Indeed, since the demonstration in earlier studies that extracted phytocannabinoids had some potential in neuroprotection, and following the discovery of an endogenous cannabinoid system that may play a similar role, the use of cannabinoids as alternative medicines has gathered a growing research interest. A large amount of dedicated research has demonstrated the neuroprotective properties of various cannabinoids, whether endogenous or exogenous, under different conditions, and both in vitro and in vivo models. In most cases, the use of cannabinoids and the modulation of the ECS were shown to limit the effects of different insults that could lead to neuronal damage and neuronal loss [5, 18, 20–23].

The present mini-review will describe briefly the main elements of the ECS, its involvement in neuroprotection, the main mechanisms responsible for neurodegeneration, and how the cannabinoids may act at different levels to limit the impact of these critical mechanisms.

Cannabis sativa has been used for centuries by humans [24, 25], but its extracts have been studied more intensively in the last six to seven decades, after the identification of the two main constituents, the psychoactive Δ9-trans-tetrahydrocannabinol (THC) [26], and the non-psychoactive cannabidiol (CBD) ([27]; reviewed in [28, 29]). Currently, the number of natural compounds from Cannabis sativa is close to 565, among which, 120 cannabinoids have been isolated and classified [24, 25, 29]. The discovery of THC and CBD was followed by the identification of the receptors that mediate their actions: the cannabinoid receptors, and then followed by the identification of endogenous ligands for these receptors, called eCBs, within the physiological ECS [29–35].

The ECS comprises the cannabinoid receptors, their ligands eCBs, the enzymes responsible for eCBs synthesis and degradation, as well as a not well-known reuptake system. The present description is relatively succinct and for further detailed reviews please see the following [5, 21, 23, 29, 33, 35, 36].

Two major receptors, the cannabinoid CB1 and CB2 receptors (CB1Rs, CB2Rs), were identified, cloned, and extensively studied. Both receptors, CB1R and CB2R belong to the G protein-coupled receptor (GPCR) superfamily and couple to Gi/o to inhibit adenylyl cyclase and protein kinase A (PKA) (reviewed in [21, 28, 29, 32, 35]). They also modulate the mitogen-activated protein kinase (MAPK) pathway by regulating different kinases such as extracellular signal-regulated kinase (ERK), p38 kinase, and c-Jun N-terminal kinases [21, 28]. CB1R is highly expressed in the brain and represents the main target of the eCBs. CB1R is expressed at presynaptic terminals and their activation leads to the inhibition of neurotransmitter release resulting in the inhibition of excitatory and inhibitory neurotransmission [21, 28, 37, 38]. There is also evidence for CB1R expression at postsynaptic locations where its activation leads to the modulation of ion channels, notably the N-methyl-D-aspartate (NMDA) receptors [21, 39]. The CB2R is expressed mostly by immune cells and seems to be involved in the modulation of the immune system principally. In the brain, CB2R expression increases during neuroinflammation but is very low under normal physiological conditions ([40]; reviewed in [21, 41]). Both CB1R and CB2R are expressed by glial cells, astrocytes and microglia [42]. The CB1R expression has also been shown in oligodendrocytes and human vascular endothelial cells of the blood-brain barrier (BBB) [21, 43, 44]. Although CB2R expression is generally very low leading to some controversy around their existence in certain types of cells such as astrocytes, it has been shown that this receptor is increased in astrocytes and microglia after induced neurotoxicity and during neuroinflammation (reviewed in [42]).

Outside the CB1Rs and CB2Rs, multiple other receptors were shown to be involved in the transduction of cannabinoid signaling (reviewed in [28, 29, 35, 38, 45]). That is the case of nuclear peroxisome proliferator-activated receptors (PPARs) [14, 46], and some categories of transient potential receptors (TRP) such as the vanilloid type1 channel (TRPV1) [47, 48]. Also reported as potential receptors for cannabinoids are serotonin 1A receptor (5HT1A) [49], opioid receptors mu and delta (µ-OR and δ-OR) [28, 38], as well as other receptors including some orphan receptors such as GPR55, GPR119 and GPR18 [38, 50–52], and multiple receptor complexes (heteromers) formed between CB1Rs or CB2Rs and other GPCRs or non-GPCRs receptors [28, 29, 33, 46, 50, 53–56].

