitter in the brain. Once released at the presynaptic level, it activates post-synaptic α-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA), N-methyl D-aspartate (NMDA), and kainate ionotropic receptors to stimulate rapid synaptic transmission
[72]. GLU also activates G-protein-coupled metabotropic receptors (mGluRs) with slower signaling transduction kinetics. mGluRs include eight subtypes classified into three groups, termed I, II, and III, based on the sequence homology, signal transduction mechanism, and pharmacological profile
[73]. When activated by GLU, mGluRs can generate fine feedback mechanisms at the pre-synaptic level, inhibiting or potentiating the release of GLU itself or other neurotransmitters from heterologous nerve terminals
[74][75][76][77]. At the same time, mGluRs regulate critical cellular mechanisms, such as post-synaptic excitatory or inhibitory currents or the activity of glial cells surrounding the synapse, namely astrocytes and microglia
[78]. In addition to the numerous actions in glial cells, the activation of mGluRs mediates the interaction between glia and neurons
[79][80]. The latter effects are complex and bidirectional, often depending on the implicated mGluR subtypes
[80].
Microglia express both ionotropic and metabotropic GLU receptors. These receptors mediate the response to GLU and participate in neuroinflammation and neurodegeneration processes
[69]. Microglial NMDA receptors trigger neuroinflammation and neuronal death
[81] by also driving pro-inflammatory responses via poly(ADP-ribose) polymerase-1 (PARP-1)/transient receptor potential cation channel subfamily M member 2 (TRMP2) signaling
[82]. Thus, the existence of microglial NMDA receptors further offers a link between inflammation and excitotoxicity. Moreover, AMPA and kainate receptor activation triggers microglia reactivity and motor neuron toxicity
[83][84][85]. mGluRs in microglia are involved in neuroinflammation
[86], acute and chronic neurological disorders, and neurodegenerative diseases
[87].
3. Physio-Pathological Role of Group I Metabotropic GLU Receptors Expressed by Microglia
Metabotropic GLU receptors are organized into three groups, termed I, II, and III, overall including eight subtypes (mGlu1–8 receptors)
[73]. Group I include mGlu1 and mGlu5 receptors (henceforth reported as mGluR1 or mGluR5), which couple to a Gq protein, resulting in the activation of phospholipase C (PLC), and inositol triphosphate (IP3) and diacylglycerol (DAG) production. Group I receptors are mainly located in the post-synaptic compartment, and their activation increases cellular excitability. Group II, including mGluR2 and mGluR3, and group III, including mGluR4 and mGluR6–8, couple to Gi/Go proteins and inhibit adenylate cyclase (AC) activity and cyclic adenosine monophosphate (AMP) formation. Due to their widespread expression throughout the nervous system and the modulation of the relevant mechanisms that they participate in, mGluRs represent promising therapeutic targets to shape the microglia phenotype
[88][89]. Some mGluR ligands are currently under clinical development regarding the treatment of various disorders, such as X Fragile, schizophrenia, Parkinson’s disease (PD), L-dopa-induced dyskinesias, generalized anxiety disorder, and chronic pain
[90][91][92].
Microglia cell lines and primary cultures from the cerebral cortex express mGluR5 mRNA and protein
[87][93][94]. Although some evidence indicates that quiescent microglia in the healthy brain do not express mGluR5
[95], after spinal cord or head trauma, activated microglia in the vicinity of the lesion significantly express this receptor
[87][96].
Evidence demonstrating the presence of mGluR1 and mGluR5 in microglia has been obtained using selective receptor agonists and antagonists. The expression of the mGluR5a mRNA and the stimulation of calcium signaling by the mGluR1/5 agonist trans-(1S,3R)-1-amino-1,3-cyclopentane dicarboxylic acid (1S,3R-ACPD) took place in cultured microglia, indicating the expression of the mGluR5a variant in these cells
[93]. However, Whittemore et al.
[97] obtained conflicting results since they found no intracellular calcium signaling modification in microglia stimulated with 1S,3R-ACPD. The reasons for this discrepancy need to be clarified.
Other group I mGluR agonists trigger the activation of PLC in microglia cultures, which leads, again, to the release of calcium and activation of protein kinase C (PKC)
[98]. In turn, PKC activation can cause changes in rectifying potassium channel expression and they shape microglia from the ameboid to the ramified phenotype
[99]. Signaling downstream group I mGluRs include mitogen-activated protein kinase (MAPK), extracellular signal-regulated kinase 1 (ERK1), and extracellular signal-regulated kinase 2 (ERK2), which are inhibited by selective mGluR5 and mGluR1 antagonists, such as 2-Methyl-6-(phenylethynyl)pyridine (MPEP) and 7-(Hydroxyimino)cyclopropan[b]chromen-1a-carboxylate ethyl ester (CPCCOEt
[98][100]). In cultured microglia, the selective activation of mGluR5 by (RS)-2-chloro-5-hidroxyphenylglycine (CHPG), or the combination of the mixed mGluR1/5 agonist [(S)-3,5-Dihydroxyphenylglycine] (DHPG) and the selective mGluR1 antagonist CPCCOEt, attenuated the lipopolysaccharide (LPS) or INFγ-induced activation
[94][101], also reducing the accumulation of ROS, the production of TNF-α, and the levels of iNOS with consequent NO release. Moreover, PLC and PKC inhibitors and calcium chelators attenuated the anti-inflammatory events following mGluR5 activation, suggesting that the mGluR5 activation in microglia involves the Gq protein signal transduction pathway
[102].
