Receptor-independent actions of cannabinoids on cell membranes: Focus on endocannabinoids

Abstract

Cannabinoids are a structurally diverse group of mostly lipophilic molecules that bind to cannabinoid receptors. In fact, endogenous cannabinoids (endocannabinoids) are a class of signaling lipids consisting of amides and esters of long-chain polyunsaturated fatty acids. They are synthesized from lipid precursors in plasma membranes via Ca²⁺ or G-protein-dependent processes and exhibit cannabinoid-like actions by binding to cannabinoid receptors. However, endocannabinoids can produce effects that are not mediated by these receptors. In pharmacologically relevant concentrations, endocannabinoids modulate the functional properties of voltage-gated ion channels including Ca²⁺ channels, Na⁺ channels, various types of K⁺ channels, and ligand-gated ion channels such as serotonin type 3, nicotinic acetylcholine, and glycine receptors. In addition, modulatory effects of endocannabinoids on other ion-transporting membrane proteins such as transient potential receptor-class channels, gap junctions and transporters for neurotransmitters have also been demonstrated. Furthermore, functional properties of G-protein-coupled receptors for different types of neurotransmitters and neuropeptides are altered by direct actions of endocannabinoids. Although the mechanisms of these effects are currently not clear, it is likely that these direct actions of endocannabinoids are due to their lipophilic structures. These findings indicate that additional molecular targets for endocannabinoids exist and that these targets may represent novel sites for cannabinoids to alter either the excitability of the neurons or the response of the neuronal systems. This review focuses on the results of recent studies indicating that beyond their receptor-mediated effects, endocannabinoids alter the functions of ion channels and other integral membrane proteins directly.

Introduction

In biological systems, our understanding of cannabinoid actions has advanced significantly by a series of recent discoveries. First, the CB₁ and CB₂ receptors for Δ⁹-tetrahydrocannabinol (THC), the principal psychotropic component of marijuana, were identified, cloned and expressed functionally, heralding the beginning of the molecular era in cannabinoid research (Matsuda et al., 1990, Munro et al., 1993, Matsuda, 1997, Onaivi et al., 2002). Both receptors belong to the G-protein-coupled receptor family and couple to adenylyl cyclase as well as other second messenger systems such as protein kinase C (PKC) and Ca²⁺ (Howlett & Mukhopadhyay, 2000, McAllister & Glass, 2002). CB₁ receptors are located predominantly on the presynaptic terminals of central and peripheral nerves and inhibit transmitter release through their coupling to Ca²⁺ and K⁺ channels. CB₂ receptors, on the other hand, are found mainly on peripheral tissues and the immune system (Howlett et al., 2002).

Another major development was the discovery of endogenous ligands or endocannabinoids for cannabinoid receptors (Devane et al., 1992, Mechoulam et al., 1998). The best-known and most thoroughly investigated endocannabinoids are derivatives of arachidonic acid (AA), namely anandamide (N-arachidonoylethanolamine, AEA) and 2-arachidonoylglycerol (2-AG) (Devane et al., 1992, Mechoulam et al., 1998, De Petrocellis et al., 2004). Finally, in the mid-nineties, relatively specific antagonists such as SR141716A and SR144528 were developed for CB₁ and CB₂ receptors, respectively (for a review, see Shire et al., 1999). Cannabinoid-receptor mediated actions of endocannabinoids, natural and synthetic cannabinoids on both central and peripheral nervous systems, as well as on the cardiovascular system, are well documented in comprehensive reviews (Kunos et al., 2000, Wilson & Nicoll, 2002, Freund et al., 2003, Gerdeman et al., 2003, Pacher et al., 2005).

Following the discovery of AEA, it was found that some of the pharmacological effects of AEA are not sensitive to SR141716A treatments (for a review, see Wiley & Martin, 2002). Recent studies carried out in CB₁ receptor knock-out mice showed that despite the lack of CB₁ receptors, AEA still exerted cannabimimetic-like activity in the tetrad test and other behavioral tests (Di Marzo et al., 2000, Baskfield et al., 2004). Similarly, SR141716A did not antagonize the effects of mAEA in the open-field test (Jarbe et al., 2003). In vitro studies have also found that AEA continues to stimulate [³H]GTP-γ-S binding in brain membranes from CB₁ knock-out mice and this stimulatory effect of AEA was insensitive to both CB₁ and CB₂ receptor antagonism (Di Marzo et al., 2000). In light of recent studies suggesting that several antipsychotics including clozapine, haloperidol, thioridazine, and chlorpromazine, known to act through their effects on 5-hydroxytryptamine (5-HT) and/or dopamine (DA)-activated G-protein coupled receptors (GPCR), also mimic the effect of cannabinoids in the widely used tetrad test in mice (Wiley & Martin, 2003); it is likely that other GPCR may also be involved in some of the actions of endocannabinoids observed in cannabinoid-receptor knock-out mice.

