1521-0103/351/2/457–466$25.00 THE JOURNAL OF PHARMACOLOGY AND EXPERIMENTAL THERAPEUTICS Copyright ª 2014 by The American Society for Pharmacology and Experimental Therapeutics

http://dx.doi.org/10.1124/jpet.114.217125 J Pharmacol Exp Ther 351:457–466, November 2014

Cannabidiol Improves Vasorelaxation in Zucker Diabetic Fatty Rats through Cyclooxygenase Activation Amanda J. Wheal, Mariateresa Cipriano, Christopher J. Fowler, Michael D. Randall, and Saoirse Elizabeth O’Sullivan Pharmacology Research Group, School of Life Sciences, University of Nottingham Medical School, Queen’s Medical Centre, Nottingham, United Kingdom (A.J.W., M.D.R.); School of Medicine, University of Nottingham Medical School, Royal Derby Hospital, Derby, United Kingdom (S.E.O.); and Department of Pharmacology and Clinical Neuroscience, Umeå University, Umeå, Sweden (M.C., C.J.F.) Received May 30, 2014; accepted September 4, 2014

ABSTRACT Cannabidiol (CBD) decreases insulitis, inflammation, neuropathic pain, and myocardial dysfunction in preclinical models of diabetes. We recently showed that CBD also improves vasorelaxation in the Zucker diabetic fatty (ZDF) rat, and the objective of the present study was to establish the mechanisms underlying this effect. Femoral arteries from ZDF rats and ZDF lean controls were isolated, mounted on a myograph, and incubated with CBD (10 mM) or vehicle for 2 hours. Subsequent vasorelaxant responses were measured in combination with various interventions. Prostaglandin metabolites were detected using enzyme immunoassay. Direct effects of CBD on cyclooxygenase (COX) enzyme activity were measured by oxygraph assay. CBD enhanced the maximum vasorelaxation to acetylcholine (ACh) in femoral arteries from ZDF lean rats (P , 0.01) and especially ZDF rats (P , 0.0001). In ZDF arteries, this enhancement persisted after cannabinoid receptor

Introduction Cannabidiol (CBD) is a naturally occurring molecule found in the plant Cannabis sativa. Unlike the related molecule D9-tetrahydrocannabinol (THC), it does not activate cannabinoid receptor (CB) type 1 in the brain, and it is therefore devoid of the psychotropic actions of THC. Indeed, it has been argued that CBD may mitigate the psychoses associated with

All animal studies were funded by Diabetes UK [Reference: 08/0003822]. C.J.F. received financial support from the Swedish Research Council [Grant No. 12158, Medicine] and the Research Funds of the Medical Faculty, Umeå University. The authors report no conflicts of interest. dx.doi.org/10.1124/jpet.114.217125.

(CB) type 1, endothelial CB, or peroxisome proliferator–activated receptor-g antagonism but was inhibited by CB2 receptor antagonism. CBD also uncovered a vasorelaxant response to a CB2 agonist not previously observed. The CBD-enhanced ACh response was endothelium-, nitric oxide–, and hydrogen peroxide–independent. It was, however, COX-1/2– and superoxide dismutase–dependent, and CBD enhanced the activity of both purified COX-1 and COX-2. The CBD-enhanced ACh response in the arteries was inhibited by a prostanoid EP4 receptor antagonist. Prostaglandin E2 metabolite levels were below the limits of detection, but 6-keto prostaglandin F1a was decreased after CBD incubation. These data show that CBD exposure enhances the ability of arteries to relax via enhanced production of vasodilator COX-1/2–derived products acting at EP4 receptors.

cannabis abuse (see Schubart et al., 2011 and references therein). A CBD/THC combination, nabiximols (Sativex; GW Pharmaceuticals, Salisbury, UK), is currently licensed internationally for the treatment of multiple sclerosis, and CBD alone (Epidiolex; GW Pharmaceuticals) was recently approved as an investigational new drug by the Food and Drug Administration in children with medically intractable epilepsy. The therapeutic potential of CBD has been widely explored in preclinical models of many other disorders ranging from type 1 and type 2 diabetes (see Di Marzo et al., 2011; Horvath et al., 2012, for recent reviews) to prostate cancer (De Petrocellis et al., 2013). With respect to the former, CBD protects the bloodretinal barrier in models of hyperglycemia (El-Remessy et al.,

ABBREVIATIONS: ACh, acetylcholine; 2-AG, 2-arachidonoylglycerol; AH6809, 6-isopropoxy-9-xanthone-2-carboxylic acid; AM251, N-(piperidin-1-yl)5-(4-iodophenyl)-1-(2,4-dichlorophenyl)-4-methyl-1H-pyrazole-3-carboxamide; AM630, 6-iodo-2-methyl-1-[2-(4-morpholinyl)ethyl]-1H-indol-3yl](4-methoxyphenyl)methanone; CAY10441, 4,5-dihydro-N-[4-[[4-(1-methylethoxy)phenyl]methyl]phenyl]-1H-imadazol-2-amine; CB, cannabinoid receptor; CBD, cannabidiol; CBe, endothelial cannabinoid receptor; COX, cyclooxygenase; CP55940, (2)-cis-3-[2-hydroxy-4-(1,1-dimethylheptyl) phenyl]-trans-4-(3-hydroxypropyl)cyclohexanol; DETCA, sodium diethyldithiocarbamate trihydrate; FAAH, fatty acid amide hydrolase; GW9662, 2-chloro-5-nitro-N-phenylbenzamide; HU308, 4-[4-(1,1-dimethylheptyl)-2,6-dimethoxyphenyl]-6,6-dimethylbicyclo[3.1.1]hept-2-ene-2-methanol; IP, prostaglandin I2; L161982, N-[[49-[[3-butyl-1,5-dihydro-5-oxo-1-[2-(trifluoromethyl)phenyl]-4H-1,2,4-triazol-4-yl]methyl][1,19-biphenyl]-2-yl]sulfonyl]-3methyl-2-thiophenecarboxamide; L-NAME, NG-nitro-L-arginine methyl ester; O1918, 1,3-dimethoxy-5-methyl-2-[(1R,6R)-3-methyl-6-(1-methylethenyl)-2-cyclohexen-1-yl]benzene; PEG, polyethylene glycol; PG, prostaglandin; PPAR, peroxisome proliferator–activated receptor; Rmax, maximal vasorelaxant response; SNP, sodium nitroprusside; SOD, superoxide dismutase; THC, D9-tetrahydrocannabinol; TRAM-34, 1-[(2-chlorophenyl) diphenylmethyl]-1H-pyrazole; U46619, (5Z)-7-[(1R,4S,5S,6R)-6-[(1E,3S)-3-hydroxy-1-octenyl]-2-oxabicyclo[2.2.1]hept-5-yl]-5-heptenoic acid; URB597, cyclohexylcarbamic acid 39-(aminocarbonyl)-[1,19-biphenyl]-3-yl ester; ZDF, Zucker diabetic fatty. 457

458

O’Sullivan et al.

