Endocannabinoid system

Overview

The endocannabinoid system (ECS) tightly regulates the probability of neurotransmitter release in a host of neuronal tissues like the hippocampus, amygdala, basal ganglia, and cerebellum [1] [2] [3] Data from immunologists, developmental biologists, and embryologists show that the ECS is also intimately involved in crosstalk between lymphocytes and synchronizing the timing between uterine receptivity and embryo implantation [4][5]. In the central nervous system, however, the ECS is involved in a specific type of retrograde signaling in which endogenous cannabinoids (endocannabinoids) are released from the post-synaptic neuron and bind to nearby cannabinoid receptor 1 (CB1) -expressing presynapses; this binding elicits a sharp, but temporary, suppression of presynaptic neurotransmitter release [6]. In this way, the post-synaptic neuron can regulate its own excitability by adjusting synaptic inputs via endocannabinoids that act as retrograde messengers. The endocannabinoid system is thought to be important in long-term potentiation and memory[7] [8], motor learning [9], and synaptic plasticity [10].

Introduction

Endocannabinoid synthesis & release

In standard neurotransmission, the pre-synaptic neuron releases neurotransmitter into the synaptic cleft which binds to cognate receptors expressed on the post-synaptic neuron. Upon binding, the neuron depolarizes. This depolarization facilitates the influx of calcium into the neuron; this increase in calcium activates an enzyme called transacylase which catalyzes the first step of endocannabinoid biosynthesis by converting phosphatidylethanolamine, a membrane-resident phospholipid into N-acyl-phosphatidylethanolamine (NAPE). Experiments have shown that both phospholipase C and phospholipase D cleave NAPE to yield anandamide and 2-AG, respectively (Okamoto et al., 2004; Liu et al, 2006). In phospholipase D (PLD) knockouts, the PLD-mediated cleavage of NAPE is reduced, not abolished, in low calcium concentrations, suggesting multiple, distinct pathways are involved in 2-AG biosynthesis (Leung et al., 2006). Once released into the extracellular space by a putative endocannabinoid transporter, messengers are vulnerable to glial inactivation. Endocannabinoids are uptaken via a putative transporter and degraded by fatty acid amide hydrolase (FAAH) which cleaves anandamide and 2-AG to arachidonic acid & ethaloamine and arachidonic acid & glycerol, respectively (reviewed in Pazos et al., 2005). While arachidonic acid is a substrate for leukotriene and prostaglandin synthesis, it is unclear whether this degradative byproduct has novel functions in the CNS (Yamaguchi et al., 2001; Brock, T., 2005). Emerging data in the field also points to FAAH being expressed in the postsynaptic neuron, suggesting it also contributes to the clearance and inactivation of anandamide and 2-AG by endocannabinoid reuptake.

Endocannabinoid binding & signal transduction

While there have been some exciting papers that have linked concurrent stimulation of dopamine and CB1 receptors to an acute rise in cAMP production, it is accepted that CB1 activation causes an inhibition of cyclic adenosine monophosphate (or cAMP) when activated alone. This inhibition of cAMP is followed by phosphorylation and subsequent activation of not only a suite (p38/p42/p44) of MAP kinases but also the PI3/PKB and MEK/ERK pathway (Galve-Roperh et al., 2002; Davis et al., 2005; Jones et al., 2005; Graham et al., 2006). Results from rat hippocampal gene chip data after acute administration of tetrahydrocannabinol showed an increase in the expression of myelin lipid protein, endoplasmic proteins, cytochrome oxidase, and two cell adhesion molecules: NCAM, and SC1; decreases in expression were seen in both calmodulin and ribosomal RNAs (Kittler et al., 2000). In addition, CB1 activation has been demonstrated to increase the activity of transcription factors like c-Fos and Krox-24 (Graham et al., 2006).

Endocannabinoid binding & alterations in neuronal excitability

The molecular mechanisms of CB1-mediated changes to the membrane voltage have also been studied in detail. CB1 agonists reduce calcium influx by blocking the activity of voltage-dependent N-, P/Q- and L-type calcium channels [11] [12] In addition to acting on calcium channels, Gi/o and Gs, subunits of G protein-coupled receptors, activation has also been shown to modulate potassium channel activity. Recent studies have found that CB1 activation facilitates GIRK, a potassium channel belonging to the Kir3 family (Guo and Ikeda, 2004)[13]. Corroborating Guo and Ikeda, Binzen et al. performed a series of immunohistochemistry experiments that demonstrated CB1 co-localized with GIRK and Kv1.4 potassium channels, suggesting that these two may interact in physiological contexts [14]. In the central nervous system, CB1 receptors, for the most part, influence neuronal excitability indirectly, by reducing the impact of incoming synaptic input [15]. This mechanism ("presynaptic inhibition") is believed to occur when a neuron ("postsynaptic") releases endocannabinoids in a retrograde fashion, binding to CB1 receptors expressed on nerve terminals of an input neuron ("presynaptic"). CB1 receptors then reduce the amount of neurotransmitter released, so that subsequent input from the presynaptic neuron has less of an impact on the postsynaptic neuron. It is likely that presynaptic inhibition uses many of the same ion channel mechanisms listed above, although recent evidence has shown that CB1 receptors can also regulate neurotransmitter release by a non-ion channel mechanism, i.e. through Gi/o mediated inhibition of adenylyl cyclase and Protein Kinase A [16] Still, direct effects of CB1 receptors on membrane excitability have been reported, and strongly impact the firing of cortical neurons [17]. In a series of behavioral experiments, Palazzo et al. demonstrated that NMDA, an ionotropic glutamate receptor, and the metabotropic Glutamate receptors (mGluRs) work in concert with CB1 to induce analgesia in mice, although the mechanism underlying this effect is unclear. Together, these findings suggest that CB1 influences neuronal excitability by a variety of mechanisms, and these effects are relevant to perception and behavior.