Some endocannabinoid actions that are mediated by G-proteins or are sensitive to pertussis toxin (PTX) but are not blocked by the specific cannabinoid receptor antagonists have been proposed to be mediated by a third cannabinoid receptor distinct from CB₁ and CB₂ (for reviews, see Wiley & Martin, 2002, Begg et al., 2005). In addition to activation of GPCR, earlier studies demonstrated that AEA interacts directly with binding of radioligands to L-type Ca²⁺ channels (Johnson et al., 1993, Shiamsue et al., 1996) and modulates the function of the serotonin type 3 (5-HT₃) receptors in a cannabinoid-receptor independent manner (Fan, 1995, Oz et al., 1995).

Radioligand displacement studies indicate that the K_i values of THC for CB₁ and CB₂ receptors are in the ranges of 35–80 nM and 4–75 nM, respectively. The K_i values for AEA and 2-AG are in the ranges of 60–540 nM and 60–470 nM for CB₁ receptors and in the ranges of 370 nM–1.9 μM and 150 nM–1.4 μM for CB₂ receptors, respectively (for a review, see Howlett et al., 2002). It is important to note that the psychotropic effects of THC occur at concentrations in the low to mid nM-range (Huestis, 2002). Thus, the receptor-independent effects of THC that occur at low micromolar concentrations were thought to be relevant only to THC toxicity. However, due to their high lipophilicity (Roth & Williams, 1979, Tian et al., 2005), cannabinoids are expected to reside nearly exclusively within the membrane bilayer; cannabinoid concentrations in lipid phases might be much higher than those in aqueous solutions. Since the time to reach equilibrium for the cannabinoids within membrane bilayer is unknown, equilibrium may not be achieved within the time course of in vitro experiments. Under these conditions, in vitro experiments with high concentrations of THC may better mimic in vivo equilibrium conditions. Furthermore, plasma concentrations of THC following cannabis smoking peak at 200–600 nM within 10 min and the onset of subjective effects are similarly rapid (Huestis, 2002). Thus, in vitro conditions testing relatively high concentrations (high nanomolar to low micromolar) of CB receptor ligands over short time periods (for 5–10 min) may mimic THC's pharmacological actions.

Endocannabinoids are produced intracellularly and interact first with plasma membranes before they are detected in extracellular compartments. Currently, there is no information available on the intracellular concentrations of endocannabinoids, though this parameter, to a large extent, is under the control of the experimenter during in vivo or in vitro studies using paradigms not utilizing endocannabinoid synthesis.

The purpose of this review is to point out receptor-independent effects of cannabinoids that are widespread and largely ignored. Although some of the results with AEA were reviewed previously (Di Marzo et al., 2002), this review attempts to compile an exhaustive volume of data on receptor-independent effects of cannabinoids and is intended as a guide for the researchers in the field and not as a conceptual overview. While some of the results are highlighted in Table 1, Table 2, Table 3, Table 4, details of the individual studies cannot be condensed into table format and are therefore given in the text. These tables are complementary to text information and are not designed to summarize related chapters. In this review, the actions of endocannabinoids are termed receptor-independent or direct when the effects of endocannabinoids are observed in a system that does not contain any of the known cannabinoid receptors (e.g. Xenopus oocytes, neurons from superior cervical ganglion or a particular cell line) or the effects of endocannabinoids are not sensitive to specific antagonists for the known cannabinoid receptors.

In addition to the eicosanoid-based molecules such as AEA and 2-AG, there are 3 different classes of cannabinoid receptor ligands that are employed currently in cannabinoid pharmacology. These consist of the classical agonist phytochemical THC, the non-classical cannabinoids, typified by the agonist CP55,940 and the aminoalkylindoles such as WIN55,212-2 (Pertwee, 1997). Although this review intends to focus mainly on the direct effects of endocannabinoids on ion transport through the cell membrane, studies of relevant receptor-independent effects of THC and other cannabinoid receptor ligands were also included. In addition, the results from endocannabinoid-related compounds including noladin ether (Howlett et al., 2002, De Petrocellis et al., 2004), virodhamine, (Porter et al., 2002, De Petrocellis et al., 2004), oleamide (Fowler, 2004, Leggett et al., 2004), and metabolites of endocannabinoids such as AA are also included.