2006); decreases insulitis and inflammation in nonobese diabetesprone mice (Weiss et al., 2006, 2008); decreases endothelial cell activation and monocyte adhesion in hyperglycemia (Rajesh et al., 2007); decreases neuropathic pain in streptozotocindiabetic mice (Toth et al., 2010); and decreases myocardial dysfunction, cardiac fibrosis, oxidative stress, and inflammation in a mouse model of type 1 diabetic cardiomyopathy (Rajesh et al., 2010). Indeed, there is much preclinical evidence to suggest that CBD is beneficial throughout the cardiovascular system, as recently reviewed (Stanley et al., 2013a). CBD is far from a selective compound and produces effects mediated by a number of target sites of action. It has a relatively poor ability to inhibit the binding of the synthetic cannabinoid agonist [3H]CP55940 [(2)-cis-3-[2-hydroxy-4-(1,1-dimethylheptyl) phenyl]-trans-4-(3-hydroxypropyl)cyclohexanol] to CB1 and CB2 receptors but is possibly a functional antagonist (or inverse agonist) of these receptors (Pertwee, 2008). Other receptor sites implicated in the actions of CBD include the orphan G protein–coupled receptor GPR55, the putative endothelial cannabinoid receptor (CBe), the transient receptor potential vanilloid 1 receptor, a1-adrenoceptors, m-opioid receptors, and serotonin receptor 1 subtype A receptors (Pertwee, 2008). CBD also activates and has physiologic responses mediated by peroxisome proliferator–activated receptor (PPAR)-g (O’Sullivan et al., 2009; De Filippis et al., 2011; Esposito et al., 2011). Additional mediators of CBD actions include enzymes capable of metabolizing endogenously produced cannabinoids (termed endocannabinoids), such as lipoxygenases, cyclooxygenase (COX)-2, and fatty acid amide hydrolase (FAAH) (Watanabe et al., 1996; Costa et al., 2004; Massi et al., 2008; Takeda et al., 2011). We recently reported that in vitro incubation with CBD at 1 or 10 mM improved vasorelaxation in aortae and femoral arteries of Zucker diabetic fatty (ZDF) rats (Stanley et al., 2013b). The aim of the present study was to determine the mechanisms underlying this effect of CBD in the vasculature of the ZDF rat. We hypothesized that these could involve activation of PPARg (O’Sullivan et al., 2009) and/or actions mediated by key vascular enzymes, such as COX and superoxide dismutase (SOD) (O’Sullivan et al., 2006), and represent a facet of the beneficial effects of CBD in diabetes or indeed other vascular disorders.

4.9 mN in warmed (37°C), gassed (95% O2/5% CO2), modified KrebsHenseleit buffer (118 mM NaCl, 4.7 mM KCl, 1.2 mM MgSO4, 1.2 mM KH2PO4, 25 mM NaHCO3, 10 mM D-glucose, and 2 mM CaCl2). PowerLab recording systems (A&D Instruments, Abingdon, UK) were used to record changes in tension. Contraction with high-K1 buffer

Materials and Methods Animals. Male ZDF rats (Charles River Laboratories, Wilmington, MA) (n 5 51) were housed in groups of 2–4 in the University of Nottingham Biomedical Services Unit in a 12-hour light/dark cycle at 22 6 2°C with access to chow and water ad libitum. Homozygote recessive males (fa/fa) develop obesity, hyperlipidemia, fasting hyperglycemia, and type 2 diabetes (ZDF rats; n 5 35). Wild types (1/1) and heterozygous (fa/1) lean genotypes remain normoglycemic (ZDF lean controls; n 5 16). At 12–13 weeks of age, the rats were stunned by a blow to the head and killed by cervical dislocation. Postmortem blood glucose concentrations were measured using an Accu-Chek Aviva analyzer (Roche Diagnostics Ltd., Mannheim, Germany). All procedures were in accordance with the UK Home Office Animal (Scientific Procedures) Act 1986 and Nottingham University Ethical Review Panel. ZDF rats were heavier than ZDF lean control rats [384 6 2 g and 305 6 14 g, respectively (mean 6 S.D.); P , 0.0001, Student’s t test]. Blood glucose levels were also higher in ZDF rats than ZDF lean rats (22.2 6 5.1 mM and 7.6 6 0.7 mM, respectively; P , 0.0001). Experimental Design. Sections of femoral artery (∼2 mm; up to 16 segments of femoral artery per rat) were mounted on a myograph (Danish Myo Technology, Aarhus, Denmark) at a basal tension of

Fig. 1. Effects of CBD on vasorelaxant responses to ACh. Arteries from ZDF lean rats (A) or ZDF rats (B) were incubated for 2 hours with vehicle or 10 mM CBD. Some arteries were washed out following the 2-hour incubation period (C). Data are mean 6 S.E.M. Data were compared using unpaired Student’s t test. **P , 0.01; ***P , 0.001; ****P , 0.0001, CBD versus vehicle.

