Endocannabinoids and cannabinoid receptors in ischaemia–reperfusion injury and preconditioning


Section on Oxidative Stress and Tissue Injury, Laboratory of Physiological Studies, NIAAA, National Institutes of Health, 5625 Fishers Lane, MSC-9413, Bethesda, MD 20892-9413, USA. E-mail: pacher@mail.nih.gov


Ischaemia–reperfusion (I/R) is a pivotal mechanism of organ injury during stroke, myocardial infarction, organ transplantation and vascular surgeries. Ischaemic preconditioning (IPC) is a potent endogenous form of tissue protection against I/R injury. On the one hand, endocannabinoids have been implicated in the protective effects of IPC through cannabinoid CB1/CB2 receptor-dependent and -independent mechanisms. However, there is evidence suggesting that endocannabinoids are overproduced during various forms of I/R, such as myocardial infarction or whole body I/R associated with circulatory shock, and may contribute to the cardiovascular depressive state associated with these pathologies. Previous studies using synthetic CB1 receptor agonists or knockout mice demonstrated CB1 receptor-dependent protection against cerebral I/R injury in various animal models. In contrast, several follow-up reports have shown protection afforded by CB1 receptor antagonists, but not agonists. Excitedly, emerging studies using potent CB2 receptor agonists and/or knockout mice have provided compelling evidence that CB2 receptor activation is protective against myocardial, cerebral and hepatic I/R injuries by decreasing the endothelial cell activation/inflammatory response (for example, expression of adhesion molecules, secretion of chemokines, and so on), and by attenuating the leukocyte chemotaxis, rolling, adhesion to endothelium, activation and transendothelial migration, and interrelated oxidative/nitrosative damage. This review is aimed to discuss the role of endocannabinoids and CB receptors in various forms of I/R injury (myocardial, cerebral, hepatic and circulatory shock) and preconditioning, and to delineate the evidence supporting the therapeutic utility of selective CB2 receptor agonists, which are devoid of psychoactive effects, as a promising new approach to limit I/R-induced tissue damage.

British Journal of Pharmacology (2008) 153, 252–262; doi:10.1038/sj.bjp.0707582; published online 19 November 2007








ischaemic preconditioning




Ischaemic–reperfusion (I/R) injury is the principal cause of tissue damage occurring in conditions such as stroke, myocardial infarction, cardiopulmonary bypass and other vascular surgeries, and organ transplantation, as well as a major mechanism of end-organ damage complicating the course of circulatory shock of various aetiologies. In all these conditions, the initial trigger of the damage is the transient disruption of the normal blood supply to target organs followed by reperfusion. Reperfusion of ischaemic tissues is the ultimate treatment to reduce tissue injury. Unfortunately, reperfusion itself inflicts additional tissue damage mediated by reactive oxygen (superoxide anion, hydrogen peroxide and hydroxyl radical) and reactive nitrogen species (for example, peroxynitrite) upon reperfusion, as well as to the rapid transcriptional activation of an array of proinflammatory genes (reviewed in Ferdinandy and Schulz, 2003; Podgoreanu et al., 2005; Pacher et al., 2005e, 2007; Ungvari et al., 2005). Direct consequences are the local sequestration and activation of polymorphonuclear leukocytes, leading to a rapid amplification of the initial inflammatory response and reactive oxygen species generation, so-called ‘respiratory burst’ (Lucchesi, 1990). Additional sources of increased reactive oxygen species generation during I/R can be xanthine and NAD(P)H oxidases, mitochondria, COX and uncoupled nitric oxide synthases (reviewed in Griendling et al., 2000; Ungvari et al., 2005; Pacher et al., 2006b). The burst of reactive oxygen and nitrogen species immediately upon reperfusion initiates a chain of deleterious cellular responses eventually leading to endothelial inflammatory response and dysfunction, adherence of neutrophils and lymphocytes to the endothelium, transendothelial migration of inflammatory cells, the release of various harmful mediators, cellular calcium overload, and eventually cell death and organ dysfunction.

Ischaemic preconditioning (IPC), first introduced by Murry et al. (1986) is a potent endogenous form of protection against I/R injury. In hearts and various other organs, IPC (brief episode(s) of ischaemia applied before the main I/R) reduces infarct size and enhances the recovery of organ function (Yellon and Hausenloy, 2005). Preconditioning can also be achieved with bacterial endotoxins and various other chemicals and when brief episodes of ischaemia are applied following the ischaemic period (the latter is called postconditioning (for reviews see Yellon and Hausenloy, 2005; Bolli, 2007; Hausenloy and Yellon, 2007).

