Abstract

Introduction

(−)-Cannabidiol (CBD) is the major nonpsychotropic cannabinoid compound derived from the plant Cannabis sativa, commonly known as marijuana. CBD was first isolated in 1940 and its structure and stereochemistry determined in 1963 [1,2]. Interest in exploiting CBD therapeutically was initially focused on its interactions with the primary psychotropic ingredient of Cannabis, Δ⁹-THC and its sedative and antiepileptic effects, and later its antipsychotic and anxiolytic actions and utility in treating movement disorders [3]. As chronicled elsewhere [3], the last several years have seen a renewed interest in CBD due to the discovery of its antioxidative, anti-inflammatory, and neuroprotective effects, actions that occur for the most part independently of the canonical cannabinoid CB₁ and CB₂ receptors [1,4]. CBD may prove to have therapeutic utility in a number of conditions involving both inflammation and oxidative stress, including Parkinson’s disease, diabetes, rheumatoid arthritis, Alzheimer’s disease, and ischemia-reperfusion injury.

The contribution of the endocannabinoid system to inflammation and regulation of the immune system is an area of intense study that is beyond the scope of this article, and the reader is referred to several recent excellent reviews [5–8]. However, a brief overview of the system is helpful in discussing CBD. The endocannabinoid system comprises the following: (1) the G protein-coupled cannabinoid receptors CB₁ and CB₂, which are located in both the central nervous system and periphery; (2) their arachidonate-based lipid ligands, e.g., 2-arachidonoylglycerol (2-AG) and anandamide (N-arachidonoylethanolamine, AEA) and (3) the enzymes that synthesize and degrade these ligands. The endocannabinoid system plays a role in a variety of physiological processes including appetite, pain sensation, and mood. Evidence indicates that both CB₁ and CB₂ are expressed by cells of the immune system and are upregulated in the activation state. Levels of CB₂ appear to be higher than those of CB₁ with decreasing amounts of CB₂ in human B cells, NK cells, monocytes, polymorphonuclear neutrophils and T cells [6]. Macrophages and related cells, microglia and osteoclasts, express both cannabinoid receptors. CB₂ activation of immune cells is associated with changes in cytokine release and migration [6].

Biochemistry of Cannabidiol

CBD (Fig. 1) is a resorcinol-based compound that was shown to have direct, potent antioxidant properties by cyclic voltammetery and a spectrophotometric assay of oxidation in a Fenton reaction [9]. In an in vitro glutamate neuronal toxicity model, CBD was shown to be more protective than either α-tocopherol or vitamin C and comparable to butylated hydroxytoluene (BHT); although as noted by the authors, CBD unlike BHT does not seem to promote tumors [9]. CBD was also reported to act as an antioxidant at submicromolar concentrations in preventing serum-deprived cell death of cultured human B lymphoblastoid and mouse fibroblasts cells [10]. The antioxidant chemistry of CBD may have utility in vivo as well. The protective effects of CBD in a rat binge ethanol-induced brain injury model [11] and a rat model of Parkinson’s disease [12] were ascribed to its antioxidant properties. As will become clear from this review, however, the anti-oxidant actions ascribed to CBD in various in vivo models of human diseases likely exceed those attributable to its chemistry alone. Rather, the therapeutic anti-oxidant properties of CBD would seem to result in no small measure from its modulation of cell signaling events that underlie the self-sustaining cycle of inflammation and oxidative stress.

Figure 1

Chemical structure of cannabidiol (CBD).

Mechanisms of Action

Several interactions with relevance to the immune system and oxidative stress are discussed here. First, despite having low affinity for CB₁ and CB₂ receptors, CBD has been shown to antagonize the actions of cannabinoid CB₁/CB₂ receptor agonists in the low nanomolar range, consistent with non-competitive inhibition [13]. At 1–10 µM, CBD appears to function as an inverse agonist at both CB₁ and CB₂ receptors [13]. Second, CBD acts as an inhibitor (IC₅₀ = 28 µM) of fatty acid amide hydrolase (FAAH), the major enzyme for endocannabinoid breakdown. Because FAAH activity correlates with gastrointestinal mobility, CBD may have utility in treating intestinal hypermotility associated with certain inflammatory diseases of the bowel [14].

Third, CBD is a competitive inhibitor with an IC₅₀ in the nanomolar range of adenosine uptake by the equilibrative nucleoside transporter 1 (ENT1) of macrophages and microglial cells, the resident macrophage-like immune cells of the brain. By increasing exogenous adenosine, which in turn activates the A_2A adenosine receptor, CBD exerts immunosuppressive actions on macrophages and microglial cells as evidenced by decreased TNFα production after treatment with lipopolysaccharide (LPS) [15,16]. CBD may thus be of benefit in treating neueodegenerative diseases associated with hyperactivation of microglial, as well as retinal neuroinflammation seen in such conditions as uveitis, diabetic retinopathy, age-related macular degeneration, and glaucoma. Note, however, that adenosine activates other receptors besides A_2A that often have opposing consequences on immune regulation and inflammation [17,18]. In several in vivo models of neurodegeneration or inflammation, moreover, the beneficial effects of CBD were demonstrated not to involve adenosine receptors.