Eicosanoids N-arachidonoyl ethanolamide (anandamide, AEA) and 2-arachidonoyl glycerol (2-AG) represent the primary eCBs and are the most studied and characterized [34, 35, 38, 57–59]. Both AEA and 2-AG act as endogenous ligands for CB1Rs and CB2Rs [28, 34, 35]. There is however a long list of endogenous congeners capable of modulating CB1Rs or CB2Rs or related receptors, among them 2-AG ether [59], virodhamine, with antagonistic activity at the CB1R [60], and oleamide [61]. 2-AG is a full agonist for both cannabinoid receptors, whereas AEA acts as a partial agonist, and both eCBs show higher affinity for the CB1R in comparison to the CB2R [28, 54, 62].

It is well known that these lipidic molecules are synthesized on demand by the actions of specific lipases in response to increased intracellular Ca²⁺ levels and are immediately released. Both AEA and 2-AG are produced by post-synaptic neurons by the action of different enzymes and act via retrograde transsynaptic action to activate CB1R. AEA is produced by the action of two main enzymes, Ca²⁺-dependent N-acyltransferase and N-acyl-phosphatidylethanolamine-hydrolyzing phospholipase D (NAPE-PLD) from membrane glycerophospholipids and its levels are regulated by its hydrolysis by fatty acid amide hydrolase (FAAH) (reviewed in [5, 21, 29, 45]). 2-AG is also produced in post-synaptic neurons on demand from diacylglycerol (DAG) in a two-step process involving the action of phospholipase C, and diacylglycerol lipase (DAGL) ([58]; reviewed in [35]). Other studies show that 2-AG is also synthesized through two additional major pathways: from 2-acyl lysophosphatidic acid (LPA) by 2-LPA phosphatase, and from 2-acyl lyso-phosphatidylinositol (LPI) by lyso-phospholipase C (reviewed in [5, 35, 45]). An important enzyme, the monoacylglycerol lipase (MAGL), controls 2-AG levels by assuming its degradation. The degradation of 2-AG occurs in the pre-synaptic neurons after release from post-synaptic neurons, whereas the degradation of AEA occurs in the post-synaptic neurons, where it is synthesized [5, 21]. In addition, other enzymes, such as cyclooxygenase-2 (COX2) and lipoxygenases can metabolize AEA or 2-AG into other bioactive derivative compounds, some of which may also behave as endocannabinoid-like molecules [21, 29, 63, 64]. Some of these derivatives may signal either through the CB1Rs or CB2Rs or through non-cannabinoid targets [21, 29, 63, 64]. Also important in the mechanism of eCB action is their possible removal from the synaptic space by a reuptake mechanism followed by hydrolysis [65]. There is a need for more work in this regard to identify these potential specific endocannabinoid transporters [65].

There are multiple mechanisms by which activation of the cannabinoid system by either exogenous cannabinoids or mobilization of eCBs exerts neuroprotective effects. One key mechanism is by reducing neuroinflammation, a major contributor to neuronal damage and cell loss [35, 124]. Secondly, the cannabinoid system has been shown to modulate oxidative stress, another key factor in the pathogenesis of various degenerative neurological disorders. Further, the regulation of excitotoxic and apoptotic pathways, which play crucial roles in neurodegenerative diseases, is a third mechanism by which the cannabinoid system exerts its neuroprotective effects, as will be developed in the next paragraphs.