Circulatory shock classifies a syndrome precipitated by a systemic derangement in perfusion leading to widespread cellular hypoxia and vital organ dysfunction. Dependent on its initial pathophysiological mechanisms, shock is subdivided in three main categories, namely cardiogenic, haemorrhagic and septic shock. In the advanced stages, all shock states evolve to a common clinical picture characterized by profound tissue ischaemia, cardiovascular failure, the activation of cellular cytotoxic effectors (polymophonuclear leukocytes) and the upregulation of an array of proinflammatory genes, leading to systemic inflammation, organ dysfunction and death (Hotchkiss and Karl, 2003). From a clinical point of view, at present no therapy is available to limit reperfusion injury, which highlights the importance of a better understanding of its underlying mechanisms, to devise better future therapeutic approaches.

Numerous studies have suggested that the endocannabinoid system may modulate I/R injury (reviewed in Lamontagne et al., 2006; Pacher et al., 2006a). Endocannabinoids have been implicated in the protective effects of IPC through cannabinoid (CB) receptor-dependent and -independent mechanisms; however, they may also contribute to the cardiovascular collapse associated with myocardial infarction and circulatory shock (reviewed in Pacher et al., 2006a). To date, two CB receptors have been identified by molecular cloning: the CB1 and CB2 receptors. The CB1 receptor is abundantly expressed in brain tissue (Matsuda et al., 1990), but is also present in peripheral tissues including vasculature (Gebremedhin et al., 1999; Liu et al., 2000), heart (Batkai et al., 2004b; Pacher et al., 2005d) and liver (Batkai et al., 2001; Engeli et al., 2005; Osei-Hyiaman et al., 2005; Teixeira-Clerc et al., 2006). The CB2 receptor was previously considered to be expressed primarily in immune and haematopoietic cells (Munro et al., 1993; reviewed in Pacher et al., 2006a). However, more recent studies have also found CB2 receptors in brain (Van Sickle et al., 2005), myocardium (Mukhopadhyay et al., 2007), cardiomyoblasts (Shmist et al., 2006; Mukhopadhyay et al., 2007) and endothelial cells of various origins (Blazquez et al., 2003; Zoratti et al., 2003; Golech et al., 2004; Mestre et al., 2006; Rajesh et al., 2007a, 2007b). The synthetic and natural ligands (the latter called endocannabinoids: arachidonoyl ethanolamide or anandamide (AEA) and 2-arachidonoylglycerol (2-AG)) of CB receptors exert various anti-inflammatory and neuroprotective (Panikashvili et al., 2001, 2005, 2006) effects by inhibiting the generation and release of proinflammatory cytokines and mediators (reviewed in Mechoulam et al., 2002a, 2002b; Klein, 2005; Pacher et al., 2006a). The pharmacological modulation of the endocannabinoid system represents a promising strategy in various cardiovascular, inflammatory, metabolic, gastrointestinal and liver disorders (reviewed in Di Marzo et al., 2004; Pacher et al., 2005a, 2006a; Pertwee, 2005b; Mackie, 2006; Mallat et al., 2007). The selectivity of the endocannabinoids and synthetic ligands used in various I/R studies towards CB1/CB2 receptors are summarized in Table 1 (for excellent detailed overviews on the subject and also on the development of CB1/CB2 receptor knockout mice, see Howlett et al., 2002; Pertwee, 2005a).

Table 1.  Ranges of Ki values for certain cannabinoid CB1 and/or CB2 receptor agonists or antagonists/inverse agonists for the in vitro displacement of [3H]CP55940, [3H]HU243 or [3H]BAY38-7271 from CB1- and CB2-specific binding sites (based on Howlett et al., 2002; Pertwee, 2005a; for details see therein)
Agonist/ligandCB1 Ki value (nM)CB2 Ki value (nM)Reference
CB1-selective agonists
 ACEA1.4–5.29195 to >2000Pertwee (2005a)
 R-(+)-methanandamide17.9–28.3815–868Pertwee (2005a)
Agonists without any marked CB1/CB2 selectivity
 Anandamide61–543279–1940Pertwee (2005a)
 BAY38-72711.855.96Pertwee (2005a)
 2-Arachidonoyl glycerol58.3–472145–1400Pertwee (2005a)
 HU-2100.0608–0.10.17–0.524Pertwee (2005a)
 CP559400.5–50.69–2.8Pertwee (2005a)
 Δ9-THC5.05–53.33.13–75.3Pertwee (2005a)
 R-(+)-WIN 55,212-21.89–1230.28–16.2Pertwee (2005a)
CB2-selective agonists
 JWH01538313.8Pertwee (2005a)
 JWH1336773.4Pertwee (2005a)
 HU-308>10 00022.7Pertwee (2005a)
 O-19665055±98423±2.1Wiley et al. (2002)
 O-38531509±1486.0±2.5Zhang et al. (2007)
CB1-selective antagonist/inverse agonists
 SR141716A1.8–11.8515–13 200Pertwee (2005a)
 AM281124200Pertwee (2005a)
 AM2517.492290Pertwee (2005a)
 LY32013514114 900Pertwee (2005a)
CB2-selective antagonist/inverse agonists
 AM630515231.2Pertwee (2005a)
 SR14452850.3 to >10 0000.28–5.6Pertwee (2005a)