Fourth, CBD has been shown to have potent actions in attenuating oxidative and nitrosative stress in several human disease models, although the exact mechanism is unclear. For instance, CBD pretreatment was found to attenuate high glucose-induced mitochondrial superoxide generation and NF-κB activation in human coronary artery endothelial cells, along with nitrotyrosine formation and expression of inducible nitric oxide synthase (iNOS) and adhesion molecules ICAM-1 and VCAM-1 [19]. Notably, high glucose-induced transendothelial migration of monocytes, monocyte-endothelial adhesion, and barrier disruption were attenuated as well. These findings lend support to the conclusion that that CBD may have therapeutic utility in treating diabetic complications and atherosclerosis. In another study, CBD was reported to reduce expression of reactive oxygen species (ROS) generating NADPH oxidases, as well as iNOS and nitrotyrosine generation in a cisplatin nephropathy model in vivo, consequently lessening cell death in the kidney and improving renal function [20]. From these studies, it is tempting to speculate that CBD may act directly at the level of the mitochondrion or nucleus to oppose oxidative/nitrosative stress.

Fifth, at low micromolar concentrations, CBD was found to inhibit indoleamine-2,3-dioxygenase activity thereby suppressing tryptophan degradation by mitogen-stimulated peripheral blood mononuclear cells and LPS-stimulated myelomonocytic THP-1 cells in vitro [21]. Based on this finding, CBD might be useful therapeutically to counter the increased risk of depression in diseases associated with immune activation and inflammation, which often lead to decreased tryptophan, the precursor of serotonin. Finally, CBD has been shown to act as an antagonist at G protein-coupled receptor 55 (GPR55) and as an antagonist or agonist at several transient receptor potential (TRP) channels; however, these observations are controversial and the pharmacophysiological significance of these interactions is not known [1,4].

Actions on Immune Cells

CBD has been shown to modulate the function of the immune system. Overall these actions may be nuanced and concentration-dependent, but in general include suppression of both cell-mediated and humoral immunity and involve inhibition of proliferation, maturation, and migration of immune cells, antigen presentation, and humoral response [1,13]. Key aspects are discussed here. In most in vivo models of inflammation, CBD attenuates inflammatory cell migration/infiltration (e.g. neutrophils) [22]. During neuroinflammation, activated microglial cells migrate towards the site of injury where they release pro-inflammatory cytokines and cytotoxic agents, including ROS. Although important in removal of cellular debris and fighting infection, activated microglial cells often exacerbate local cell damage. CBD was shown to inhibit activated microglial cell migration by antagonizing the abnormal-cannabidiol (Abn-CBD)-sensitive receptor at concentrations < 1 µM [23]. Evidence that the Abn-CBD receptor is the orphan G protein-coupled receptor GPR18 was recently reported [24]. CBD was also shown to block endotoxin-induced oxidative stress resulting from retinal microglial cell activation in uveitis [25]. CBD blocked the immediate activation of NADPH oxidase as well as a second wave of ROS formation and the associated TNFα secretion and p38 MAPK activation. The direct antioxidant property of CBD is unlikely to be the entire explanation for these actions as they occurred at a concentration of 1 µM. Inhibition of adenosine uptake as discussed previously may have been involved. However, a complete understanding of the anti-inflammatory actions of CBD on microglial cells is not yet available. Recently, through an unidentified mechanism, CBD was reported to suppress LPS-induced pro-inflammatory signaling in cultured microglial cells, including NF-κB and STAT1 activation, while enhancing STAT3-related anti-inflammatory signaling [26].

CBD induces apoptosis of monocytes and certain normal and transformed lymphocytes, including thymocytes and splenocytes, through oxidative stress and increased ROS levels [27–31]. The basis for this action appears to be glutathione depletion due to adduct formation with the reactive metabolite of CBD, cannabidiol hydroxyquinone, thereby triggering cell death through caspase 8 activation and/or the intrinsic apoptotic pathway. Increased ROS from the upregulation of NADPH oxidases via an undefined mechanism may contribute to cell death as well [31]. A recent study assessed the impact of repeated administration of relatively low levels of CBD to adult male Wistar rats on peripheral blood lymphocyte subset distribution [32]. At 2.5 mg/kg/day for 14 days, CBD did not produce lymphopenia, but increased the total number of natural killer T (NKT) cells and percentage numbers of NKT and natural killer (NK) cells. A dose of 5 mg/kg/day did have a lymphopenic effect, but by reducing B, T, Tc and Th lymphocytes. Thus, CBD would appear to suppress specific immunity, while enhancing nonspecific antitumor and antiviral immune response. As discussed by the authors [32], the lymphopenic effect of CBD was observed at a concentration shown to be efficacious in a number of animal models of neurodegenerative and inflammatory diseases, including blocking the progression of collagen-induced arthritis in a murine model of rheumatoid arthritis, decreasing damage to pancreatic islets in the NOD mouse model of type 1 diabetes, lessening hyperalgesia in rat models of neuropathic and inflammatory pain, and preventing cerebral ischemia in gerbils.