CB1 -/- phenotype

Neuroscientists often utilize transgenic CB1 knockout mice (i.e. the mice have had the gene encoding the CB1 receptor deleted or removed) to discern novel roles for the ECS. While CB1 knockout mice are healthy and live into adulthood, there are some differences among mice without CB1 and wild-type (i.e "normal" mice with the receptor intact); When under a high-fat diet CB1 knockout mice tend to about sixty percent leaner and slightly less hungrier than wildtype [18]. Compared to wildtype, CB1 knockout mice exhibit severe deficits in motor learning, memory retrieval, and increased difficulty in completing the Morris water maze [19] [20] [21]. There is also evidence indicating that these knockout animals have an increased incidence and severity of stroke and seizure (Parmentier et al., 2002; Marsicano et al., 2003).

ECS changes induced by cannabis consumption

Memory

Mice treated with tetrahydrocannabinol show suppression of long-term potentiation in the hippocampus - a process that is essential for the formation and storage of long-term memory [22]. These results concur with anecdotal evidence suggesting that smoked preparations of Cannabis Sativa attentuates short-term memory [23]. Indeed, mice without the CB1 receptor show enhanced memory and long-term potentiation indicating that the endocannabinoid system may play a pivotal role in the extinction of old memories. Interestingly, recent research reported in a 2005 Journal Of Clinical Investigation article [24] indicate that the high-dose treatment of rats with the synthetic cannabinoid, HU-210 over a period of a few weeks resulted in stimulation of neural growth in the rats' hippocampus region, a part of the limbic system playing a part in the formation of declarative and spatial memories.

Appetite

Those who use cannabis are familiar with its appetite-enhancing effects. Emerging data suggests that THC act via CB1 receptors on hypothalamic nuclei, thus directly increasing appetite[25][26]. It is thought that hypothalamic neurons tonically produce endocannabinoids that work to tightly regulate hunger. Interestingly, the amount of endocannabinoids produced is inversely correlated with the amount of leptin in the blood[27]. For example, mice without leptin not only become massively obese but have higher-than-normal levels of hypothalamic endocannabinoids [28]. Similarly, when these mice were treated with an endocannabinoid antagonist, such as Rimonabant, food intake was reduced[29]. When the [CB1] receptor is knocked out in mice, these animals tend to be leaner and less hungry than wild-type (or "normal") mice While there is need for more research, these results (and others) suggest that exogenous cannabinoids (as from smoking marijuana) in the hypothalamus activates a pathway responsible for food-seeking behavior [30].

ECS and multiple sclerosis

Historical records from ancient China and Greece suggest that preparations of Cannabis Sativa were commonly prescribed to ameloriate multiple sclerosis-like symptoms such as tremors and muscle pain; unfortunately, however, treatment with marinol hasn’t shown the same efficacy as inhaled Cannabis [31][32]. Due to the illegality of Cannabis and rising incidence of multiple sclerosis patients who self-medicate with the drug, there has been much interest in exploiting the endocannabinoid system in the cerebellum to provide a legal and effective relief (reviewed in Pertwee, 2001). In mouse models of multiple sclerosis, there is a profound reduction and reorganization of CB1 receptors in the cerebellum (Cabranes et al., 2006). Serial sections of cerebellar tissue subjected to immunohistochemistry revealed that this aberrant expression occurred during the relapse phase but returned to normal during the remitting phase of the disease (Cabranes et al., 2006). There is recent data indicating that CB1 agonists promote the in vitro survival of oligodendrocytes, specialized support glia that are involved in axonal myelination, in the absence of growth and trophic factors; in addition, these agonist have been shown to promote mRNA expression of myelin lipid protein. (Kittler et al., 2000; Mollna-Holgado et al., 2002). Taken together, these studies point to the exciting possibility that cannabinoid treatment may not only be able to attenuate the symptoms of multiple sclerosis but also improve oligodendrocyte function (reviewed in Pertwee, 2001; Mollna-Holgado et al., 2002).

Role in human female reproduction

The developing embryo expresses cannabinoid receptors early in development that are responsive to anandamide which is secreted in the uterus. This signaling is important in regulating the timing of embryonic implantation and uterine receptivity. In mice, it has been shown that anandamide modulates the probability of implantation to the uterine wall. For example, in humans, the likelihood of miscarriage increases if uterine anandamide levels are too high or low [33]. These results suggest that proper intake of exogenous cannabinoids (e.g. marijuana) can decrease the likelihood for pregnancy[34][35].

Role in hippocampal neurogenesis

In the adult brain, the endocannabinoid system facilitates neurogenesis ("birth of new neurons") of hippocampal granule cells[36][37]. In the subgranular zone of the dentate gyrus, multipotent neural progenitors (NP) give rise to daughter cells that, over the course of several weeks, mature into granule cells whose axons project to and synapse onto dendrites on the CA3 region [38]. Very recent data suggests that the maturing granule cells are dependent on a reelin, a molecular guidance cue, for proper migration through the dentate gyrus (Gong et al., 2007). NPs in the hippocampus have been shown to possess FAAH and express CB1 and utilize 2-AG (Aguado et al., 2005). Intriguingly, CB1 activation by endogenous or exogenous promote NP proliferation and differentiation; this activation is absent in CB1 knockouts and abolished in the presence of antagonist (Aguada et al., 2005; Jiang et al., 2005).