Mechanisms of CBD-Improved Vasorelaxation (62.5 mM NaCl, 59.4 mM KCl, 1.2 mM MgSO4, 1.2 mM KH2PO4, 25 mM NaHCO3, 10 mM D-glucose, and 2 mM CaCl2) was used to demonstrate the viability of the mounted vessels. Then arteries were incubated for 2 hours with either 10 mΜ CBD or vehicle (5 ml of ethanol in 5 ml, 0.1%). Following this, arteries were contracted with an a1-adrenoceptor agonist, methoxamine (10–100 mΜ), to reach a stable tone (at least 4.9 mN above basal tone). In some cases, U46619 [(5Z)-7-[(1R,4S,5S,6R)-6-[(1E,3S)-3-hydroxy-1-octenyl]-2-oxabicyclo [2.2.1]hept-5-yl]-5-heptenoic acid], a thromboxane A2 mimetic, at up to 100 nM was also added to achieve the required level of tone or to help stabilize tone. Cumulative concentration-response curves to

459

acetylcholine (ACh), sodium nitroprusside (SNP), or the CB2 receptor agonist HU308 (4-[4-(1,1-dimethylheptyl)-2,6-dimethoxyphenyl]-6,6dimethylbicyclo[3.1.1]hept-2-ene-2-methanol; 1 nM to 30 mM) were constructed. Potential target sites of action were examined using a range of antagonists (or enzyme inhibitors) acting at PPARg (GW9662 [2-chloro-5-nitro-N-phenylbenzamide]; 1 mM), CB1 (AM251 [N-(piperidin1-yl)-5-(4-iodophenyl)-1-(2,4-dichlorophenyl)-4-methyl-1H-pyrazole3-carboxamide]; 1 mM), CB2 (AM630 [6-iodo-2-methyl-1-[2-(4-morpholinyl) ethyl]-1H-indol-3-yl](4-methoxyphenyl)methanone]; 1 mM), the putative CBe (O1918 [1,3-dimethoxy-5-methyl-2-[(1R,6R)-3-methyl-6-(1-

Fig. 2. CB receptor involvement. ZDF arteries were incubated for 2 hours with vehicle or 10 mM CBD and coincubated with 1 mM AM251 (A), 1 mM AM630 (B), 1 mM URB597 (C), 1 mM O1918 (D), or 1 mM GW9662 (E). (F) A concentration-response curve to the CB2 agonist HU308 performed in the presence of CBD and/or indomethacin. Data are mean 6 S.E.M. For (A–E), Rmax values were compared using unpaired Student’s t test. In (F), Rmax values were compared using one-way analysis of variance, but those coincubated with vehicle and indomethacin could not be included in the comparisons due to not being able to fit a curve for those data. *P , 0.05; **P , 0.01; ***P , 0.001, CBD versus vehicle.

460

O’Sullivan et al.

methylethenyl)-2-cyclohexen-1-yl]benzene]; 1 mM), and FAAH (URB597 [cyclohexylcarbamic acid 39-(aminocarbonyl)-[1,19-biphenyl]-3-yl ester]; 1 mM). In some arteries, the intima was rubbed with forceps to remove the endothelium. The role of endothelium-derived hyperpolarization was investigated by inhibition of nitric oxide synthase with NG-nitro-Larginine methyl ester (L-NAME; 300 mM), inhibition of COX with indomethacin (3 mM), and blockade of small and intermediate calciumactivated potassium channels with apamin (0.5 mM) and TRAM-34 (1-[(2-chlorophenyl)diphenylmethyl]-1H-pyrazole; 10 mM), respectively. The involvement of COX was investigated by coincubation with a nonselective COX inhibitor, indomethacin (3 mM) or flurbiprofen (10 mM), or a COX-2–selective inhibitor, nimesulide (10 mM). The prostaglandin D2/prostaglandin E prostanoid receptor antagonist AH6809 (6-isopropoxy-9-xanthone-2-carboxylic acid; 1 mM), the EP4 receptor antagonist L161982 (N-[[49-[[3-butyl-1,5-dihydro-5-oxo1-[2-(trifluoromethyl)phenyl]-4H-1,2,4-triazol-4-yl]methyl][1,19-biphenyl]2-yl]sulfonyl]-3-methyl-2-thiophenecarboxamide; 1 mM), and the prostaglandin I2 (IP; prostacyclin) antagonist CAY10441 (4,5-dihydroN-[4-[[4-(1-methylethoxy)phenyl]methyl]phenyl]-1H-imadazol-2-amine; 100 nM) were used. To establish any role for hydrogen peroxide and superoxides, arteries were coincubated with polyethylene glycol (PEG)– catalase (300 U/ml) to break down active endogenous hydrogen peroxide or sodium diethyldithiocarbamate trihydrate (DETCA; 300 mM) to inhibit SOD. COX Activity Assays. The method used was that of Meade et al. (1993) with minor modifications (Onnis et al., 2010). Briefly, buffer containing 1 mM hematin, 2 mM phenol, 5 mM EDTA, substrate [10 mM arachidonic acid or 2-arachidonoylglycerol (2-AG)], and 0.1 M Tris-HCl, pH 7.4 at room temperature, was added to an oxygen electrode chamber with an integral stirring unit (Oxygraph System; Hansatech Instruments, King’s Lynn, UK) (final assay volume, 2 ml). The oxygen

electrode was calibrated with respect to air pressure and ambient temperature. After the addition of 20 ml of test compound (vehicle, CBD, or, as a positive control, flurbiprofen; CBD 6 AM630), a baseline was established over a period of 5 minutes. Reactions were started by addition of ovine COX-1 or human recombinant COX-2 (200 U per assay), as appropriate, and the oxygen consumption was measured in 10-second blocks. Data are presented as the change in oxygen consumption due to substrate oxygenation (micromoles per liter) from the point of addition of the enzyme. Prostaglandin Metabolites. The 5 ml of Krebs buffer used to incubate arteries for the 2-hour incubation and ACh concentrationresponse curve was collected and frozen at 280°C. In this, using commercially available enzyme immunoassays (Cayman Chemicals, Cambridge Bioscience, Cambridge, UK), two stable derivatives of prostaglandin (PG) E2 (13,14-dihydro-15-keto PGA2 and 13,14-dihydro15-keto PGE2) and a PGI2 (prostacyclin) metabolite (6-keto PGF1a) were measured. Data Analysis and Statistical Procedures. GraphPad Prism (GraphPad Software, Inc., La Jolla, CA) was used to plot the data as mean percentage relaxation, with error bars representing S.E.M. and n being the number of arteries from different animals. Data in Fig. 1 (where n . 8) were tested for normality using the D’Agnostino-Pearson omnibus normality test. The software was used to fit the sigmoidal concentration-response curve to the mean data using a logistic equation. Percentage maximal vasorelaxant responses (Rmaxs) calculated from these mean data were compared between strains and treatments using unpaired Student’s t test or one-way analysis of variance followed by Bonferroni post hoc test, with P , 0.05 taken as significant. Drugs, Chemical Reagents, and Other Materials. Unless otherwise stated, all chemicals were purchased from Sigma-Aldrich (Dorset, UK). Apamin, GW9662, L161982, URB597, O1918, AM251,