As mentioned, the connection between neuroinflammation, excitotoxicity, oxidative stress, and neurodegeneration is rather multifaceted and interconnected (Figure 1). To simplify, neurodegeneration could be considered as resulting from the overlapping effects of these three mechanisms (see for more detailed reviews [5, 10, 11, 17, 84, 125–130]). Neuroinflammation is regarded as the brain’s response to injuries and disease processes, which may involve the activation of glial cells, microglia and astrocytes, compromising the integrity of the BBB and its functions, increasing its permeability, with infiltration of immune cells. While the first response is protective, its persistence can lead to the release of pro-inflammatory cytokines such as tumor necrosis factor-α (TNF-α) and interleukin-1β (IL-1β), which proliferate the inflammation and other molecules such as chemokines (e.g., COX2) that recruit additional immune cells, that prove to be harmful to and destroy the neurons. Elevated glutamate can lead to excitotoxicity, with excessive activation of glutamate receptors notably of extra-synaptic NMDA receptors [10, 17, 131] causing excessive calcium influx into neurons. This may trigger a cascade of events including increased generation of ROS, which are unstable, free radical, highly reactive molecules, generated within mitochondria, leading to mitochondrial dysfunction, contributing to oxidative stress and neuronal damage. An imbalance between the production of ROS and the ability to enzymatically detoxify these molecules in the brain characterizes oxidative stress [10, 17]. Neuroinflammation can exacerbate oxidative stress by increasing the production of ROS, which further damages cellular components including DNA, RNA, proteins, and lipids. In summary, the combination and interconnectedness of these factors create a vicious cycle where each one exacerbates the others, resulting in progressive neuronal damage and cell death that ultimately leads to neurodegeneration (Figure 1). As will be detailed bellow, several studies have demonstrated the ability of cannabinoids to mitigate neuronal damage and death in various pathological conditions by modulating these main mechanisms.

While describing the utility of cannabinoids and the ECS in oxidative stress, the strong interactions among neuroinflammation, excitotoxicity, and oxidative stress are paramount and have been reported in neurodegenerative diseases such as AD, PD, and Huntington’s disease [137, 180, 191–195]. ROS are regular products of the mitochondrial respiratory chain, and their regulation is important for cell survival. However, both excitotoxicity and the activation of microglia lead to an increase in ROS and reactive nitrogen species (RNS) [9, 11, 21, 127, 128, 196]. The accumulation of ROS and RNS contributes highly to oxidative stress, which has many detrimental consequences such as degradation of DNA, carbohydrates and lipid peroxidation, and mitochondrial functions [196–201]. Though some of these molecules do not escape the cell membrane, others that are more stable (e.g., H₂O₂) can traverse the cell membrane and spread to adjacent tissues. Multiple studies suggest that oxidative stress plays a critical role in the progression of different diseases including CNS disorders and neurodegenerative diseases [196–201], although the use of antioxidants showed limited or, in some cases, negative results [202, 203].

Interestingly, activated oxidative stress pathways were shown to impair ECS-mediated signaling, whereas activation of CB1R as well as the upregulation of brain CB2R reduce oxidative stress in the brain. This results in reduced neuroinflammation and consequently, attenuated neuronal and tissue damage. Multiple studies have shown that activation of the ECS has neuroprotective effects at different levels. For instance, the inhibition of FAAH by URB597, and the consequent increase in the endogenous levels of AEA, which prevented excitotoxic damage, attenuated motor and biochemical (lipid peroxidation and protein carbonylation) alterations in rats, while preserving the structural integrity of the striatum and inhibiting the neuronal loss [204]. This study among many others is in favor of the idea that pharmacological manipulation of the ECS plays a neuroprotective role against excitotoxic insults and the resulting oxidative stress as shown by the biochemical results in the central nervous system [204]. In another study, the cannabinoid trans-caryophyllene has been shown to increase neuronal viability through inhibition of mitochondrial depolarization and oxidative stress, and by increasing the expression of BDNF in rat neuronal-glial cultures [205]. Many studies showed that CB2R expression is increased in microglia and other immune cells and may participate in the reduction of oxidative stress [184]. This is only a limited window into a large volume of research showing that the ECS and cannabinoids are involved in the neuroprotective effects against oxidative stress, but are not antioxidants by themselves.