Pain

Neuropathic pain is associated with microglia activation in the spinal cord and brain and their subsequent release of pro-inflammatory cytokines, such as interleukin-6 (IL-6), interleukin-1β (IL-1β), and tumor necrosis factor-α (TNFα) [33]. The etiology of neuropathic pain, which is common in cancer, diabetes, multiple sclerosis, and peripheral nerve injury, is poorly understood, but recent evidence indicates that increased ROS generation by microglial cells is the critical initiating factor [34]. The drug Sativex, which consists of Δ⁹-THC and CBD, is approved in several countries for treatment of central and peripheral neuropathic pain and for spasticity associated with multiple sclerosis [35]. In a mouse model of type I diabetic peripheral neuropathic pain, intranasal or intraperitoneal administration of a moderate-high dose of CBD attenuated tactile allodynia and thermal hypersensitivity without affecting the diabetic state [36]. The antinociceptive effects of CBD were associated with less of an increase in microglial density and p38 MAPK activity in the dorsal spinal cord. Finally, the anti-inflammatory and immunosuppressive actions of CBD may be of use in treating rheumatoid arthritis and the associated pain [37,38].

Diabetes and Diabetic Complications

CBD was shown to reduce either the initiation of diabetes or the development of overt or latent diabetes in non-obese diabetes-prone (NOD) mice by reducing insulitis [39,40]. This action was accompanied by a shift in the immune response from a dominant Th1 pattern with pro-inflammatory cytokines to a Th2 pattern with increased levels of the anti-inflammatory cytokine IL-10. Major effectors of β-cell death in type 1 diabetes are various free radicals and oxidant species, including NO, and infiltrating macrophages are one source of high concentrations of NO and inflammatory cytokines that further enhance NO and ROS formation [41]. CBD was also shown to be effective in blocking ROS-induced up-regulation of surface adhesion molecules on endothelial cells due to high glucose and in preserving endothelial barrier function [19,42]. Adhesion of monocytes followed by their transmigration into the subendothelial space is an early event in atherosclerosis, the most common macrovascular complication of diabetes, and may contribute as well to diabetic retinopathy [19,42,43]. The anti-inflammatory actions of CBD may also protect retinal neurons in diabetes by attenuating activation and ROS generation by Müller glia, thus preventing tyrosine nitration and inhibition of Müller cell glutamine synthetase and the consequent accumulation of glutamate, which in turn leads to oxidative stress-induced death of retinal neuronal cells [44].

In a mouse model of type I diabetic cardiomyopathy, both pre- and post-treatment with CBD attenuated cardiac fibrosis and cell death, myocardial dysfunction, inflammation, oxidative/nitrosative stress, and the activation of related signaling pathways [45]. CBD attenuated diabetes-induced activation in the heart of the key pro-inflammatory transcription factor, NF-κB and its consequences, e.g. expression of ICAM-1, iNOS, VCAM-1, and TNFα. These observations underscore the point that CBD likely attenuates inflammation far beyond its antioxidant properties per se. CBD also reduced high glucose-induced increases in both cytosolic and mitochondrial reactive oxygen and nitrogen species generation in primary human cardiac myocytes, which was accompanied by reduced NF-κB activation and cell death. These findings indicate that CBD may have great therapeutic potential in alleviating cardiac complications of diabetes.

Hypertension

Although CBD has not been considered for treating hypertension, a parallel between the role of microglia in diabetes and hypertension deserves mention. Activation of microglia within the paraventricular nucleus (PVN) was recently shown to contribute to neurogenic hypertension resulting from chronic angiotensin II infusion in the rat [46]. Microglia activation was associated with enhanced expression of pro-inflammatory cytokines, the acute administration of which into the left ventricle or PVN resulted in increased blood pressure. The hypertensive action of angiotensin II infusion could be blocked by overexpression of IL-10 in the PVN or intracerebroventricular infusion of minocycline, supporting the involvement of ROS.

The immune system contributes as well to systemic endothelial dysfunction observed in hypertension [47]. Local production of angiotensin II by activated leukocytes within the vessel wall is thought to reduce endothelial function and NO production, leading to attenuated vasodilation and increased blood pressure, through the production of inflammatory cytokines and ROS [48,49]. Interestingly, recent evidence has shown that the initial stimulus for peripheral leukocyte activation in angiogensin II-induced hypertension is the increase in blood pressure that results from stimulation of cells within the anteroventral third ventricle (AV3V) of the brain by angiotensin II [50].

Ischemia Reperfusion Injury

Redox stress and ROS produced by ischemia-reperfusion of organs activates the immune system, which aids in repair by removing debris and stimulating remodeling. An excessive or prolonged inflammatory response, however, may prove detrimental to organ function by exacerbating ROS production and causing death of the parenchyma. Several hours after ischemia-reperfusion in the heart, a model of myocardial infarction, neutrophils accumulate in the myocardium [51]. Several lines of evidence suggest that this accumulation of neutrophils worsens injury to the myocardium [51]. In rats, treatment with CBD for 7 days following a 30 minute occlusion of the left anterior descending coronary artery markedly reduced infarct size, myocardial inflammation and IL-6 levels and preserved cardiac function [52]. In addition, the number of leukocytes infiltrating the border of the infarcted area was dramatically reduced. CBD has been shown to inhibit stimulated migration of neutrophils [22]. CBD treatment was also recently shown to reduce neutrophil migration in a rat model of periodontitis [53]. Hyperactive neutrophils exacerbate periodontal tissue injury and lead to tooth loss in part by excessive ROS formation in individuals with refractory periodontitis [54]. Finally, pre- or post-ischemic treatment with CBD was shown to have a prolonged and potent protective action in cerebral ischemia. The neuroprotective actions of CBD were attributed to reduced neutrophil accumulation and myeloperoxidase activity [55], as well as decreased high mobility group box 1 (HMGB1) expression by microglia [56].

