Role of ionotropic cannabinoid receptors in peripheral antinociception and antihyperalgesia

Abstract

Despite the wealth of information on cannabinoid-induced peripheral antihyperalgesic and antinociceptive effects in many pain models, the molecular mechanism(s) for these actions remains unknown. Although metabotropic cannabinoid receptors have important roles in many pharmacological actions of cannabinoids, recent studies have led to the recognition of a family of at least five ionotropic cannabinoid receptors (ICRs). The known ICRs are members of the family of transient receptor potential (TRP) channels and include TRPV1, TRPV2, TRPV4, TRPM8 and TRPA1. Cannabinoid activation of ICRs can result in desensitization of the TRPA1 and TRPV1 channel activities, inhibition of nociceptors and antihyperalgesia and antinociception in certain pain models. Thus, cannabinoids activate both metabotropic and ionotropic mechanisms to produce peripheral analgesic effects. Here, we provide an overview of the pharmacology of TRP channels as ICRs.

Introduction

Although cannabinoids have been used for millennia for treating pain and other symptoms, their mechanisms of action remain obscure. With the heralded identification of multiple G-protein-coupled receptors (GPCRs) mediating cannabinoid effects nearly two decades ago, the mystery of cannabinoid pharmacology was thought to be solved [1,2]. However, continued studies demonstrate that many cannabinoid effects cannot be attributed solely to the CB₁ and CB₂ metabotropic GPCRs. For example, several important cannabinoid actions, such as peripherally mediated antihyperalgesia (see Glossary), persist in CB₁- and CB₂-gene knockout animals [3,4]. In addition, the dual generation of both neuroprotective and neurotoxic effects of cannabinoids [5,6] cannot be explained by the presence of known metabotropic cannabinoid receptors (MCRs). For example, the CB₁ antagonist rimonabat is neuroprotective in a model of cerebral ischemia, and these effects are independent of actions on MCRs [6]. These and other findings have led to the suggestion of a ‘CBx’ – an unknown cannabinoid receptor [7]. Here, we focus on the role of transient receptor potential (TRP) channels serving as ionotropic cannabinoid receptors (ICRs) and contributing to the pharmacology of cannabinoids, with an emphasis on their key role in the pain system.

The detection of tissue injury is an essential function of somatosensation and is mediated primarily by a specialized class of peripheral nociceptive afferent neurons. Although tissue injury is detected via peripheral nociceptive neurons, the actual perception of pain occurs in the central nervous system (CNS) and is subject to both stimulatory and inhibitory central modulation [8]. Peripheral nociceptive neurons are classically organized by morphological attributes or conduction velocity properties, although contemporary research has focused on their expression of receptors and ion channels as a functional type of classification [9,10]. The growing recognition of receptors expressed on nociceptors has prompted research on peripheral mechanisms regulating their activity. For example, peripheral administration of cannabinoids into inflamed tissue inhibits nociceptor activity and produces a peripherally mediated antihyperalgesia [11]. Additional analgesic mechanisms are engaged after central administration of cannabinoids. Thus, drugs can modulate peripheral or central targets to produce peripherally or centrally mediated analgesia.

The classical cannabinoid receptors: metabotropic receptors: metabotropic receptors

The first family of cannabinoid receptors identified were the metabotropic GPCRs including CB₁, CB₂, GPR55 and possibly GPR119 and peroxisome-proliferator-activated receptors (PPARs) [1,2,7,12]. The expression pattern for CB₁ (throughout the CNS and heart, gastrointestinal tract, kidney, spleen, liver, lung, testis, uterus and muscle), CB₂ (primarily immune cells and some neurons) and GPR55 (adrenal tissue, ileum, jejunum, frontal cortex and striatum) is broad, implicating a complex modulation of multiple physiologic systems by cannabinoids [13,14]. These MCRs signal primarily via G_i/o-related pathways, although coupling to G_s and G_q signaling pathways have been reported under certain experimental conditions [15]. Thus, agonist-directed trafficking is observed with these receptors as with other GPCRs. Activation of the MCRs leads to generation of the cannabinoid tetrad of behavior in addition to many other of the classical effects attributed to cannabinoid pharmacology [16]. In general, two approaches have been used experimentally to implicate an observed effect with the MCRs. First, pretreatment with a cannabinoid antagonist is predicted to block the observed effect. However, many of these compounds have inverse agonist activities that complicate interpretation, and other compounds such as AM251 might have known antagonist actions against CB₁ with only recently recognized actions against other receptors (i.e. antagonist or inverse agonist actions at GPR55) [7,14]. A second approach is the use of genetic knockout animals. Interestingly, the knockout studies implicate the existence of non-CB₁ and/or non-CB₂ cannabinoid receptors [17]. It is possible that these effects might be attributed to additional GPCRs activated by cannabinoids. However, the finding that certain cannabinoid actions persist after pretreatment with either pertussis toxin or GDPβS compounds capable of blocking GPCR functions provides strong evidence of a second family of non-MCRs [18,19]. In addition, parallel studies indicate that application of cannabinoids generates a slow inward current in the TRPV1 and TRPA1 ionotropic receptors [20,21]. Taken together, these studies have expanded the scope of cannabinoid research to focus on ionotropic receptor systems.

