Cannabidiol, a nonpsychoactive Cannabis constituent, protects against myocardial ischemic reperfusion injury
Cannabidiol (CBD) is a major, nonpsychoactive Cannabis constituent with anti-inflammatory activity mediated by enhancing adenosine signalling. Inasmuch as adenosine receptors are promising pharmaceutical targets for ischemic heart diseases, we tested the effect of CBD on ischemic rat hearts. For the in vivo studies, the left anterior descending coronary artery was transiently ligated for 30 min, and the rats were treated for 7 days with CBD (5 mg/kg ip) or vehicle. Cardiac function was studied by echocardiography. Infarcts were examined morphometrically and histologically. For ex vivo evaluation, CBD was administered 24 and 1 h before the animals were killed, and hearts were harvested for physiological measurements. In vivo studies showed preservation of shortening fraction in CBD-treated animals: from 48 ± 8 to 39 ± 8% and from 44 ± 5 to 32 ± 9% in CBD-treated and control rats, respectively (n = 14, P < 0.05). Infarct size was reduced by 66% in CBD-treated animals, despite nearly identical areas at risk (9.6 ± 3.9 and 28.2 ± 7.0% in CBD and controls, respectively, P < 0.001) and granulation tissue proportion as assessed qualitatively. Infarcts in CBD-treated animals were associated with reduced myocardial inflammation and reduced IL-6 levels (254 ± 22 and 2,812 ± 500 pg/ml in CBD and control rats, respectively, P < 0.01). In isolated hearts, no significant difference in infarct size, left ventricular developed pressures during ischemia and reperfusion, or coronary flow could be detected between CBD-treated and control hearts. Our study shows that CBD induces a substantial in vivo cardioprotective effect from ischemia that is not observed ex vivo. Inasmuch as CBD has previously been administered to humans without causing side effects, it may represent a promising novel treatment for myocardial ischemia.
cannabinoids are natural and synthetic compounds structurally or pharmacologically related to the constituents of the plant Cannabis sativa or to the endogenous agonists (endocannabinoids) of the cannabinoid CB1 and CB2 receptors (for recent reviews see Refs. 17, 22, and 24). In a rat model, it was shown that anandamide, an endogenous cannabinoid (8), has a physiological triphasic effect on the cardiovascular system consisting of an initial reduction in heart rate (HR) and blood pressure followed by a brief pressor response and then a prominent reduction of blood pressure and HR (29). It has also been shown that endocannabinoids have a protective effect against myocardial ischemia and can help preserve coronary endothelial function during ischemia (15, 27, 28). These effects are receptor mediated and can be inhibited by specific CB1 and CB2 receptor blockers (15, 27, 28). In one study, a potent CB1 and CB2 receptor agonist, HU-210, was shown to substantially reduce the myocardial necrotic zone after left anterior descending coronary artery (LAD) ligation (28); in another trial, HU-210 was shown to significantly reduce the ischemic arrhythmic effect in treated animals. These effects were mediated by the CB2 receptor (15, 27). Other studies have shown that the cardioprotective effect of heat and ischemic preconditioning is mediated, partly at least, by CB2 receptors and can be abolished by CB2 receptor antagonists (3, 13). HU-210 was also shown to increase coronary blood flow and to be associated with the reduced remodelling in infarcted rat hearts (31, 32). These results indicate a possible role for the endocannabinoid system in stress-induced preconditioning.
Cannabidiol (CBD) is a major cannabinoid constituent of Cannabis (18). In contrast to tetrahydrocannabinol, CBD binds very weakly to CB1 and CB2 receptors (18, 24). Contrary to most cannabinoids, CBD does not induce psychoactive or cognitive effects. CBD has been shown to have anti-inflammatory properties (6, 16, 33). CBD (together with tetrahydrocannabinol) has been successfully tested in a few preliminary human trials related to autoimmune diseases such as rheumatoid arthritis (2) and multiple sclerosis (4, 26, 38) and is well tolerated by humans (7, 18, 24).
The mechanisms of action of CBD remain obscure. It has a mild effect on vanilloid VR1 receptors (1, 25). Recently, it was proposed that the anti-inflammatory effects of CBD might be mediated by enhancement of adenosine signalling through inhibition of its uptake (5). This effect apparently involves the adenosine A2A receptor (5).
In view of the known cardioprotective effect of cannabinoids, the reduction in infarct size induced by adenosine receptor activation (34), the enhanced A2A receptor signalling by CBD (5), the anti-inflammatory effect of CBD, which might prevent immune-related myocardial injury (14), and its lack of psychotropic or cognitive effects (24), we studied the possible cardioprotective effect of CBD on ischemic rat hearts. We also tested whether any protective effect is exerted directly on the myocardium or via mediators of immune mechanisms.
Male Sprague-Dawley rats (275–300 g body wt) were used: 7 animals were used in the CBD-treated and vehicle-injected arms of the Langendorff isolated rat heart model, and 14 animals were used in the CBD-treated and vehicle-injected arms of the whole animal study. The Institutional Animal Study Review Board approved the study.
CBD was extracted from Cannabis resin (hashish) and purified as previously described (9). The crystalline CBD (66–67°C melting point) presented a single peak on gas chromatography analysis and is ≥99% pure (9). CBD was dissolved in absolute ethanol, with an equal volume of a detergent (Cremophor), and the solution was mixed until homogeneous. Saline was added to a final ratio of 1:1:18 ethanol-detergent-saline. The final concentration of 5 mg·kg body wt−1·day−1 was injected intraperitoneally.