Depression

CBD is reported to have anti-depressive actions, the basis for which is not established although activation of 5-HT_1A receptors may be involved at least at higher concentrations [13,57,58]. Growing evidence in recent years has implicated pro-inflammatory cytokines, free radical species, and oxidants in the etiology of depression [59,60]. One explanation is that the resultant oxidative stress adversely affects glial cell function and leads to neuron damage in the brain.

Neurodegenerative Diseases

Microglial hyperactivation is a common feature of a number of neurodegenerative diseases, including Parkinson’s, Alzheimer’s, Huntington’s, amyotrophic lateral sclerosis (ALS) and multiple sclerosis(MS) [61,62]. Activated microglia produce a number of pro- and anti-inflammatory cytokines, chemokines, glutamate, neurotrophic factors, prostanoids, and a variety of free radicals that together create a state of oxidative stress. Alzheimer’s disease, which is the most common form of dementia, is characterized by the deposition of “senile” plaques that are sites of microglia activation and inflammation. The resultant oxidative stress is a critical factor in the pathophysiology of Alzheimer’s [63]. The plaques are composed of insoluble aggregates of the beta-amyloid peptide (Aβ), which self-assembles as monomers, oligomers and finally fibrils. Recent evidence shows that the oligomeric form of beta-amyloid is the most neurotoxic species and is most effective as a chemotactic agent for microglia and stimulator of microglial oxidative stress [61,64]. Activated microglia are a major contributor of inflammatory factors in Alzheimer’s and secret a number of pro-inflammatory cytokines, which ironically further enhance Aβ production by neuronal cells [65]. In addition, an inflammatory state was shown to block the ability of microglia to phagocytize fibrillar Aβ [66]. Aging was also shown to negatively impact on the ability of microglia to internalize Aβ [67]. Microglia from aged mice were also shown to be less responsive to stimulation and to secrete greater amounts of IL-6 and TNFα compared to microglia of younger mice. Aged microglia also had lower levels of glutathione, suggesting an increased susceptibility to the harmful effects of oxidative stress. Finally, although controversial, evidence has been put forward suggesting that bone marrow-derived monocytic cells may somehow gain access to the diseased brain in Alzheimer’s and be better at phagocytosing amalyoid plaques than resident microglia [65,68].

Based on rather scant evidence, some have proposed that CBD might have utility in treating neurodegenerative diseases [1,3,69–71]. CBD was shown to have a protective effect on cultured rat pheocromocytoma PC12 cells exposed to Aβ [72,73]. In a concentration-dependent manner, CBD increased cell survival, while decreasing ROS and nitrite production, lipid peroxidation, and iNOS protein expression. CBD was shown to have anti-inflammatory actions in vivo in a mouse model of Alzheimer’s neuroinflammation induced by injection of human Aβ into the hippocampus. CBD dose-dependently attenuated Aβ-induced glial fibrillary acidic protein (GFAP) mRNA, iNOS and IL-1β protein expression, and NO and IL-1β release [74]. In a recent study, CBD was found to protect against amphetamine-induced oxidative protein damage in a rat model of mania and to increase brain-derived neurotrophic factor (BDNF) expression levels in the reversal protocol [75]. Results of these preclinical studies are persuasive and support the need for double-blind placebo controlled trials to assess the therapeutic utility of CBD in patients with neurodegenerative diseases.

Obesity and the Metabolic Syndrome

Metabolic syndrome is a combination of medical disturbances including central obesity, glucose intolerance, hypertension, and dyslipidemia that increases the risk for developing cardiovascular diseases and type 2 diabetes. Adipocyte dysfunction leading to a low-grade chronic inflammatory state is thought to underpin the etiology of the metabolic syndrome [76]. Metabolic overload of adipocytes causes production of ROS, pro-inflammatory cytokines, and adipokines that activate inflammatory genes and stress kinases and interfere with insulin signaling [76,77]. Saturated fatty acids also activate toll-like receptors on adipocytes and macrophages, components of the innate immune system, to induce production of proinflammatory cytokines and chemokines. Enhanced mitochondrial flux together with relative hypoxia due to adipocyte tissue hypertrophy, endothelial cell apoptosis, and inflammation-impaired angiogenesis further enhances ROS generation. Enhanced rupture of adipocytes due to excessive hypertrophy attracts and activates macrophages that further exacerbate the inflammatory state through the production of inflammatory cytokines and ROS. The chronic inflammatory state compromises the ability of adipose tissue to absorb incoming fat leading to fat build up in other organs, including liver, heart, and skeletal muscle, and creating a local inflammatory state that progresses to insulin resistance in those organs as well. Increased ROS levels are thought to be the major contributing factor to insulin resistance [78,79].