In this review, we will discuss the triggers and sources of endocannabinoid production during various forms of I/R injury (myocardial, cerebral, hepatic and retinal ischaemia, and circulatory shock) and preconditioning, as well as the diverse role of these novel mediators and their receptors in these processes. We will also overview the accumulating evidence obtained through the use of various synthetic CB1/CB2 receptor ligands, with particular focus on the novel role of CB2 receptors, suggesting that the modulation of the endocannabinoid system can be therapeutically exploited in various forms of I/R injury.

Myocardial I/R and preconditioning

Initial studies used isolated heart preparations to study the role of endocannabinoid system in myocardial I/R and preconditioning. Lagneux and Lamontagne (2001) implicated for the first time the involvement of the endocannabinoid system in endotoxin (lipopolysaccharide (LPS))-induced preconditioning against myocardial I/R injury, based on the assumption that LPS increases endocannabinoid production in inflammatory cells (Varga et al., 1998; Maccarrone et al., 2001; Liu et al., 2003). They compared the effects of 90 min of low-flow ischaemia followed by 60-min reperfusion at normal flow in isolated hearts from rats pretreated with LPS or saline. LPS pretreatment reduced infarct size and enhanced functional recovery upon reperfusion compared to controls, which could be attenuated by the CB2 antagonist SR144528, but not by the CB1 antagonist SR141716, suggesting the involvement of myocardial CB2 receptors in the observed LPS-induced cardioprotection (Lagneux and Lamontagne, 2001). In a consequent study, in which the preconditioning was triggered by heat stress, SR144528 but not SR141716 also abolished the infarct-size-reducing effect of heat stress (Joyeux et al., 2002). The conclusion of these early studies was that the protection afforded by LPS- or heat stress-induced preconditioning was mediated by endocannabinoids acting on CB2 receptors. In contrast, in preconditioning induced by a brief period of ischaemia (5 min), either CB2 or CB1 receptor blockade could abolish the protection, and both CB1 and CB2 receptors were implicated in the preservation of the endothelium-dependent, 5-HT-induced vasodilation by IPC (Bouchard et al., 2003). Palmitoylethanolamide or 2-AG, but not AEA, added to the perfusion medium of isolated rat hearts afforded protection against ischaemia by decreasing myocardial damage and infarct size and by improving myocardial functional recovery (Lepicier et al., 2003). SR144528 completely blocked the cardioprotective effect of both palmitoylethanolamide and 2-AG, whereas SR141716 only partially inhibited the effect of 2-AG only (Lepicier et al., 2003). Similarly, CB1 and CB2 agonists ACEA and JWH015 also reduced infarct size in this model, and the CB2 receptor-mediated cardioprotection by palmitoylethanolamide involved activation of p38/ERK (extracellular signal-regulated kinase)1/2 kinases and PKC (Lepicier et al., 2003). On the contrary, Underdown et al. (2005) have found that the infarct-size-reducing effect of AEA could be equally well antagonized by both CB1 and CB2 antagonists; however, it could not be mimicked by selective CB1 or CB2 agonists, suggesting the involvement of a site distinct from CB1 or CB2 receptors. Another recent study using a model of delayed preconditioning in rats induced by administration of the nitric oxide donor nitroglycerin for 24 h via transdermal application suggested that the protective effect of nitroglycerin against myocardial infarction is mediated via CB1 receptors. Nitroglycerin increased the myocardial content of 2-AG, but not AEA (Wagner et al., 2006). The major limitation of the above-mentioned studies is the use of ex vivo models (for example, buffer-perfused isolated heart preparations) that could not address the question of whether endocannabinoids or synthetic agonists can modulate endothelial or immune cell activation and interactions, which are pivotal events in the sequel of reperfusion damage (reviewed in Pertwee, 2005a; Lamontagne et al., 2006; Pacher et al., 2006a). Despite the above-mentioned limitation, these pioneering studies importantly implied the possible contribution of functional CB2 receptors in cardiomyocytes and/or endothelial cells responsible, at least in part, to the protective effects of preconditioning. Indeed, consequent studies have demonstrated the presence of CB2 receptors in myocardium (Mukhopadhyay et al., 2007), cardiomyoblasts (Shmist et al., 2006; Mukhopadhyay et al., 2007) and endothelial cells of various origins (Blazquez et al., 2003; Zoratti et al., 2003; Golech et al., 2004; Mestre et al., 2006; Rajesh et al., 2007a, 2007b). Consistently with the beneficial effect of CB2 receptor activation on cardiomyocytes, a recent study demonstrated that delta(9)-tetrahydrocannabinol (THC) protected H9c2 cardiomyoblasts subjected to hypoxia in vitro presumably via CB2 receptor activation and increased nitric oxide production (Shmist et al., 2006).