ICRs regulate sensory neuron activities

Emerging data from several studies has led to the recognition of a second family of cannabinoid receptors, namely the ICRs. The ICRs are either activated or antagonized by cannabinoids in a variety of cell types by nM–μM concentrations of these compounds (Table 1). To date, the known ICRs belong to the TRP family of channels. The TRP channels are a broad family of ligand-gated ion channels that generate an inward flow of cations upon activation. Studies reported to date indicate that cannabinoids gate at least five distinct ICRs (Table 1).

Table 1.

Action of cannabinoids on ionotropic cannabinoid receptors^a

Cannabinoid	Type	Action on ICR	MCR	Currentb
Anandamide	Endogenous	TRPV1 [21]	CB1; CB2	200–500 pA
		>0.3 μMc	>10 nM
NADA	Endogenous	TRPV1 [61]	CB1; CB2	300–700 pA
		>10 nM	>100 nM
5′,6'-EA	Endogenous	TRPV4 [62]	NAd	20–50 pA
		>1 μM
ACEA	Synthetic	TRPV1 [4,23]	CB1	300–700 pA
		>5 μM	>1 nM
Δ9-THC	Plant	TRPV2 [63]; TRPA1 [30]	CB1; CB2	100–200 pA
		>10 μM	>10 nM
Cannabinol	Plant	TRPV2 [63]; TRPA1 [30]	NAd	50–100 pA
		>10 μM
Cannabidiol	Plant	TRPV2 [63]; TRPA1 [30]	NAd	50–100 pA
		>10 μM
Cannabigerol	Plant	TRPM8 [46]	NAd	NA
		NA
AM404	Synthetic	TRPV1 [64,65]	AEA-trans	200–500 pA
		>1 μM	NA
WIN55212	Synthetic	TRPA1 [4]	CB1; CB2	200–300 pA
		>5 μM	>10 nM
AM1241	Synthetic	TRPA1-TRPV1e	CB2	100–200 pA
		>25 μM	>1 nM

^aAbbreviations: Δ9-THC, delta(9)-tetrahydrocannabinol; 5',6'-EA, 5',6'-epoxyeicosatrienoic acid; AEA-trans, antagonist for putative anandamide transporter; ICR, ionotropic cannabinoid receptor; MCR, metabotropic cannabinoid receptor; NA, non-applicable; NADA , N-arachidonoyl-dopamine.

^bApproximate value of current magnitudes in sensory neurons.

^cApproximate threshold concentrations of cannabinoids to activate receptors.

^dCannabinoid-like compounds do not activate CB₁ and CB₂.

^eActivation of TRPA1 and TRPV1 co-expressing cells.