Macrophages, both resident and to a greater extent bone marrow-derived, play a critical role in initiating adipose tissue dysregulation and inflammation in the metabolic syndrome and together with adipocytes constitute a paracrine loop that sustains the chronic inflammatory state [80]. Macrophages secrete TNFα which acts on hypertrophied adipocytes to downregulate adiponectin and induce pro-inflammatory cytokines and lipolysis. The released free fatty acids act in turn on the toll-like receptor 4 (TLR4) of macrophages to induce production of pro-inflammatory cytokines, including TNFα. Both macrophages and adipocytes secrete monocyte chemotactic protein 1 (MCP1), which serves to recruit more macrophages to the adipose tissue.

Recent evidence has revealed that most macrophages in obese adipose tissue are polarized towards the M1 or classically activated, pro-inflammatory state, as opposed to the M2 or alternatively activated, anti-inflammatory state [80,81]. Th1 cytokine interferon gamma (IFNγ), microbial byproducts (e.g., LPS), and free fatty acids from visceral adipose tissue promote polarization towards the M1 state, whereas Th2 cytokines IL-4 and IL-13 promote polarization towards the M2 phenotype. Ligand-dependent transcription factors peroxisome proliferator activated receptors (PPARs) play a key role in determining the M1/M2 phenotype [81,82]. Activation of PPARγ or PPARδ promotes differentiation towards the M2 phenotype, while PPARγ activation inhibits M2 to M1 phenotype switch and represses the M1 pro-inflammatory gene expression profile. Of interest, CBD as well as some other cannabinoids, has been shown to activate PPARγ, possibly through direct binding [83,84]. Although tonic activation of CB₁ receptors by endocannabinoids is implicated in the development of abdominal obesity and CB₁ antagonists and inverse agonist reduce obesity, their clinical use is problematic due to serious neuropsychiatric effects [85]. Given its anti-inflammatory actions and PPARγ agonism, CBD might serve as the basis for design of a new anti-obesity drug [1]. In this regard, a cautionary note regarding the PPARγ agonism associated with CBD should be sounded, although this was observed only in very high concentrations and only in vitro, which is that several PPARγ agonists have been retracted because of various problems [86,87].

Atherosclerosis

Atherosclerosis is an inflammatory disease in which monocytes/macrophages play a critical role in the initiation and progression, as well as rapture, of the atherosclerotic plaque [88]. Plaques form in the arterial wall at areas of disturbed flow and endothelial dysfunction (Fig. 2). The initiating event is the transcytosis of low density lipoprotein (LDL) into the subendothelial space where it is trapped by binding to proteoglycans of the extracellular matrix [88,89]. LDL is oxidized by various cells including macrophages, first to minimally modified LDL (mmLDL) and then extensively oxidized LDL (oxLDL). The former activates endothelial cells to secrete various factors that attract monocytes and to express adhesion molecules that support the binding and transmigration of monocytes into the subendothelial space. Once there, monocytes differentiate into macrophages under the influence of cytokines and oxLDL. Macrophages take up oxLDL and differentiate into foam cells that secrete a number of cytokines and growth factors that sustain the inflammatory response and stimulate migration of smooth muscle and endothelial cells into the intima. Continued oxLDL uptake by foam cells combined with impaired cholesterol efflux results in their apoptosis and exposure of thrombogenic lipids [88,89]. A number of events in monocyte/macrophage physiology may be potential therapeutic targets for dealing with atherosclerosis and are discussed in detail elsewhere [88,89].

Figure 2

Inflammation and oxidative stress in atherosclerotic plaque formation. Endothelial dysfunction causes monocyte activation and their binding to endothelial cells, via the production of MCP-1, its binding to CCR2 receptors, and the upregulation of adhesion molecules on endothelial cells (1). Monocytes cross the endothelium and differentiate into macrophages (2). Due to ROS, LDL that traverses the endothelium is converted to mmLDL and oxLDL. Macrophages accumulate oxLDL through scavenger receptors and are turned into foam cells (3). Along with T cells, foam cells produce inflammatory mediators that stimulate migration of smooth muscle and endothelial cells into the intima (4). Figure taken with permission from Reference 88.

ROS play a pivotal role in atheroma development and macrophages are the major source for ROS with NADPH oxidase, cyclooxygenases (COX), lipooxygenases (LOX), iNOS, and myeloperoxidase contributing [88,89]. ROS participate in atherosclerosis in part by causing LDL oxidation, activating stress signaling pathways, inducing apoptosis, and facilitating plaque rupture [88]. Based on their ability to inhibit 15-LOX, CBD and its mono- and dimethylated derivatives have been proposed as potentially useful in treating atherosclerosis [90]; however, the question of whether 15-LOX has a detrimental or beneficial role in atherosclerosis is unsettled [91]. Nevertheless, a growing body of evidence supports the utility of targeting endocannabinoid signaling, particularly that of macrophages, in the treatment of atherosclerosis [92]. Differentiation of human monocytes, including that induced by oxLDL, results in a change in their CB₁ and CB₂ expression profile such that CB₁ becomes more prominent [93]. Activation of macrophage CB₁ receptor was shown to upregulate the CD36 scavenger receptor and cholesterol accumulation by macrophages/foam cells [94]. CB₁ receptor activation of human macrophages was linked to ROS generation via p38 MAPK activation, as well as production of TNFα and MCP1 [93]. In contrast, activation of the CB₂ receptor was shown to attenuate the pro-inflammatory actions of the CB₁ receptor through activation of the Ras family small G protein, Rap1 [93]. Consistent with these findings, a nonselective CB₁/CB₂ receptor agonist reduced oxLDL-induced ROS generation and TNFα secretion via the CB₂ receptor of murine macrophage, which in contrast to human macrophages do not express much CB₁ receptor [93,95]. Such tantalizing findings have fueled the idea that the endocannabinoid system may be a avenue for further drug development in dealing with atherosclerosis, likely involving a role for CBD as well [7].