Fig. 3. Nitric oxide and endothelium-derived hyperpolarization. ZDF arteries were incubated for 2 hours with vehicle or 10 mM CBD after endothelium denudation (A) or coincubated with L-NAME (300 mM) (B), L-NAME and indomethacin (3 mM) (C), or L-NAME and indomethacin combined with apamin (0.5 mM) and TRAM-34 (10 mM) (D). Data are mean 6 S.E.M. Rmax values were compared using unpaired Student’s t test. ****P , 0.001, CBD versus vehicle.

Mechanisms of CBD-Improved Vasorelaxation and AM630 were bought from Tocris Bioscience (Bristol, UK). U46619 was purchased from Enzo Life Sciences (Exeter, UK). CAY10441, ovine COX-1, and human recombinant COX-2 were purchased from Cayman Chemical Co. (Ann Arbor, MI). In the myography experiments, CBD was a gift from GW Pharmaceuticals. For the experiments with purified COX enzymes, CBD was obtained from Tocris. Stock solutions of CBD, U46619, GW9662, O1918, indomethacin, nimesulide, flurbiprofen, CAY10441, AH6809, and HU308 were made to 10 mM in ethanol; TRAM-34, AM251, AM630, and L161982 were made to 10 mM in dimethylsulfoxide; and URB597 was made to 1 mM in dimethylsulfoxide. Stock solutions of apamin (0.5 mM), L-NAME (100 mM), PEG-catalase (100,000 U/ml), methoxamine, ACh, SNP, and TRAM-34 (all 10 mM) were made in distilled water. DETCA was dissolved directly into Krebs. Serial dilutions of compounds were made daily.

Results CBD Improves Vasorelaxant Responses. Incubation for 2 hours with 10 mM CBD enhanced the maximum vasorelaxation compared with vehicle (5 ml of ethanol). In ZDF

461

lean rats, CBD caused an ∼15% increase in the maximal response to ACh (Rmax values: vehicle, 51 6 2%; CBD, 59 6 2%; P , 0.01, unpaired Student’s t test; Fig. 1A). In ZDF rats, CBD caused an ∼40% increase in vasorelaxation (Rmax values: vehicle, 46 6 1%; CBD, 65 6 1%; P , 0.0001, unpaired Student’s t test; Fig. 1B). The CBD-induced enhancement of ACh responses was still apparent when the preparations were washed out after the 2-hour incubation period (P , 0.001; Fig. 1C). The vasorelaxant response to SNP was enhanced in ZDF lean but not ZDF rats after 10 mM CBD (Rmax values: ZDF lean ethanol-incubated, 55 6 6%, n 5 7; ZDF lean CBD-incubated, 72 6 2%, n 5 7; P , 0.05; data not shown). CB Receptor Involvement. CBD-enhanced vasorelaxation was still observed following coincubation of the arteries with the CB1 receptor antagonist AM251 (P , 0.01; Fig. 2A), the FAAH inhibitor URB597 (P , 0.01; Fig. 2C), the CBe antagonist O1918 (P , 0.05; Fig. 2D), and the PPARg antagonist GW9662 (P , 0.001; Fig. 2E). However, the CB2

Fig. 4. COX involvement. ZDF arteries were incubated for 2 hours with vehicle or 10 mM CBD and coincubated with 3 mM indomethacin (A), 1 mM AH6809 (B), 10 mM flurbiprofen (C), 1 mM L161982 (D), 10 mM nimesulide (E), or 100 nM CAY10441 (F). Data are mean 6 S.E.M. Rmax values were compared using unpaired Student’s t test. *P , 0.05; ***P , 0.001, CBD versus vehicle.

462

O’Sullivan et al.

receptor antagonist AM630 inhibited the effects of CBD (Fig. 2B). To assess a potential role for CB2 receptors in the effects of CBD, the CB2 receptor agonist HU308 was cumulatively added to precontracted arteries. HU308 did not cause vasorelaxation of femoral arteries in control conditions. However, following a 2-hour incubation with CBD, a concentrationdependent vasorelaxation to HU308 was observed (Rmax, 35 6 4%; Fig. 2F), which was not affected by the presence of indomethacin. Potential Involvement of the Endothelial Pathways. The CBD-enhanced ACh response persisted after removal of the endothelium (P , 0.0001; Fig. 3A), inhibition of nitric oxide (P , 0.0001; Fig. 3B), or a combination of L-NAME and indomethacin (P , 0.0001; Fig. 3C). However, in the presence of a combination of inhibitors (apamin, TRAM-34, L-NAME, and indomethacin) accepted as blocking endothelium-derived hyperpolarization factor, the effect of CBD was no longer significant (P 5 0.0829; Fig. 3D). Potential Involvement of COXs. Indomethacin (Fig. 4A), flurbiprofen (Fig. 4C), and nimesulide (Fig. 4E) abolished the enhancement of ACh-induced vasorelaxation by CBD in ZDF rats. The CBD-induced enhanced ACh response persisted in the presence of the prostaglandin D2/prostaglandin E prostanoid receptor antagonist AH6809 (P , 0.05; Fig. 4B) and the IP (prostacyclin) receptor antagonist CAY10441 (P ,