The majority of ICRs are expressed in nociceptive sensory neurons, which can detect and respond to noxious mechanical, thermal and chemical stimuli [22]. Therefore, it could be predicted that activation of sensory neurons, by cannabinoid gating of inward currents generated by these ICRs, could result in nociception and, ultimately, pain perception [23–25]. However, a vast majority of behavioral studies indicate that cannabinoids do not produce nociception [19,23,26] but instead induce a peripherally mediated and efficacious antihyperalgesia and antinociception [27–29]. The interpretation of these data is complex because cannabinoids activate ionotropic receptors and generate inward currents but still produce a profound antihyperalgesia. One possible hypothesis addressing this issue is that partial activation of ICRs does not necessarily generate excitation (i.e. action potential) of nociceptors. From this perspective, it is interesting to note that cannabinoids are not full agonists for TRP channels [4,23,30]. Indeed, cannabinoids typically evoke a slow generation of small inward currents and Ca²⁺ accumulation [4,19,20,30] (Table 1). As a result, cannabinoid-gated responses might not reach the threshold levels required to excite nociceptors. Moreover, slow depolarization of nociceptor membrane potentials might lead to inactivation of voltage-gated channels that, in turn, inhibits the generation of action potentials [31].

Because cannabinoids can trigger peripherally mediated antihyperalgesia and antinociception, it is important to address whether their modulation of ICRs regulates peripheral nociceptor activity. Application of cannabinoids elevates internal Ca²⁺ ([Ca²⁺]_i) levels in nociceptors because known ICRs (TRP channels) are permeable to Ca²⁺ (Table 1). An elevation of [Ca²⁺]_i can ignite numerous cellular cascades, including induction of Ca₂₊-dependent kinases, phosphatases and phosphatidylinositol-4,5-bisphosphate (PtdIns(4,5)P₂) [32–35]. It is well established that these enzymes and lipids can be effective modulators of activities of the TRPV1 and TRPA1 channels [33,34,36–39], which have crucial roles in regulating nociception [40–43]. Thus, one hypothesis is that certain cannabinoids regulate nociceptors by activating Ca²⁺-permeable channels. As detailed later, the remarkable outcome of this activation leads to a desensitization of nociceptor activities. Recent studies have indeed demonstrated that cannabinoids that selectively activate TRPV1 (e.g. arachidonoylchloro-ethanolamide [ACEA]) or TRPA1 (e.g. WIN55212 or AM1241) induce a homologous and cross-desensitization (i.e. the desensitization of one channel after activation of another channel) of the activities of both of these TRP channels [4,19,20], ultimately resulting in peripherally mediated antihyperalgesia [4,19]. The simultaneous desensitization of multiple TRP channels might lead to a broader inhibition of nociceptor responsiveness than that observed by inhibition of only one channel and, thus, represents a potentially novel approach for the generation of analgesic compounds. Studies conducted in cell expression systems indicate that TRPV1-selective cannabinoids can only desensitize TRPA1 when TRPV1 is co-expressed. Similarly, TRPA1-selective cannabinoids desensitize TRPV1 only under conditions in which TRPA1 is present [4]. Thus, the cross-desensitization produced by these TRP-selective cannabinoids requires the co-expression of both TRPV1 and TRPA1.

To understand how activation of ICRs leads to inhibition of nociceptors, molecular mechanisms of desensitization of TRP channels by ICR-activating cannabinoids were investigated. The results indicate that cross-desensitization between the TRPA1 and TRPV1 channels in sensory neurons seems to involve multiple separate mechanisms.

First, TRPV1 is desensitized by TRPV1- and TRPA1-activating cannabinoids via Ca²⁺-dependent calcineurin (phosphatase 2B)-induced dephosphorylation of the channel [19,20] (Figure 1). A similar mechanism is employed in the desensitization of TRPV1 by the TRPA1-specific agonist mustard oil (MO) [33,44]. Interestingly, Ca²⁺-induced PtdIns(4,5)P₂ biosynthesis, a mechanism involved in tachyphylaxis of capsaicin responses in heterologous expression systems [34,38], does not play a part in desensitization of TRPA1 by activation of TRPV1 or in cross-desensitization of TRPV1 by activation of the TRPA1 channel in sensory neurons [33]. Furthermore, Ca²⁺-dependent activation of kinases can be responsible for recovery from desensitization [37]. Overall, Ca²⁺-dependent desensitization prevails over sensitization because calcineurin is more effectively induced by low concentrations of Ca²⁺ than kinases [45]. Thus, calcineurin-mediated dephosphorylation is an important mechanism for cannabinoid desensitization of TRPV1 and compounds that block calcineurin produce a significant inhibition of peripheral cannabinoid antihyperalgesia [18] (Figure 1).

Figure 1.