Opposing regulatory effects of CB₁ and CB₂ receptors on inflammation and oxidative/nitrative stress is a general theme that has significance in atherosclerosis, as well as other human maladies. CB₂ activation in endothelial cells, which play a key role in development of early atherosclerosis and any inflammatory response, decreases activation and the inflammatory response [96], while CB₁ activation in human coronary artery endothelial cells was reported to induce ROS-dependent and -independent MAPK activation and cell death [97]. CB₁ cannabinoid receptors promote oxidative stress and cell death in murine models of doxorubicin-induced cardiomyopathy and in human cardiac myocytes [98]. In contrast, CB₂ activation was found to reduce oxidative stress and neutrophil infiltration in the infracted mouse myocardium [99]. In nephropathy, CB₂ limits oxidative/nitrosative stress, inflammation, and cell death [100], while activation of CB₁ cannabinoid receptors promote oxidative/nitrosative stress, inflammation, and cell death [101].

Conclusions

Inflammation and oxidative stress are intimately involved in the genesis of many human diseases. Unraveling that relationship therapeutically has proven challenging, in part because inflammation and oxidative stress “feed off” each other. However, CBD would seem to be a promising starting point for further drug development given its anti-oxidant (although relatively modest) and anti-inflammatory actions on immune cells, such as macrophages and microglia. CBD also has the advantage of not having psychotropic side effects. Studies on models of human diseases support the idea that CBD attenuates inflammation far beyond its antioxidant properties, for example, by targeting inflammation-related intracellular signaling events. The details on how CBD targets inflammatory signaling remain to be defined. The therapeutic utility of CBD is a relatively new area of investigation that portends new discoveries on the interplay between inflammation and oxidative stress, a relationship that underlies tissue and organ damage in many human diseases.

Abstract

1. Introduction

It is well-established that Δ′-tetrahydrocannabinol (Δ′-THC), the major psychoactive constituent in the medicinal/drug-type variety of the cannabis plant, displays a number of biological activities including anti-atherosclerotic, anti-proliferative and endocrine disrupting properties (Steffens et al., 2005; Watanabe et al., 2005; Takeda et al., 2008a, 2009a, 2011a; Guindon and Hohmann, 2011). There is also growing experimental evidence suggesting that cannabidiol (CBD), found principally in the fiber-type of cannabis, also possesses biological activities. For example, CBD has been reported as an inhibitor of human glioma cell migration and an inducer of programmed cell death in tumor cells, including breast cancer cells (Vaccani et al., 2005; Ligresti et al., 2006; Shrivastava et al., 2011). In fresh plants, the concentrations of neutral cannabinoids, including CBD, are much lower than those of cannabinoid acids that contain a carboxyl group (–COOH) in the structures. CBD is produced largely through non-enzymatic decarboxylation of its acidic precursor, cannabidiolic acid (CBDA), during extraction from the leaf materials of the plant (see Fig. 1) (Yamauchi et al., 1967). Although studies on the biological efficacies of CBD are ongoing, there is a relative lack of information regarding the potential biological activities of CBDA. We have reported that CBDA is a selective cyclooxygenase-2 (COX-2) inhibitor (Takeda et al., 2008b, 2009a), and it has been reported by others that CBDA is a necrosis-inducing factor for the leaf cells of cannabis plant (Morimoto et al., 2007). However, no investigations have evaluated potential anti-migration effects of CBDA on human breast cancer cells. Among the available experimental human breast cancer cell lines, MDA-MB-231 cells have been established as a valuable mesenchymal breast cancer model (see Fig. 2A–b) for pre-clinical studies as they are highly aggressive, both in vitro and in vivo (Price et al., 1990; Rochefort et al., 2003; Takeda et al., 2011b).

Open in a separate window

Fig. 1

Chemical structures of CBDA and CBD. In the fiber-type cannabis plant, the concentration of CBD is much lower than that of its precursor CBDA. CBD is formed artificially from CBDA by non-enzymatic decarboxylation during extraction step (Yamauchi et al., 1967).