0.001; Fig. 4F). However, the EP4 antagonist L161982 abolished the enhancement of ACh-induced vasorelaxation by CBD (Fig. 4D). COX Activity Assays. Arachidonic acid is metabolized by COX-1 and COX-2, and the endocannabinoid 2-AG is metabolized by COX-2 but not COX-1. CBD enhanced the metabolism of arachidonic acid by both COX-1 and COX-2, and that of 2-AG by COX-2 (Fig. 5A). In all cases, a two-way analysis of variance matching both time and CBD indicated that there was a significant (P , 0.0001) interaction term time  CBD, indicating that the effect of CBD is dependent on the incubation time used. For the nine experiments undertaken with 10 mM CBD and COX-1 (Fig. 5, A and B), this was confirmed by expressing the values at 30 and 120 seconds (after subtraction of the values at 10 seconds due to the short lag phase seen in the assays) as percentages of the corresponding vehicle controls: for the short incubation time, the mean value for 10 mM CBD was 91% of control (95% confidence interval, 63–119%), whereas at the 120-second time point, the corresponding value was 174% (95% confidence interval, 140–209%; P , 0.005, one-sample t test). Because AM630 has structural similarity to indomethacin, we tested whether this compound might interact with COX-1. On its own, 30 mM AM630 inhibited COX-1 activity (Fig. 5B). AM630 also inhibited the ability of CBD to enhance COX-1 metabolism. The initial inhibitory effect of AM630 on COX-1

Fig. 5. CBD and COX activity. (A) Changes in activities of ovine COX-1 (toward 10 mM arachidonic acid) and human recombinant COX-2 by CBD. F30 refers to 30 mM flurbiprofen, as a positive control. (B) Changes in activities of ovine COX-1 by CBD (10 mM) and AM630 (30 mM). Data shown are mean 6 S.E.M. (n = 4). DMSO, dimethylsulfoxide; EtOH, ethanol. (C) PGI2 metabolite content in the Krebs-Henseleit buffer used in experiments with arteries taken from lean or ZDF rats that had been incubated for 2 hours with vehicle or 10 mM CBD (mean 6 S.E.M.; n = 7). Values were compared using oneway analysis of variance plus Bonferroni post hoc test on selected pairs. *P , 0.05, CBD versus vehicle.

Mechanisms of CBD-Improved Vasorelaxation

463

activity occurs faster than the ability of CBD to increase COX-1 activity (Fig. 5B). Thus, a two-way analysis of variance for matching for both CBD and AM630 of the change in oxygen tension between 10 and 30 seconds gave F1,3(CBD) 5 0.22 (P . 0.6); F1,3(AM630) 5 14 (P , 0.05); and F1,3(CBD  AM630) 5 30 (P , 0.05). The corresponding values for the change in oxygen tension between 10 and 120 seconds were F1,3(CBD) 5 10 (P , 0.05); F1,3(AM251) 5 12 (P , 0.05); and F1,3(CBD  AM630) 5 5.5 (P 5 0.1). PG Metabolites. After the ACh concentration-response curves, the incubate was tested for PG metabolites. The amount of PGE metabolites in most of the samples was below the minimum detection limit of the assay (data not shown). However, levels of 6-keto PGF1a were detectable, and ZDF femoral arteries treated with CBD had significantly less 6-keto PGF1a than those in the other conditions (Fig. 5C). Potential Involvement of Hydrogen Peroxide and SOD. CBD enhanced vasorelaxation to ACh in the presence of PEG-catalase (metabolizes hydrogen peroxide) (P , 0.01; Fig. 6A) or catalase plus L-NAME and indomethacin (P , 0.001; Fig. 6B). However, the SOD inhibitor DETCA inhibited the effects of CBD on ACh (Fig. 6C). Mechanisms of CBD-Enhanced ACh-Induced Vasorelaxation in ZDF Lean Rats. In ZDF lean rats, ACh caused a concentration-dependent vasorelaxation that was enhanced by CBD (Fig. 1A). As observed in the ZDF rats, these actions of CBD were blocked by coincubation with indomethacin (Fig. 7A), nimesulide (Fig. 7B), L161982 (Fig. 7C), DETCA (Fig. 7D), or AM630 (Fig. 7E). The effects of CBD were still observed following removal of the endothelium in arteries from the ZDF lean rats (Fig. 7F).

Discussion This study was designed to identify the mechanisms of action underlying the improvements in endothelial function observed in arteries from diabetic rats after prolonged incubation with CBD. CBD enhances COX activity in arteries, leading to the production of vasodilator prostanoids acting at the EP4 receptor (see Fig. 8 for a summary diagram of the proposed mechanisms of action). This represents another facet in the therapeutic potential of CBD in diabetes, which has already been shown to protect the blood-retinal barrier and endothelial cell activation in hyperglycemia (El-Remessy et al., 2006; Rajesh et al., 2007) and to decrease insulitis and inflammation, neuropathic pain, and myocardial dysfunction in diabetic mice (Weiss et al., 2006, 2008; Rajesh et al., 2010; Toth et al., 2010). CBD Enhances ACh Responses in Diabetic Rats: A Role for the Endocannabinoid System. Our previous findings that CBD enhances ACh responses (Stanley et al., 2013b) are confirmed by the current data. This response persisted after washout of arteries, suggesting intrinsic alterations in the arteries. Use of appropriate antagonists ruled out activation of CB1, CBe, or PPARg by CBD. It should be noted that the size of the enhancement caused by CBD visually appears to be reduced by the CBe antagonist O1918, although the effect is still significant. This might suggest an involvement of this receptor. However, because the effect of CBD persisted in the absence of an endothelium, this is unlikely. A proposed mechanism of action for the antitumor effect of CBD is via the inhibition of FAAH and consequent

Fig. 6. Hydrogen peroxide and SOD. ZDF arteries were incubated for 2 hours with vehicle or 10 mM CBD and coincubated with 300 mM PEG-catalase alone (A) or in combination with 3 mM indomethacin and 300 mM L-NAME (B) or with 300 mM DETCA (C). Data are mean 6 S.E.M. Rmax values were compared using unpaired Student’s t test. **P , 0.01; ***P , 0.001, CBD versus vehicle.

increase in local levels of endocannabinoids (Massi et al., 2008). We previously showed that incubation of diabetic arteries with either anandamide or 2-AG can also improve ACh responses in diabetic rats (Stanley et al., 2013b), and endocannabinoids themselves are vasorelaxants (Randall et al., 2004); this could therefore explain the effect of CBD. However, when this was tested by coincubation of arteries with an FAAH inhibitor, the effects of CBD persisted.