Loane et al.
[94] showed that microglia express functional mGluR5, whose activation decreases the release of inflammatory molecules and the impact on neurotoxicity. The inhibition of NADPH oxidase mediated the protective effects of mGluR5 activation in microglia
[103], a mechanism of microglia-mediated neurotoxicity common to numerous neurodegenerative diseases
[104].
Another molecular pathway linked to the beneficial effects of mGluR5 in microglia is the brain-derived neurotrophic factor/tyrosine-protein kinase B (BDNF/TrKB) cascade
[105]. Indeed, the activation of mGluR5 by CHPG protects from oxygen–glucose deprivation (OGD) and reperfusion-induced cytotoxicity, apoptosis, the accumulation of ROS, and the release of inflammatory cytokines in the microglial BV2 cell line
[105]. mGluR5 activation also triggers the protein kinase B/glycogen synthase kinase 3β/cAMP-response element binding protein (Akt/GSK-3β/CREB) pathway, resulting in the inhibition of GSK-3β expression, increased phosphorylation of CREB, and reduced expression of inflammation-related genes in microglia cells
[106].
Although initially only mRNA coding for mGluR5 was found in microglia cells, and cultured microglia apparently do not express mGluR1, in fact, the mGluR1 subtype appears to be present in this cell population
[93], even if less expressed with respect to mGluR5
[102]. In addition to microglia, mGluR1 is also expressed by several cells within the CNS, including neurons, meningeal cells, astrocytes, and T and B cells
[93][107], and mGluR1 agonists boost T cell proliferation and promote the activation of the MAPK signaling cascade, increasing inflammation
[108]. Additionally, mGluR1 immunoreactivity has been reported in a subset of the microglia/macrophage cell lineage in human multiple sclerosis (MS) lesions
[109]. Therefore, mGluR1 antagonists might also have multiple therapeutic applications.
3.1. Role of Group I Metabotropic GLU Receptors expressed by Microglia cells in amyotrophic lateral sclerosis
Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disease characterized by the progressive loss of upper and lower motor neurons (MNs). In the early phase, symptoms are muscle weakness, followed by a gradual loss of muscle control and contraction, atrophy, and paralysis, which lead to death by respiratory failure [110]. Around 90% of cases of ALS are sporadic or unrelated to a specific etiological or hereditary genetic cause. However, 10% of patients are familial, attributable to specific transmissible genetic mutations [111][112].
Although the initial ALS approach was mainly focused on neurons, growing and unequivocal evidence indicates that non-neuronal cells also play a key role in the pathogenesis of ALS, thus contributing to the definition of ALS as a non-cell-autonomous disease [113][114][115][116][117][118]. Glial cells, such as astrocytes and microglia, can participate in the local inflammatory response with peripheral lymphocytes and macrophages. They acquire a reactive phenotype, migrate to the lesion site, proliferate, and secrete pro-inflammatory and neurotoxic mediators [119][120][121][122][123][124][125][126][127]. Glial activation modifies the expression of a wide range of soluble molecules, such as cytokines and chemokines, DAMPs, reactive nitrogen species (RNS), and ROS, giving rise to profound changes in fundamental aspects of the interactions between glia and neurons [128].
Microglia become reactive before the symptomatic phase of the disease [52][129], concomitantly with the first loss of neuromuscular junctions [130] and MNs [131].
Several findings indicate that, during ALS progression, microglia cells do not undergo a stage-dependent transition [132]; instead, they show the coexistence of the different phenotypes, thus producing a peculiar functional profile as a complex outcome of multiple regulation factors.
Differing from other pathological or traumatic conditions, the involvement of group I mGluRs in the modulation of the microglia phenotype in ALS is poorly documented. Nevertheless, mGluR1, mGluR5, and microglia cells are indeed a promising target for ALS treatment and other neurodegenerative diseases in which neuroinflammation plays a pivotal role [70][87][89][133]. In ALS patients, mGluR1 and mGluR5 mRNAs are abundantly expressed in the dorsal horn rather than in the ventral horn of the spinal cord. Of note, spared MNs express abundant mGluR5, while vulnerable MNs do not [134]. Aronica and Collaborators showed that mGluR1 and mGluR5 were highly represented in neuronal cells throughout the human spinal cord, with mGluR1 having high expression in ventral horn neurons. In contrast, intense mGluR5 immunoreactivity was observed in the dorsal horns [135]. This different CNS area- and cell-specific expression of group I mGluRs highlights an intriguing clue possibly linked to the selective vulnerability of MNs in ALS [136][137][138][139]. Regarding glial cells, only sparse astrocytes showed weak to moderate staining for mGluR1 and mGluR5 in the spinal cords of healthy patients [135][140]. In ALS patients, the mGluR1 and mGluR5 immunolabeling intensity increased in cells with an astroglia morphology in the grey and white matter. At the same time, their expression in neurons was comparable to that observed in healthy subjects [135][140].