Open in a separate window

Fig. 2

Effect of CBDA on the vertical migration of highly aggressive human breast cancer MDA-MB-231 cells. (A) Morphologies of two human breast cancer cell lines; MCF-7 cells (a) and MDA-MB-231 cells (b). MCF-7 cells display epithelial morphology (collective) and MDA-MB-231 cells display single elongated morphology (mesenchymal). (B) MDA-MB-231 cells were exposed for 12 h to CBDA (5, 10, 25 μM) and CBD (5, 10, 25 μM). After the treatments, cell viability was measured according to the methods described in Section 2. Data are expressed as the percent of vehicle-treated group (indicated as Cont.), as mean ± S.D. (n = 6). *Significantly different (p < 0.05) from the vehicle-treated control. (C) Transwell migration assays (vertical migration) were performed to determine MDA-MB-231 cell migration 12 h after treatments with 5 μM, 10 μM, or 25 μM CBDA and 5 μM CBD. Data are expressed as the percent of vehicle-treated group (indicated as Cont.), as mean ± S.D. (n = 8). *Significantly different (p < 0.05) from the vehicle-treated control. N.D., not detectable due to complete inhibition of the migration.

It is well-recognized that Rho family small GTPases (~21 kDa) regulate actin cytoskeletal dynamics, thereby affecting multiple cellular functions including cell mobility and polarity (Jaffe and Hall, 2005; Yamazaki et al., 2005). In the GTP-bound state, Rho family small GTPases are active and able to interact with specific downstream effectors such as Rhotekin, which lead to the translocation and activation of the effectors, and induction of various intracellular responses. In these respects, altered Rho GTPase activity or expression is implicated in cancer progression (Ellenbroek and Collard, 2007). Among the Rho subfamily members that include the isoforms RhoA, RhoB, and RhoC, RhoA has specifically been identified as an inhibitor of cancer cell mobility, including breast cancer cells (Vial et al., 2003; Simpson et al., 2004; Vega et al., 2011). Post-translational regulation of Rho activity has been demonstrated for RhoA. This Rho protein is phosphorylated in vitro and in vivo by kinases such as a cAMP-dependent protein kinase (PKA) on serine 188 (Ser188) (Lang et al., 1996; Dong et al., 1998; Ellerbroek et al., 2003), and it is generally accepted that phosphorylation of this site is an important negative regulation of the RhoA activity, leading to termination of the signaling process.

In this study we demonstrate for the first time that (i) CBDA is an inhibitor of MDA-MB-231 breast cancer cell migration, and that (ii) the mechanism responsible for the inhibitory effects of CBDA likely involves activation of RhoA via inhibition of PKA.

2. Materials and methods

3. Results and discussion

3.1. Effects of CBDA on the migration of MDA-MB-231 cells

When compared with the epithelial-like morphology of MCF-7 cells that are capable of forming domes, with close contact between cells (collective, Fig. 2A–a), MDA-MB-231 cells exhibit a mesenchymal-like morphology (mesenchymal, Fig. 2A–b), a hallmark feature of tumor aggressiveness (Price et al., 1990; Rochefort et al., 2003; Jaffe and Hall, 2005; Yamazaki et al., 2005; Takeda et al., 2011b). Thus, we selected the use of MDA-MB-231 cells to investigate whether CBDA can inhibit the cell migration under the incubation period and its concentration that do not affect overall MDA-MB-231 cell growth (Fig. 2B, see also Fig. 6B). In this study we could not directly compare the efficacy of CBDA and CBD-mediated inhibition of the migration of MDA-MB-231 cells because of CBD’s strong cytostatic effect (see also Ligresti et al., 2006); 10 μM and 25 μM CBD treatment for 12 h significantly reduced MDA-MB-231 cell viability (Fig. 2B). Thus, we compared 5 μM or 25 μM CBDA treatments with that of 5 μM CBD. As shown in Figs. 2C, 3A–a/b and B, CBDA inhibited migration of MDA-MB-231 cells in a concentration-dependent manner, whereas 5 μM CBD exhibited no significant inhibitory effects on migration. Futher, CBDA’s significant abrogation of cell migration was detected even 48 h after its exposure (Fig. 3B). It is noteworthy that CBDA-mediated attenuation of MDA-MB-231 cell migration was more pronounced when assessing vertical migration versus horizontal migration. Although in this study we could not detect CBD-mediated inhibition of the migration of MDA-MB-231 cells, it has been reported that low concentrations of CBD (i.e., 0.1–1.5 μM), treated once a day up to 72 h, suppress cell migration (McAllister et al., 2007). The discrepancy between the latter report and the result presented here may be attributed to the difference in the experimental conditions employed. In short, our data strongly suggest that CBDA is an effective inhibitor of breast cancer cell migration in vitro.

Open in a separate window

Fig. 3

Effect of CBDA on the horizontal migration of MDA-MB-231 cells. (A) Wound-healing assays (horizontal migration) were performed to determine the effects of CBDA or CBD on MDA-MB-231 cell migration. (a) Representative images of the migrating cells were captured 12 h after vehicle (indicated as Cont.), 5 μM or 25 μM CBDA and 5 μM CBD treatments. (b) Migration data presented in panel (a) was assessed on the basis of percent wounded area filled in. Data are expressed as the percent of vehicle-treated group (indicated as Cont.), as mean ± S.D. (n = 8). (B) Migration data was assessed on the basis of percent wounded area filled in 12, 24, or 48 h after treatments with 25 μM CBDA. Data are expressed as the percent of vehicle-treated group (indicated as Cont.), as mean ± S.D. (n = 8) *Significantly different (p < 0.05) from the respective vehicle-treated controls.