Mechanisms for cannabinoid cross-desensitization of TRPV1 and TRPA1. Cannabinoids desensitize TRPV1 via activation of calcineurin and dephosphorylation of the ion channel. Homologous desensitization of TRPV1 can occur by application of TRPV1-selective cannabinoids (e.g. ACEA), and heterologous desensitization of TRPV1 can occur by administration of TRPA1-selective cannabinoids (e.g. WIN55212). Cannabinoids desensitize TRPA1 via activation of a calcium-independent pathway. Abbreviations: DAG, diacylglycerol; IP₃, inositol (1,4,5)-trisphosphate; PHD, pleckstrin homology domain; PIP₂, phosphatidylinositol (4,5)-bisphosphate; PLC, phospholipase C.

A second potential mechanism for TRPA1 desensitization is via a Ca²⁺-dependent depletion of PtdIns(4,5)P₂ [33]. However, unlike capsaicin [33,34], tested cannabinoids cannot induce Ca²⁺-dependent depletion of PtdIns(4,5)P₂, possibly owing to their slow or reduced accumulation of intracellular Ca²⁺ levels [20]. Thus, this mechanism does not seem to be utilized for cannabinoid desensitization of TRP channels.

A third mechanism for desensitization of TRPA1 is via activation of a Ca²⁺-independent pathway [33,44]. This pathway is evident for TRPA1 desensitization by low concentrations of MO [46] and possibly by a partial agonist for the TRPA1 channel, notably the cannabinoid WIN55212 [4,33]. Altogether, certain cannabinoids are able to inhibit nociception by activating ICRs – TRP channels.

Roles of ICRs in peripheral antihyperalgesia

There is broad agreement that cannabinoids can produce peripherally mediated antihyperalgesic and antinociceptive effects by any of several proposed mechanisms [4,19,47–49]. Experimental findings support at least four distinct hypotheses (Figure 2). The first hypothesis is that cannabinoids mediate their actions via different metabotropic or ionotropic receptors. Thus, peripherally restricted doses of WIN55212 significantly inhibit capsaicin-induced nocifensive behavior in wild-type animals, and this effect is lost in either TRPV1 or CB₁-gene knockouts but not in mice with genetic deletion of the CB₂ receptor [4]. Thus, cannabinoids might activate multiple receptor mechanisms to produce peripherally mediated antihyperalgesia. The second hypothesis is based upon the observation that cannabinoid receptor systems are differentially activated in various pain models (e.g. inflammatory versus neuropathic versus basal pain thresholds) or by distinct experimental approaches (e.g. local versus systemic versus intrathecal injection). For example, certain immune cells expressing CB₂ might have greater contributions to nociceptor activation in inflammatory pain models compared with neuropathic pain models. In addition, inflammation has been reported to trigger upregulation of the CB₁ receptor in sensory neurons [50]. The third hypothesis is based upon the finding that cannabinoids can modulate sensory neurons in addition to non-neuronal peripheral cells. Thus, the overall cannabinoid effect might be an integration of activity across several peripheral cell types (e.g. nociceptor terminals or keratinocytes [32]). Fourth, it is possible that ICRs and MCRs can functionally cooperate under certain conditions [5,51] because ionotropic and metabotropic receptors are co-expressed in many cells [52].

Figure 2.

Proposed mechanisms for the peripherally mediated antihyperalgesic effects of cannabinoids. (a) Hypothesis 1 proposes that cannabinoids produce peripheral antihyperalgesia by activation of both MCRs and ICRs. (b) Hypothesis 2 proposes that different pain models (or routes of drug injection) selectively activate either MCR or ICR pathways. (c) Hypothesis 3 proposes that the net effect of cannabinoids on modulating nociception is due to activation of MCRs and/or ICRs located on multiple cell types that are capable of interacting. (d) Hypothesis 4 proposes that MCRs and ICRs interact in the same cell leading to desensitization of activity. Note that the selective involvement of MCR and/or ICR in each hypothesis is for illustrative purposes only.