In a clinically more relevant rat model of I/R injury, both AEA and HU-210 decreased the incidence of ventricular arrhythmias and reduced infarct size, presumably through the activation of CB2 but not CB1 receptors (Krylatov et al., 2001, 2002a, 2002b; Ugdyzhekova et al., 2001, 2002). In a mouse model of myocardial I/R induced by coronary artery ligation, the reduction of leukocyte-dependent second wave of myocardial damage subsequent to the initial IR injury was attributed to CB2 receptor activation, since the protection afforded by WIN 55,212-2 could be prevented by AM630, but not by the CB1 antagonist AM251 (Di Filippo et al., 2004). Two recent studies using rat models of acute and chronic myocardial infarction demonstrated that endocannabinoids contribute to the hypotension and cardiodepression associated with acute cardiogenic shock, which could be attenuated by CB1 antagonists (Wagner et al., 2001, 2003).

Collectively, although the role of CB receptors and endocannabinoids in protection afforded by preconditioning against myocardial I/R is still a very controversial issue requiring further clarification by using knockout mice and more selective ligands for CB2 receptors, the findings implicating the importance of CB2 receptor, presumably both on endothelial and inflammatory cells, and perhaps on cardiomyocytes, are very encouraging.

Cerebral I/R (stroke)

Ischaemic stroke, resulting from the reduction of cerebral blood flow in the territory of a major cerebral artery due to its transient or permanent occlusion by local thrombosis or embolus is the second leading cause of death in industrialized countries and the leading medical cause of acquired adult disability. One in six patients die in the first 4 weeks following ischaemic stroke, and half of the survivors are permanently disabled in spite of the best efforts to rehabilitate them to avoid complications (Klijn and Hankey, 2003). A cascade of complex molecular events is set in motion during cerebral ischaemia and culminates in neuronal cell death. Improving our understanding of these events might help to devise novel therapies to limit neuronal injury in stroke patients, a concept termed ‘neuroprotection’ (Lees et al., 2006).

The endocannabinoid system may represent a pivotal neuroprotective mechanism both in acute forms of neuronal injury (for example, stroke and traumatic brain injury) and in various chronic neurodegenerative disorders, including multiple sclerosis, Parkinson's disease, Huntington's disease, Alzheimer's disease and amyotrophic lateral sclerosis (reviewed in Mechoulam et al., 2002b; Croxford, 2003; Sarne and Mechoulam, 2005; Mackie, 2006; Pacher et al., 2006a). Even though the exact mechanisms of these neuroprotective effects are not completely understood, numerous CB receptor-dependent as well as receptor-independent processes have been suggested to be involved, which include, but are not limited to: (1) modulation of immune responses and the release of inflammatory mediators by CB1, CB2 and non-CB1/CB2 receptors on neurons, astrocytes, microglia, macrophages, neutrophils and lymphocytes (Walter and Stella, 2004; Klein, 2005); (2) modulation of synaptic plasticity and excitatory glutamatergic transmissions via presynaptic CB1 receptors (Freund et al., 2003; Piomelli, 2003); (3) activation of cytoprotective signalling pathways (for example, PKB/Akt, PKA or neurotrophic factors) (Pacher et al., 2006a); (4) modulation of calcium homoeostasis and excitability via interactions with Ca2+, K+ and Na+ channels, gap junctions and intracellular Ca2+ stores, NMDA receptors (Freund et al., 2003; Piomelli, 2003; Pacher et al., 2006a); (5) CB1 receptor-mediated central hypothermia, presumably by decreasing metabolic rate and oxygen demand; (6) antioxidant properties of CBs (Hampson et al., 2000); (7) modulation of endothelial activation and inflammatory response, leukocyte rolling, adhesion to the endothelium, transmigration and activation presumably by CB2 receptors.