Open in a separate window

Fig. 6

CBDA inhibits PKA activity. (A) (a) A representative photograph of phosphorylated bands and non-phosphorylated bands of PKA-specific substrate peptide in MDA-MB-231 cells incubated with vehicle (time 0, lane 3), vehicle (48 h, lane 4), and 25 μM CBDA (48 h, lane 5) was shown. For positive and negative controls, the reactions were performed in the presence (lane 1) or absence (lanes 2) of PKA catalytic subunit, respectively. An arrow in the image indicates wells that samples are applied. (b) The relative intensity of phosphorylated bands in MDA-MB-231 cells is shown. Data are expressed as fold change vs. non-CBDA-treated group (left panel; lanes 4 vs. 5), as mean ± S.D. (n = 3). *Significantly different (p < 0.05) from the vehicle-treated control. (B) MDA-MB-231 cells were treated with vehicle (indicated as Cont.) or 25 μM CBDA for 48, 72, or 96 h. After the treatments, cell viability was measured according to the methods described in Section 2. Data are expressed as the percent of vehicle-treated group, as mean ± S.D. (n = 6). *Significantly different (p < 0.05) from the vehicle-treated control. N.S., not significant. (C) A model of CBDA-mediated RhoA activation through PKA inhibition.

3.2. COX-2 inhibitory activity of CBDA is not critical in the attenuation of MDA-MB-231 cell migration

It has been reported that selective COX-2 inhibitors interfere with the migration of certain cancer cell types (Banu et al., 2007; Kim et al., 2010). Previously, we reported that CBDA is an effective COX-2 inhibitor (Takeda et al., 2008b, 2009a). Therefore, we initially hypothesized that CBDA may exert its inhibitory effects on MDA-MB-231 cell migration via COX-2 inhibition. However, as demonstrated by the data shown in Fig. 4, vertical migration of MDA-MB-231 cells was not influenced by 25 μM DuP-697 (an established selective COX-2 inhibitor), nor SC-560 (an established selective COX-1 inhibitor), although the same concentration of CBDA completely inhibited cell migration. In addition, horizontal migration was not affected by DuP-697 (data not shown). These results indicate that COX-2 activity is not an essential factor for the migration of MDA-MB-231 cells, and that other pathway(s) are likely to be involved in the anti-migration effects of CBDA.

Open in a separate window

Fig. 4

COX-2 inhibitory activity of CBDA is not essential to attenuate MDA-MB-231 cell migration. Transwell migration assays were performed to determine the effects of COX-2 selective inhibitors (25 μM CBDA or 25 DuP-697) and COX-1selective inhibitor (25 μM SC-560) on MDA-MB-231 cell migration 12 h after their respective treatments. Data are expressed as the percent of vehicle-treated group (indicated as Cont.), as mean ± S.D. (n = 8). *Significantly different (p < 0.05) from the vehicle-treated control. N.D., not detectable due to complete inhibition of the migration.

3.3. CBDA-mediated stimulation of RhoA activity

In an effort to investigate CBDA’s mode of action, we focused on the Rho family of small GTPases, in particular the isoform RhoA, since active RhoA has been reported to inhibit breast cancer cell mobility in vitro (Vial et al., 2003; Simpson et al., 2004; Vega et al., 2011). We discovered that the active RhoA levels were remarkably stimulated by 25 μM CBDA, especially at 48 h post-treatments (Fig. 5A and B). In accordance with the RhoA activation time-course, horizontal migration of MDA-MB-231 cells was significantly inhibited by 25 μM CBDA at 48 h (see Fig. 3B). GTPγS, a hydrolysis-resistant GTP analog, was used as positive control for the experiment (Fig. 5A, left portion). Because RhoA can be regulated by post-transcriptional modification via phosphorylation on Ser188, leading to inactivation of the protein (Lang et al., 1996; Dong et al., 1998; Ellerbroek et al., 2003), we also assessed the phosphorylation status of RhoA following 25 μM CBDA exposure. Although total RhoA levels were not changed, the phosphorylation levels at Ser188 on RhoA were reduced by CBDA in a time-dependent manner, as compared with vehicle-treated groups (Fig. 5A and B). Thus, there exists an apparently negative relationship between the levels of active RhoA and phosphorylated RhoA, suggesting that CBDA activates RhoA via an inhibition of its phosphorylation.

Open in a separate window

Fig. 5

CBDA stimulates RhoA activity. (A) RhoA affinity pull-down assays were used to determine the level of active RhoA according to the methods described in Section 2. Resulted pull-down samples were subjected to Western blot analyses using an anti-RhoA antibody. Active RhoA was increased by 25 μM CBDA (indicated as +) in a time-dependent manner. Western blot analyses were also performed using an anti-RhoA antibody specific to RhoA phosphorylated at Ser188 and a β-actin antibody. (B) Results are expressed as the ratio of active RhoA to total RhoA protein in each cell lysate. Data are expressed as the fold change vs. vehicle-treated group (indicated as −), as mean ± S.D. (n = 3). *Significantly different (p < 0.05) from the vehicle-treated control.

Acknowledgments

Abbreviations

Footnotes

References