464

O’Sullivan et al.

Fig. 7. CBD-enhanced vasorelaxation in ZDF lean rats. Arteries from ZDF lean rats were incubated for 2 hours with vehicle or 10 mM CBD and coincubated with 3 mM indomethacin (A), 10 mM nimesulide (B), 1 mM L161982 (C), 300 mM DETCA (D), or 1 mM AM630 (E) or were endotheliumdenuded (F). Data are mean 6 S.E.M. Rmax values were compared using unpaired Student’s t test. ***P , 0.001, CBD versus vehicle.

Treatment with a CB2 antagonist inhibited the effects of CBD. As CBD does not activate CB2 (Pertwee, 2008), two possible explanations are considered: 1. The antagonism produced by AM630 is an off-target effect. AM630 is widely used as a CB2 receptor antagonist/ inverse agonist and blocks this receptor at midnanomolar concentrations (Ross et al., 1999). However, relatively little is known about the interaction of the compound with other targets. Because AM630 shares some of the structural components of indomethacin, we tested whether this compound might also interact with COX-1. Inhibitory effects of AM630 on COX-1 were seen at 30 mM, but the effects on COX-1 were modest at 1 mM, which was the concentration used in the vascular studies. Thus, although we feel it unlikely that the sensitivity to AM630 relates to its action on COX-1, the present study underlines the importance of using selective (i.e., #1 mM) concentrations of this compound to produce a complete block of CB2 receptors without producing off-target effects.

2. CBD has allosteric effects on the CB2 receptor. CBD has allosteric effects at glycine (Foadi et al., 2010) and opioid (Kathmann et al., 2006) receptors. Although many compounds cause allosteric modulation of CB1, less is known about modulation of CB2. To support this theory, we found that CBD was able to induce a concentrationdependent vasorelaxant response to a synthetic CB2 ligand that did not cause vasorelaxation on its own (this is typical of CB2 ligands; see Stanley and O’Sullivan, 2014a). Thus, CBD must be altering receptor function rather than increasing local levels of endogenous CB2 agonists. Either way, enhancing CB2 responses, which are typically associated with anti-inflammatory actions, might explain some of the effects of CBD in diabetes. A Role for the Endothelium. The CBD-induced effects persisted after removal of the endothelium and nitric oxide production, ruling out these typical mediators of ACh responses. It was of interest that a significant response (20%) to ACh could be observed in arteries of diabetic rats after removal of the

Mechanisms of CBD-Improved Vasorelaxation

Fig. 8. Summary of the proposed mechanism of action of CBD-enhanced vasorelaxation in rats.

endothelium, which was observed more than after removal of the endothelium in ZDF lean controls (,5% relaxation). This suggests that smooth muscle mediators of vasorelaxation are upregulated in compensation for endothelial dysfunction in diabetes, and that CBD can enhance this response. CBD was also able to enhance the vasorelaxant response to the endotheliumindependent relaxant SNP, suggesting that vascular smooth muscle is a likely target of CBD. Inhibition of endotheliumderived hyperpolarization almost completely abolished the ACh response in diabetic arteries, and although CBD still appears to enhance the response, this was no longer significantly different from the vehicle control, so a role for endotheliumderived hyperpolarization cannot be ruled out. A Role for COX. The enhanced ACh response after CBD incubation was abolished in the presence of both COX-1 and COX-2 inhibitors. COXs are homodimeric in structure, but the two monomers behave differently, one acting as the catalytic unit and the other as an allosteric regulator (Yuan et al., 2009). Fatty acids that are not themselves substrates for COX can bind to the allosteric monomers and thereby regulate the catalytic activity of the enzymes. For example, palmitic acid increases the activity of human recombinant COX-2 (Dong et al., 2011). Interestingly, the effects of the fatty acids on COX-2 are less pronounced at higher arachidonic acid concentrations (Yuan et al., 2009). This may also be the case for CBD, given that the concentration of substrate used in the present study (10 mM) was lower than that in the study of Massi et al. (2008) (75 mM). Whatever the explanation, the present data are consistent with the hypothesis that the vascular effects of CBD seen here are due, at least in part, to a direct activation of COX by the compound. To address this hypothesis, we performed in vitro experiments, and CBD was found to increase the rate of oxygenation of arachidonic acid by both COX-1 and -2, and of 2-AG by COX-2. The activation showed some time dependency, suggesting that CBD associates rather slowly with the COX enzymes. With respect to direct effects on purified COX-2, the only data available, to our knowledge, are the studies by Massi et al. (2008) and Ruhaak et al. (2011) using arachidonic acid as substrate. Massi et al. (2008) reported that CBD inhibited ram