The first evidence for the neuroprotective effect of CBs came from the stroke research field from studies using synthetic non-psychotropic CB Dexanabinol/HU-211, which exerted its beneficial effects through CB1/CB2-independent mechanisms, in various rat and gerbil models of in vivo cerebral ischaemia (reviewed in Pacher et al., 2006a). Follow-up studies have also investigated the neuroprotective effects of CB1 receptor stimulation using synthetic agonists. The synthetic CB WIN 55,212-2 attenuated hippocampal neuronal loss following transient global cerebral ischaemia in rats and reduced infarct size after permanent focal cerebral ischaemia induced by middle cerebral artery occlusion, when given 40 min prior or 30 min after the occlusion, in a CB1-dependent manner, since the protective effect was preventable by SR141716 (Nagayama et al., 1999). WIN 55,212-2, as well as AEA and 2-AG, also protected cultured cerebral cortical neurons from in vitro glucose deprivation and hypoxia, but these effects were insensitive to CB1 and CB2 receptor antagonists (Nagayama et al., 1999; Sinor et al., 2000). In rat models of middle cerebral artery occlusion another synthetic agonist BAY38-7271 reduced infarct size even when given intravenously 4 h following the occlusion (Mauler et al., 2002). Similarly, HU-210 improved motor disability and decreased infarct size by up to 77% in a similar model (Leker et al., 2003). Pretreatment with SR141716 partially attenuated the protective effect of HU-210 indicating CB1 receptor involvement. However, the protective effect of HU-210 could be abolished completely by warming the animals to the body temperature of controls, indicating that the CB1-mediated hypothermia was responsible for the observed beneficial effects (Leker et al., 2003). Similarly, CB1-mediated hypothermia was responsible for the neuroprotective effects of Δ(9)-tetrahydrocannabinol in a mouse ischaemic model of cerebral injury (Hayakawa et al., 2004) and, perhaps, also in a rat model of global cerebral ischaemia (Louw et al., 2000). Consistent with CB1-mediated cerebroprotection, CB1 knockout mice had increased neurotoxicity to NMDA and elevated mortality from permanent focal cerebral ischaemia, increased infarct size, more severe neurological deficits after transient focal cerebral ischaemia and decreased cerebral blood flow in the ischaemic penumbra during reperfusion, as compared to wild-type controls subjected to the same insult (Parmentier-Batteur et al., 2002).

In contrast, several more recent studies do not support the neuroprotective role of endocannabinoids and CB1 receptor activation. In fact, the CB1 antagonists SR141716 and LY320135 were found to reduce infarct size and to improve neurological function in a rat model of cerebral ischaemia induced by middle cerebral artery occlusion (Berger et al., 2004; Muthian et al., 2004; Sommer et al., 2006), whereas low doses of WIN 55,212-2 had no protective effect (Muthian et al., 2004).

Recent studies have also evaluated the effects of selective CB2 agonists (O-3853, O-1966) in a stroke model. CB2 agonists significantly decreased cerebral infarction and improved motor function after 1 h middle cerebral artery occlusion followed by 23 h reperfusion in mice, by attenuating the transient ischaemia-induced increase in leukocyte rolling and adhesion to vascular endothelial cells (Zhang et al., 2007). The role of CB2 receptors in I/R injury was further supported by increased accumulation of CB2-positive macrophages derived from resident microglia and/or invading monocytes following cerebral I/R (Ashton et al., 2007).

Collectively, it appears that both CB1 agonists and antagonists may afford neuroprotective effects against cerebral I/R. The reason for the contradictory effects of pharmacological blockade vs genetic knockout of CB1 receptors is not clear, and may be related to CB1 receptor-independent effects of antagonists, but this issue needs further clarification. In the case of CB2 agonists, the most likely mechanism of protection is the attenuation of the transient I/R-induced increase in leukocyte infiltration, rolling and adhesion to vascular endothelial cells, and consequent activation.

Circulatory shock (full organ/body ischaemia and/or I/R)