Although the role of mGluR5 in regulating astrocyte function and their neurotoxic phenotype during ALS progression was largely investigated [141][142][143][144][145][146], there is only one paper describing mGluRs affecting microglia functionality [147]. Berger et al. examined in vitro the modulation of mGluRs expressed by microglial cells in two distinct models of inflammatory conditions, including microglia cell cultures obtained from rats expressing the SOD1G93A ALS-linked mutation. As expected, SOD1G93A microglia were characterized by increased neuroinflammation and enhanced reactivity. The results showed that the mGluR5 mRNA was upregulated in microglia cell cultures prepared from the brains of neonatal SOD1G93A rats. Interestingly, the exposure to LPS, mimicking an inflammatory environment, increased mGluR3 and decreased mGluR5 gene expression in both SOD1G93A and wtSOD1 microglia [147]. This evidence indicates that an inflammatory environment may trigger the opposite regulation of mGluR subtype gene expression. These events seem particularly robust in SOD1G93A microglia cultures. Thus, it will be crucial to consider the possible cell-specific receptor expression and localization when con-sidering these therapeutic targets in ALS.
Using the SOD1G93A mouse model, it has been first demonstrated that the expression and function of mGluR1 and mGluR5 were enhanced at glutamatergic synapses in the spinal cord at the early pre-symptomatic and late symptomatic stages of the disease [75][76]. These alterations could further exacerbate the excessive glutamatergic neurotransmission previously demonstrated in the spinal cords of SOD1G93A mice [148][149][150]. In subsequent studies, genetically halving mGluR1 and mGluR5, or ablating mGluR5, significantly ameliorated disease progression and survival probability in SODG93A mice [151][152][153], and produced a reduction in astrogliosis and microgliosis, always accompanied by positive outcomes for the ALS phenotype. In vivo, the beneficial effects can hardly be ascribed to a specific cell subtype; however, researchers postulated that the observed modulation of reactive astrocytes and microglia could represent a potential contribution to the improved MN survival [151][152][153]. Then, the effect of the chronic oral administration of the mGluR5 NAM CTEP [154]. Differing from the mGluR5 genetic ablation, the pharmacological treatment was started after symptom onset and maintained until the late symptomatic stage of the disease [154]. CTEP dose-dependently ameliorated the survival and clinical course in SOD1G93A mice. Of relevance, paralleling the genetic studies, chronic treatment with CTEP also reduced astrogliosis and microglia proliferation in the spinal cords of SOD1G93A mice, possibly contributing to the amelioration of the extracellular noxious milieu toward MNs, thus in turn reducing the disease severity.
Considering the dual role of microglia during ALS progression and the fact that blocking mGluR5 before or after disease onset, by genetic or pharmacological strategies, respectively, always ameliorated disease progression and reduced glial reactivity, uncertainty about the mixed effects of dampening mGluR5 in ALS arises. The negative modulation of group I mGluRs should differently affect astrocyte or microglia cells early in the pathology, when microglia probably possess an anti-inflammatory phenotype and astrocytes should start to be reactive, with respect to the late symptomatic stages, when both astrocytes and microglia are detrimental for MNs.
The studies mentioned above, including the microglia-specific effects of group I mGluR modulation, are worth considering in the potential therapeutic application of mGluR5-targeted drugs to be exploited for ALS and other neurodegenerative diseases characterized by glial activation and neuroinflammatory features. Due to the subtle modifications that the microglia phenotype may undergo during specific ALS stages, and the uncertain role played by mGluRs, the only way to shed light on this complex scenario would be to expand the studies by exploiting more powerful and “omic” approaches, including transcriptomics, proteomics, metabolomics, and epigenomics, besides those adopted till now.
The pharmacological or genetic modulation of group I mGluRs expressed by microglia cells represents an attractive multipotential therapeutic strategy for acute traumatic and chronic neurodegenerative disorders. Although the literature has frequently investigated the roles of mGluRs and related therapeutic approaches focusing on neurons, other cell types, including astrocytes and microglia, express these receptors. Many neuroprotective strategies aimed at modulating the aberrant reactive state of these cells have highlighted microglia as a new target to improve clinical outcomes in different pathological conditions, particularly neurodegenerative diseases characterized by prominent neuroinflammatory hallmarks. Group I mGluR modulation reduces inflammation, excitotoxicity, necroptosis, and apoptotic cell death.
Further collection of preclinical and clinical evidence will be essential to optimize group I mGluRs as multipotential targets to modulate the complex balance of microglia phenotypes in CNS disorders.