465

seminal vesicle COX-1– and human recombinant COX-2–catalyzed oxygenation of arachidonic acid at concentrations of $50 mM, with ∼50% reductions in activity being seen at a concentration of 1 mM. In contrast, Ruhaak et al. (2011) reported that CBD at a concentration of 100 mg/ml (∼300 mM) did not affect the production of [14C]prostaglandins following incubation of [14C]arachidonic acid with purified ram seminal vesicle COX-1, and produced an apparent increase in activity in the same assay using COX-2 purified from sheep placental cotyledons. The CBD-induced enhanced ACh response persisted in the presence of IP (prostacyclin) and DP/EP prostanoid receptor antagonists. However, an EP4 receptor antagonist abolished the enhancement of ACh-induced vasorelaxation by CBD. Interestingly, the endocannabinoid anandamide has also been shown cause vasorelaxation of the rat aorta via an arachidonic acid–COX-2–derived metabolite acting at EP4 receptors (Herradon et al., 2007), and we have found that 2-AG relaxes human mesenteric arteries partly through the EP4 receptor (Stanley and O’Sullivan, 2014b). EP4 is typically activated by PGE2. Therefore, enzyme immunoassays were performed to detect any increases in PGE2 metabolites after CBD. Unfortunately, PGE2 metabolites in all conditions were below the limit of detection, although this does not conclusively rule out an increase in the secretion of PGE2, as the arterial segments were very small compared with the volume of buffer assayed. We did, however, observe that CBD decreased the level of PGI2 metabolites. As PGI2 can have vasoconstrictor effects through the thromboxane receptor when stimulated by ACh (Liu et al., 2012, 2013), its reduction by CBD might suggest that one of the effects of CBD is to reduce the level of vasoconstrictor prostanoids produced by COX metabolism, thereby enhancing vasorelaxant capacity. The reduction in PGI2 might also reflect preferential metabolism of arachidonic acid into other, yet to be identified products that activate and cause vasorelaxation via EP4 receptors. A Role for SOD. Some of the vascular effects of cannabinoids are due to increases in SOD activity (O’Sullivan et al., 2005, 2006). Similarly, in the present study, the effects of CBD were abolished by a SOD inhibitor. One mechanism of SOD is to prevent nitric oxide from being scavenged by endogenous superoxides; however, the effects of CBD were unaffected by inhibition of nitric oxide synthase. SOD also catalyses the production of H2O2, which brings about some of the endothelium-dependent vasorelaxant effects of ACh (Matoba et al., 2000). However, the effects of CBD in the present study were not inhibited by catalase, which metabolizes H2O2 into water and oxygen, thus terminating its biologic actions. Increased SOD activity must therefore promote vasorelaxation through other mechanisms. For example, Rajesh et al. (2010) found that CBD restores SOD activity in diabetic mice, associated with reduced lipid peroxides and reactive oxygen species. ZDF Lean Controls. Although the effect of CBD was more pronounced in the diabetic rats, a small effect of CBD to enhance ACh responses was also observed in the arteries of ZDF lean rats. We therefore carried out a series of experiments that determined that the same mechanisms of action of CBD occur in the ZDF lean rats as were identified in the ZDF rats. Conclusion. CBD enhances vasorelaxant responses (to ACh or a CB2 agonist) in arteries from this rat model of type 2 diabetes. The mechanisms of action involve CB2 receptors and the enhancement of COX and SOD activity. The effect of

466

O’Sullivan et al.

CBD was endothelium-independent. Products of CBD-enhanced COX activity cause vasorelaxation through activation of vasodilator prostanoid EP4 receptors. Acknowledgments

The authors thank Dr. Richard Roberts for the use of a myograph. Authorship Contributions

Participated in research design: Wheal, O’Sullivan, Randall, Fowler. Conducted experiments: Wheal, Cipriano. Performed data analysis: Wheal, O’Sullivan, Cipriano. Wrote or contributed to the writing of the manuscript: Wheal, Cipriano, Fowler, Randall, O’Sullivan. References Costa B, Colleoni M, Conti S, Parolaro D, Franke C, Trovato AE, and Giagnoni G (2004) Oral anti-inflammatory activity of cannabidiol, a non-psychoactive constituent of cannabis, in acute carrageenan-induced inflammation in the rat paw. Naunyn Schmiedebergs Arch Pharmacol 369:294–299. De Filippis D, Esposito G, Cirillo C, Cipriano M, De Winter BY, Scuderi C, Sarnelli G, Cuomo R, Steardo L, De Man JG, et al. (2011) Cannabidiol reduces intestinal inflammation through the control of neuroimmune axis. PLoS One 6:e28159. De Petrocellis L, Ligresti A, Schiano Moriello A, Iappelli M, Verde R, Stott CG, Cristino L, Orlando P, and Di Marzo V (2013) Non-THC cannabinoids inhibit prostate carcinoma growth in vitro and in vivo: pro-apoptotic effects and underlying mechanisms. Br J Pharmacol 168:79–102. Di Marzo V, Piscitelli F, and Mechoulam R (2011) Cannabinoids and endocannabinoids in metabolic disorders with focus on diabetes. Handbook Exp Pharmacol 203:75–104. Dong L, Vecchio AJ, Sharma NP, Jurban BJ, Malkowski MG, and Smith WL (2011) Human cyclooxygenase-2 is a sequence homodimer that functions as a conformational heterodimer. J Biol Chem 286:19035–19046. El-Remessy AB, Al-Shabrawey M, Khalifa Y, Tsai NT, Caldwell RB, and Liou GI (2006) Neuroprotective and blood-retinal barrier-preserving effects of cannabidiol in experimental diabetes. Am J Pathol 168:235–244. Esposito G, Scuderi C, Valenza M, Togna GI, Latina V, De Filippis D, Cipriano M, Carratù MR, Iuvone T, and Steardo L (2011) Cannabidiol reduces Ab-induced neuroinflammation and promotes hippocampal neurogenesis through PPARg involvement. PLoS One 6:e28668. Foadi N, Leuwer M, Demir R, Dengler R, Buchholz V, de la Roche J, Karst M, Haeseler G, and Ahrens J (2010) Lack of positive allosteric modulation of mutated alpha(1)S267I glycine receptors by cannabinoids. Naunyn Schmiedebergs Arch Pharmacol 381:477–482. Herradón E, Martín MI, and López-Miranda V (2007) Characterization of the vasorelaxant mechanisms of the endocannabinoid anandamide in rat aorta. Br J Pharmacol 152:699–708. Horváth B, Mukhopadhyay P, Haskó G, and Pacher P (2012) The endocannabinoid system and plant-derived cannabinoids in diabetes and diabetic complications. Am J Pathol 180:432–442. Kathmann M, Flau K, Redmer A, Tränkle C, and Schlicker E (2006) Cannabidiol is an allosteric modulator at mu- and delta-opioid receptors. Naunyn Schmiedebergs Arch Pharmacol 372:354–361. Liu B, Luo W, Zhang Y, Li H, Zhu N, Huang D, and Zhou Y (2012) Involvement of cyclo-oxygenase-1-mediated prostacyclin synthesis in the vasoconstrictor activity evoked by ACh in mouse arteries. Exp Physiol 97:277–289. Liu B, Zhang Y, Zhu N, Li H, Luo W, and Zhou Y (2013) A vasoconstrictor role for cyclooxygenase-1-mediated prostacyclin synthesis in mouse renal arteries. Am J Physiol Renal Physiol 305:F1315–F1322. Massi P, Valenti M, Vaccani A, Gasperi V, Perletti G, Marras E, Fezza F, Maccarrone M, and Parolaro D (2008) 5-Lipoxygenase and anandamide hydrolase (FAAH) mediate the antitumor activity of cannabidiol, a non-psychoactive cannabinoid. J Neurochem 104:1091–1100. Matoba T, Shimokawa H, Nakashima M, Hirakawa Y, Mukai Y, Hirano K, Kanaide H, and Takeshita A (2000) Hydrogen peroxide is an endothelium-derived hyperpolarizing factor in mice. J Clin Invest 106:1521–1530. Meade EA, Smith WL, and DeWitt DL (1993) Differential inhibition of prostaglandin endoperoxide synthase (cyclooxygenase) isozymes by aspirin and other nonsteroidal anti-inflammatory drugs. J Biol Chem 268:6610–6614.