In addition to their well-known immunological and neurobehavioral actions, CBs and their endogenous and synthetic analogues exert complex cardiodepressive and vasodilatory effects, which have been implicated in the mechanism of hypotension associated with haemorrhagic (Wagner et al., 1997; Cainazzo et al., 2002), endotoxic (Varga et al., 1998; Liu et al., 2003, 2006; Batkai et al., 2004a; Kadoi and Goto, 2006), septic (Kadoi et al., 2005), and cardiogenic shock (Wagner et al., 2001, 2003), advanced liver cirrhosis (Batkai et al., 2001; Ros et al., 2002), cirrhotic cardiomyopathy (Gaskari et al., 2005; Pacher et al., 2005c; Moezi et al., 2006; Yang et al., 2007; Batkai et al., 2007a), doxorubicin-induced heart failure (Mukhopadhyay et al., 2007) and the shock associated with necrotizing pancreatitis (Matsuda et al., 2005). Importantly, these cardiovascular depressive effects could be prevented or reversed by pretreatment with CB1 antagonists, and are subjects of numerous comprehensive recent overviews (Randall et al., 2002; Hiley and Ford, 2004; Lamontagne et al., 2006; Lepicier et al., 2006; Mallat et al., 2007; Mendizabal and Adler-Graschinsky, 2007; Pacher et al., 2005a, 2005b, 2006a). CB receptor antagonists (for example, SR141716, AM281, AM251 and SR144528) prolonged survival in endotoxic and septic shock or necrotizing pancreatitis (Varga et al., 1998; Smith et al., 2001; Cainazzo et al., 2002; Kadoi et al., 2005; Matsuda et al., 2005; Kadoi and Goto, 2006), while increased mortality in haemorrhagic (Wagner et al., 1997) and cardiogenic shock (Wagner et al., 2001) despite the increase in blood pressure. One explanation for this puzzling controversy is that endocannabinoid-mediated vasodilation may have survival value through improving tissue oxygenation by counteracting the excessive sympathetic vasoconstriction triggered by haemorrhage or myocardial infarction, and this would be removed by CB1 blockade. In contrast, CB1 blockade may improve survival in endotoxic shock by preventing the primary hypotensive response to LPS (Randall et al., 2002; Hiley and Ford, 2004; Mendizabal and Adler-Graschinsky, 2007; Pacher et al., 2005a, 2005b, 2006a). Complicating the picture, in haemorrhagic, cardiogenic and endotoxic shock, the CB agonists HU-210, WIN 55,212-2 and THC also improved endothelial function and/or survival (Wagner et al., 1997, 2001; Varga et al., 1998; Smith et al., 2000, 2001). Since the cardiovascular dysfunction and failure in most of the above-mentioned conditions are triggered by overwhelming tissue ischaemia and/or I/R, and consequent oxidative/nitrosative stress and inflammatory response coupled with the activation of various downstream cell death pathways (reviewed in Evgenov and Liaudet, 2005; Ungvari et al., 2005; Pacher et al., 2005e, 2007), another explanation for the diverse beneficial effects of both agonists and antagonists in circulatory shock could lie in their various anti-inflammatory and/or antioxidant properties (reviewed in Walter and Stella, 2004; Klein, 2005), which may be attributed to their inverse agonistic properties or to CB1/2 receptor-independent mechanisms (reviewed in Begg et al., 2005; Pertwee, 2005a, 2005b, 2006).

Overall, it seems that both CBs and antagonists of CB receptors may have various favorable effects in rodent shock models; however, the specificity of these effects and the relevance to human circulatory shock should be established by further studies.

Hepatic I/R

Hepatic I/R injury continues to be a fatal complication that can follow liver surgery or transplantation. It is well known that hepatic I/R injury is dependent on polymorphonuclear cell (PMN) infiltration, Kupffer cell activation and inflammatory cytokine responses (Jaeschke et al., 1996, 1997, 2006; Ohkohchi et al., 1999). Adhesion molecules mediate the initial attachment of neutrophils to the activated endothelium (Carlos and Harlan, 1994; Jaeschke, 1997). On reperfusion, tumour necrosis factor-α (TNF-α) acts as a continuous stimulator for neutrophil infiltration in the liver and it also upregulates the production of cell-type-specific leukocyte chemoattractants, known as chemokines, which have also been shown to cause upregulation of cell adhesion molecules and neutrophil activation (Jaeschke, 2006). The increased inflammatory response further aggravates oxidative stress and initiates a chain of deleterious events eventually culminating in cellular dysfunction and death.

In two recent studies, we have investigated the involvement of the endocannabinoid system in an in vivo mouse model of hepatic I/R injury using selective CB2 agonists and CB2 knockout mice (Batkai et al., 2007b). Activation of CB2 receptors by JWH133 prior to the insult protected against I/R damage (measured by serum transaminases activity) by decreasing inflammatory cell infiltration, tissue and serum TNF-α, chemokines macrophage-inflammatory protein-1α (MIP-1α) and macrophage-inflammatory protein-2 (MIP-2) levels, tissue lipid peroxidation, and expression of adhesion molecule intercellular adhesion molecule-1 (ICAM-1). JWH133 also decreased the TNF-α-induced ICAM-1 and vascular cell adhesion molecule-1 expressions in human liver sinusoidal endothelial cells and the adhesion of human neutrophils to human liver sinusoidal endothelial cells in vitro. In agreement with the protective role of CB2 receptor activation, CB2−/− mice developed increased I/R-induced tissue damage and proinflammatory phenotype (Batkai et al., 2007b). In a follow-up study, we have demonstrated that the potent CB2 receptor agonist HU-308, given prior to the induction of I/R, significantly attenuated the extent of liver damage (measured by serum alanine aminotransferase and lactate dehydrogenase), decreased serum and tissue TNF-α, MIP-1α and MIP-2 levels, tissue lipid peroxidation, neutrophil infiltration, DNA fragmentation and caspase 3 activity. The protective effect of HU-308 against liver damage was also preserved when given immediately after the ischaemic episode. CB2 receptor was expressed in human liver sinusoidal endothelial cells and its activation by HU-308 also attenuated the TNF-α-induced ICAM-1 and vascular cell adhesion molecule-1 expression and the adhesion of human neutrophils to human liver sinusoidal endothelial cells in vitro. These findings, coupled with recent results from myocardial (Di Filippo et al., 2004) and cerebral I/R models (Zhang et al., 2007), and antifibrotic effects of CB2 receptor in the liver (Julien et al., 2005), suggest that selective CB2 receptor agonists may represent a novel protective strategy against hepatic and other forms of I/R injury by attenuating endothelial cell activation/inflammatory response, chemotaxis of inflammatory cells, rolling and adhesion of inflammatory cells to the endothelium, transendothelial migration, adhesion to the parenchymal cells and activation, and interrelated oxidative/nitrosative stress/inflammatory response (Figure 1).