How important are ionotropic cannabinoid mechanisms in such a diversity of analgesic cannabinoid hypotheses? Interestingly, evidence from knockout animals indicates that both pathways are equally important in at least some pain models because knockout of either ICR or MCR genes reduces peripherally mediating antihyperalgesic effects of cannabinoids [4]. Several hypotheses could explain these findings. First, ICRs and MCRs operate independently and mediate the actions of cannabinoids on different types of cells. For example, CB₂-mediated antinociception occurs via activation of CB₂ on keratinocytes [53], whereas TRPA1-mediated effects of cannabinoids take place on nociceptors [4]. Thus, the relative role of specific cell types in various pain models could alter the contribution of distinct cannabinoid receptor systems. Second, ICRs and MCRs effectively mediate antihyperalgesia or antinociception in distinct pain models. Third, MCRs could cooperate with ionotropic receptors on cells where both are co-expressed. This could result in either direct inhibition of nociceptors or promotion of Ca²⁺-dependent release of nociception inhibitory factors (i.e. opioids, endocannabinoids etc) by non-neuronal peripheral cells. When co-expressed, ICRs and MCRs can cooperate in a variety of ways. For example, the activation of CB₁ on mesencephalic dopaminergic neurons can produce intracellular 12(S)-hydroxyeicosatetraenoic acid, which is a TRPV1 agonist [5]. Cannabinoids can also be coupled to G_q/11 proteins [54,55], and this could lead to activation of the phospholipase C pathway which, in turn, could result in the activation of several TRP channels including TRPA1 [56]. In addition, fatty acid amine hydrolase (FAAH) is involved in anandamide metabolism and is co-localized with TRPV1 in several areas, indicating an important regulatory interaction [57]. Indeed, inhibition of FAAH rapidly leads to accumulation of fatty acid amides, some of which are efficacious TRP agonists [58]. Thus, URB597, a potent and systemically active inhibitor of FAAH, activates TRPA1 channels in addition to peroxisome proliferator-activated receptor-α [58,59].

Conclusion and future perspectives

Despite the wealth of information on cannabinoid-induced peripheral antihyperalgesic and antinociceptive actions in many pain models, the molecular mechanism(s) for these effects remains unknown. Recent investigation in these mechanisms has yielded the hypothesis that activation of ICRs can result in desensitization of the TRPA1 and TRPV1 channel activities, inhibition of nociceptors and antihyperalgesia and antinociception in certain pain models. One important conclusion is that, although cannabinoids differ in their activations of various receptors, they could exert inhibitory effects by acting through ICRs. In addition, the control of intracellular activities by cannabinoids could occur via two very distinct pathways (i.e. metabotropic versus ionotropic), providing multiple mechanisms for triggering antihyperalgesia and antinociception. Thus, it is conceivable that partial TRP-channel-specific agonists could constitute a novel class of peripherally selective analgesics without the typical side effects of conventional cannabinoids.

Nevertheless, many challenges still remain. Among these is the detailed characterization of molecular mechanisms responsible for desensitization of nociceptor-specific TRP channels by ICR-activating cannabinoids, the investigation of the function of ICR in peripheral cells (such as blood and skin cells) that might have roles in the process of nociception, and the evaluation of possible co-operation between MCRs and ICRs in nociceptive neurons in addition to other peripheral cells contributing to nociception.

Glossary

Activation^*

increased functional activity of neurons (e.g. action potential or exocytosis).

Allodynia

occurrence of nociception in response to a non-noxious stimulus.

Antiallodynia

reduction of allodynia.

Antihyperalgesia

reduction of hyperalgesia.

Antinociception

increase in nociceptive thresholds above basal nociceptive levels.

Basal nociceptive thresholds

stimulus intensity capable of evoking an escape response under basal (un-injured) conditions. Usually a thermal or mechanical stimulus.

Hyperalgesia

increased magnitude of nociception in response to a noxious stimulus.

Inflammatory pain models

animal models in which nociception occurs owing to injury of a peripheral tissue.

Inverse agonist

a compound that binds to the same site as an agonist but reverses the constitutive activity of the receptor.

Neuropathic pain models

animal models in which nociception occurs owing to injury to a peripheral nerve.

Nociception

the neural encoding and modulation of noxious stimuli.

Pain

an unpleasant sensory and emotional experience associated with actual or potential tissue damage.

Peripheral nociceptive neurons

specialized class of peripheral afferent neurons that detects and encodes stimuli that induce tissue injury or chemical factors released during tissue injury.

Sensitization

increased responsiveness of neurons.