Onnis V, Congiu C, Björklund E, Hempel F, Söderström E, and Fowler CJ (2010) Synthesis and evaluation of paracetamol esters as novel fatty acid amide hydrolase inhibitors. J Med Chem 53:2286–2298. O’Sullivan SE, Kendall DA, and Randall MD (2006) Further characterization of the time-dependent vascular effects of delta9-tetrahydrocannabinol. J Pharmacol Exp Ther 317:428–438. O’Sullivan SE, Sun Y, Bennett AJ, Randall MD, and Kendall DA (2009) Timedependent vascular actions of cannabidiol in the rat aorta. Eur J Pharmacol 612: 61–68. O’Sullivan SE, Tarling EJ, Bennett AJ, Kendall DA, and Randall MD (2005) Novel time-dependent vascular actions of Delta9-tetrahydrocannabinol mediated by peroxisome proliferator-activated receptor gamma. Biochem Biophys Res Commun 337:824–831. Pertwee RG (2008) The diverse CB1 and CB2 receptor pharmacology of three plant cannabinoids: delta9-tetrahydrocannabinol, cannabidiol and delta9-tetrahydrocannabivarin. Br J Pharmacol 153:199–215. Rajesh M, Mukhopadhyay P, Bátkai S, Haskó G, Liaudet L, Drel VR, Obrosova IG, and Pacher P (2007) Cannabidiol attenuates high glucose-induced endothelial cell inflammatory response and barrier disruption. Am J Physiol Heart Circ Physiol 293:H610–H619. Rajesh M, Mukhopadhyay P, Bátkai S, Patel V, Saito K, Matsumoto S, Kashiwaya Y, Horváth B, Mukhopadhyay B, Becker L, et al. (2010) Cannabidiol attenuates cardiac dysfunction, oxidative stress, fibrosis, and inflammatory and cell death signaling pathways in diabetic cardiomyopathy. J Am Coll Cardiol 56: 2115–2125. Randall MD, Kendall DA, and O’Sullivan S (2004) The complexities of the cardiovascular actions of cannabinoids. Br J Pharmacol 142:20–26. Ross RA, Brockie HC, Stevenson LA, Murphy VL, Templeton F, Makriyannis A, and Pertwee RG (1999) Agonist-inverse agonist characterization at CB1 and CB2 cannabinoid receptors of L759633, L759656, and AM630. Br J Pharmacol 126: 665–672. Ruhaak LR, Felth J, Karlsson PC, Rafter JJ, Verpoorte R, and Bohlin L (2011) Evaluation of the cyclooxygenase inhibiting effects of six major cannabinoids isolated from Cannabis sativa. Biol Pharm Bull 34:774–778. Schubart CD, Sommer IE, van Gastel WA, Goetgebuer RL, Kahn RS, and Boks MP (2011) Cannabis with high cannabidiol content is associated with fewer psychotic experiences. Schizophr Res 130:216–221. Stanley C and O’Sullivan SE (2014a) Vascular targets for cannabinoids: animal and human studies. Br J Pharmacol 171:1361–1378. Stanley CP, Hind WH, and O’Sullivan SE (2013a) Is the cardiovascular system a therapeutic target for cannabidiol? Br J Clin Pharmacol 75:313–322. Stanley CP and O’Sullivan SE (2014b) Cyclooxygenase metabolism mediates vasorelaxation to 2-arachidonoylglycerol (2-AG) in human mesenteric arteries. Pharmacol Res 81:74–82. Stanley CP, Wheal AJ, Randall MD, and O’Sullivan SE (2013b) Cannabinoids alter endothelial function in the Zucker rat model of type 2 diabetes. Eur J Pharmacol 720:376–382. Takeda S, Hirayama A, Urata S, Mano N, Fukagawa K, Imamura M, Irii A, Kitajima S, Masuyama T, Nomiyama M, et al. (2011) Cannabidiol-29,69-dimethyl ether as an effective protector of 15-lipoxygenase-mediated low-density lipoprotein oxidation in vitro. Biol Pharm Bull 34:1252–1256. Toth CC, Jedrzejewski NM, Ellis CL, and Frey WH, Jr (2010) Cannabinoid-mediated modulation of neuropathic pain and microglial accumulation in a model of murine type I diabetic peripheral neuropathic pain. Mol Pain 6:16. Watanabe K, Kayano Y, Matsunaga T, Yamamoto I, and Yoshimura H (1996) Inhibition of anandamide amidase activity in mouse brain microsomes by cannabinoids. Biol Pharm Bull 19:1109–1111. Weiss L, Zeira M, Reich S, Har-Noy M, Mechoulam R, Slavin S, and Gallily R (2006) Cannabidiol lowers incidence of diabetes in non-obese diabetic mice. Autoimmunity 39:143–151. Weiss L, Zeira M, Reich S, Slavin S, Raz I, Mechoulam R, and Gallily R (2008) Cannabidiol arrests onset of autoimmune diabetes in NOD mice. Neuropharmacology 54:244–249. Yuan C, Sidhu RS, Kuklev DV, Kado Y, Wada M, Song I, and Smith WL (2009) Cyclooxygenase allosterism, fatty acid-mediated cross-talk between monomers of cyclooxygenase homodimers. J Biol Chem 284:10046–10055.

Address correspondence to: Saoirse Elizabeth O’Sullivan, School of Medicine, University of Nottingham Medical School, Royal Derby Hospital, Uttoxeter Road, Derby DE22 3DT, UK. E-mail: [email protected]