Figure 1.

Mechanisms of CB2 receptor-dependent protection in ischaemia/reperfusion (I/R). CB2 receptor agonists may protect against I/R injury by attenuating endothelial cell activation/inflammatory response, chemotaxis of inflammatory cells, rolling and adhesion of inflammatory cells to the endothelium, transendothelial migration, adhesion to the parenchymal cells and activation, and interrelated oxidative/nitrosative stress/inflammatory response.

CB2 receptor is also detectable in human coronary artery endothelial cells by western blotting, reverse transcription-PCR, real-time PCR and immunofluorescence staining, where its activation by JWH133 or HU-308 attenuates TNF-α-induced nuclear factor-κB and RhoA activation, upregulation of adhesion molecules ICAM-1 and vascular cell adhesion molecule-1, increased expression of monocyte chemoattractant protein, enhanced transendothelial migration of monocytes and augmented monocyte–endothelial adhesion (Rajesh et al., 2007a). CB2 agonists also decreased the TNF-α- and/or endotoxin-induced ICAM-1 and vascular cell adhesion molecule-1 expression in isolated aortas and the adhesion of monocytes to aortic vascular endothelium (Rajesh et al., 2007a). Since the above-mentioned TNF-α- and endotoxin-induced phenotypic changes are critical in the initiation and progression of atherosclerosis, these findings suggest that targeting CB2 receptors on endothelial cells may explain, at least in part, the previously observed beneficial effects of THC in a mouse model of atherosclerosis (Steffens et al., 2005).

Endocannabinoids in I/R: sources, triggers and roles

Previous pioneering studies hypothesized that circulating activated macrophages and platelets are the pivotal sources of endocannabinoids during haemorrhagic shock (Wagner et al., 1997), endotoxemia (Varga et al., 1998), myocardial infarction (Wagner et al., 2001) or liver cirrhosis (Batkai et al., 2001) both in experimental animals and in humans. When isolated and injected into normal rats, these activated cells elicited SR141716-sensitive hypotension, also pointing towards the involvement of CB1 receptors in many of these conditions (Wagner et al., 1997; Varga et al., 1998; Batkai et al., 2001; Maccarrone et al., 2001; Ros et al., 2002; Liu et al., 2003).

Several studies have reported increased endocannabinoid levels following cerebral (Schmid et al., 1995; Panikashvili et al., 2001; Schabitz et al., 2002; Berger et al., 2004; Muthian et al., 2004) and hepatic and myocardial I/R injury (Wagner et al., 2001; Kurabayashi et al., 2005; Batkai et al., 2007b); however, the role of endocannabinoids and their sources in I/R injury remains to be a very controversial issue requiring further clarification. Interestingly, a recent study using a rat model of high intraocular pressure-induced retinal I/R found enhanced fatty acid amide hydrolase (FAAH) activity and downregulation of CB1 and transient receptor potential vanilloid-1 (TRPV1) receptors following I/R (Nucci et al., 2007). The I/R-induced cell death was attenuated either by the FAAH inhibitor URB597 or by the AEA stable analogue methanandamide (MetAEA), suggesting that endogenous AEA tone may play a protective role against injury (Nucci et al., 2007).

In a recent study, we attempted to identify cellular sources and triggers of endocannabinoid production using a mouse model of in vivo hepatic I/R (Batkai et al., 2007b). We found that I/R, but not ischaemia alone, triggered several-fold increases in the hepatic levels of the endocannabinoids AEA and 2-AG, which originated from hepatocytes, Kupffer and endothelial cells. Furthermore, these increases were positively correlated with the degree of tissue damage and serum TNF-α, MIP-1α and MIP-2 levels. Consistently, brief exposure of primary hepatocytes to various oxidants (H2O2, peroxynitrite) or inflammatory stimuli (TNF-α, endotoxin) (Pacher et al., 2006b, 2007) triggered marked increases in cellular endocannabinoid levels. Therefore, the important conclusions of this study are that not only inflammatory stimuli (for example, endotoxin) but also oxidative/nitrosative stress can modulate endocannabinoid levels in hepatocytes, and most likely in most other cell types too. The latter is also supported by recent findings demonstrating that the commonly used chemotherapeutic agent doxorubicin, which is known to mediate its cardiotoxicity by triggering oxidative/nitrosative stress (Pacher et al., 2003), increased endocannabinoid levels both in the myocardium in vivo and in cardiomyocytes in vitro (Mukhopadhyay et al., 2007). Similarly, up to sixfold increase in endocannabinoid AEA and/or 2-AG levels was observed in the hearts and livers of cirrhotic rats (notable cirrhotic cardiomyopathy is not associated with inflammatory cell infiltration of the myocardium; Batkai et al., 2007a). Therefore, parenchymal cells may also represent a very significant source of endocannabinoids produced in various pathological conditions associated with increased inflammation and/or oxidative tissue injury, in addition to the previously reported activated macrophages (Pacher et al., 2006a). The evidence on changes and possible regulation of endocannabinoid levels in various diseases was recently a subject of excellent overviews (Pertwee, 2005b; Di Marzo and Petrosino, 2007).

Our findings also imply that I/R-induced activation of hepatic endocannabinoids may limit hepatic injury by modulating the expression of adhesion molecules and the infiltration and activation of inflammatory cells by CB2-dependent/-independent mechanisms, which is also consistent with the emerging role of CB2 receptors in regulating microglial cell function and neuroinflammation (Walter and Stella, 2004; Maresz et al., 2005). Both mononuclear and polymorphonuclear leukocytes are known to express CB2 receptors (Klein, 2005; Pacher et al., 2006b), which could be activated on these infiltrating cells through a paracrine mechanism by endocannabinoids generated in and released from the various cell types in the liver. It is noteworthy that the endocannabinoid AEA can promote stellate cell and hepatocyte apoptosis in vitro by a mechanism not related to CB receptors (Siegmund et al., 2005, 2006).

Conclusions, future directions

There is a marked increase of endocannabinoid production in various forms of I/R (myocardial, cerebral, hepatic and circulatory shock), which correlate with the degree of tissue injury and inflammation, and may originate from virtually any cell type involved (Figure 2). Both CB1 agonists and antagonists may exert various neuroprotective effects against cerebral I/R, the specificity of which should further be studied by using knockout mice. Direct measurements should also confirm the increase of target tissue endocannabinoid levels following preconditioning (most recent evidence is based on assumptions of studies using pharmacological ligands), and experiments using knockout mice should determine the involvement of CB1/CB2 receptors. The latter is particularly important, since large body of evidence supporting the idea that CBs may mediate responses via interaction with other sites that probably represent novel CB receptor subtypes such as the putative endothelial CB receptor and GPR55 (Begg et al., 2005; Mackie and Stella, 2006; Hiley and Kaup, 2007; Johns et al., 2007), and that endocannabinoids may also behave as agonists of TRPV1 receptors under certain conditions and are substrates for COX leading the generation of various biologically active metabolites (Pacher et al., 2005b).

Figure 2.

Triggers and sources of endocannabinoids during I/R, and effects mediated via their receptors with potential relevance to pathophysiology. I/R, ischaemia/reperfusion.

Accumulating recent evidence suggests that selective CB2 agonists may protect against myocardial, cerebral and hepatic I/R injuries by decreasing the endothelial cell activation/inflammatory response, the expression of adhesion molecules, inflammatory cytokines/chemokines levels, recruitment, adhesion and activation of inflammatory cells, and interrelated oxidative/nitrosative stress. There is considerable interest in the development of selective CB2 receptor agonists, which are devoid of psychoactive properties of CB1 agonists, for various inflammatory disorders. Further studies should also establish the therapeutic window of protection during the reperfusion phase with the currently available CB2 receptor agonists, and new compounds should also be designed with better in vivo bioavailability, to devise clinically relevant treatment strategies against various forms of I/R. Nevertheless, the recently observed beneficial effects of CB2 receptor agonists in hepatic and other forms of I/R, coupled with the absence of psychoactive properties, and antifibrotic effects of CB2 receptor in the liver suggest that this approach may represent a novel promising strategy against various forms of I/R injury and other inflammatory disorders.


This work was supported by the Intramural Research Program of NIH/NIAAA (PP).

Conflict of interest

The authors state no conflict of interest.