In vivo LAD ligation procedure.
Rats were anesthetized with 10% ketamine-2% xylazine (0.85:0.15, 0.1 ml/kg im), intubated, and ventilated with a small animal respirator (Harvard). The heart was exposed via left sternotomy, and the LAD was reversibly occluded for 30 min with a snare fashioned from an encircling silk ligature and a section of 5-Fr feeding tube. Coronary occlusion and reperfusion were verified by visual inspection. After reperfusion, the released ligature was left in situ. The chest was closed in layers with uninterrupted sutures. Rats were extubated, lost body fluids were replaced by subcutaneous administration of 30–50 ml of saline, and the animals were treated with rymadil (5 mg/kg ip) analgesia for 3 days. The animals were randomized to receive intraperitoneal injections of 5 mg/kg CBD or solvent vehicle as control 1 h before the procedure and every 24 h thereafter for 7 days until they were killed. The drug and vehicle solutions were prepared in separate laboratories to ensure that the staff (e.g., echocardiography and pathology) was blinded to the treatment arm. After 7 days, animals were reanesthetized and ventilated as described above. The LAD was permanently reoccluded, and the heart was harvested for further analysis.
Ex vivo isolated Langendorff rat heart LAD ligation.
To assess the direct CBD effect on the heart, we used the isolated Langendorff rat heart model (19). Animals were intraperitoneally injected with 5 mg/kg CBD or solvent 24 and 1 h before heart removal. At 30 min before heart removal, unfractionated heparin (750 IU, 0.15 ml ip) was injected into the rat. Animals were anesthetized with 0.6–0.8 ml of 20% pentobarbital sodium by intraperitoneal injection. After it was removed, the heart was placed in cold (4°C) saline-heparin solution (100 ml of saline and 750 IU of heparin). The aorta was cannulated, and the heart was perfused with modified Krebs-Henseleit solution under a constant pressure of 90 mmHg. A water-filled latex balloon was inserted into the left ventricular (LV) cavity via the left atrium, coupled to a pressure transducer, to record LV developed pressures (LVDP). A 50-μl Hamilton syringe was used to adjust LV end-diastolic pressure (LVEDP) to 5–10 mmHg. Coronary flow (CF) was measured throughout the procedure by collection of the effluent. HR was set to 300 beats/min by an external pacemaker attached to the left auricle to adjust for possible differences in HR between the treatment groups. LVDP was continuously recorded on a polygraph. After 30 min of stabilization, a 4-0 silk suture was placed under the LAD below an imaginary interauricular line and then threaded through a buttonlike snare. This allows ligation for a prolonged period without harm to the heart. The LAD was occluded for 45 min; then the snare was released, and reperfusion through the LAD was allowed for 45 min. After 45 min of reperfusion, the LAD was reoccluded and the percent infarct size of the area at risk was determined (see below).
Echocardiographic imaging was performed using a GE Vivid3 platform equipped with a 13-MHz linear epiaortic transducer (General Electric, Haifa, Israel). The probe was positioned in a left parasternal position, and two-dimensional imaging of the heart in the short axis was performed using a high frame rate. This image was used to guide an M-mode cursor down the medial axis of the LV. Measurements were performed in triplicate using the leading edge convention for myocardial borders. The images were acquired with HR at 250–350 beats/min.
The following parameters were measured: LV end-diastolic diameter (LVEDD), LV end-systolic diameter (LVESD). Shortening fraction (SF) was calculated using these variables.
Infarct size measurements and histopathology.
Measurements were performed as previously described (20, 21, 37). Briefly, 0.5–1 ml of 1% Evans blue solution in saline was slowly infused retrogradely through the aorta into the coronary arteries to mark the area at risk (not stained). The heart was then frozen (−20°C) for 1 h in aluminum foil and cut into 2-mm-thick transverse sections from the occlusion area (LAD) to the apex (usually 4–6 slices). The slices were stained by incubation at 37°C for 15 min in 1% (wt/vol) triphenyltetrazolium chloride (TTC) in phosphate buffer (pH 7.4) and fixed in 10% (vol/vol) formaldehyde solution. The stained slices were photographed, and the area of LV at risk and the area of infarcted tissue in the risk zone were determined by planimetry using Adobe Photoshop software. Subsequently, tissue slices were fixed in 5% ammonium hydrochloride, embedded in paraffin, cut into 5-μm sections, and stained with hematoxylin and eosin and Masson's trichrome. The extent of the inflammatory response was graded on a scale of 1–3 by an experienced pathologist who was blinded to the treatment groups.
Serum TNF-α, IL-6, and C-reactive protein measurements.
Rats were bled 7 days after coronary ligation. Sera were separated and frozen. Concentration of TNF-α, Il-6, and C-reactive protein (CRP) in the sera was measured by commercial ELISA kit according to the manufacturer's protocols.
ELISA reagents for TNF-α were purchased from Biosource (Camarillo, CA), for IL-6 from AbD Serotec (Kidlington, UK) and Prospect (Rehovot, Israel), and for CRP from Immunology Consultants Laboratory (Newberg, OR).
Unpaired Student's t-test was used to compare baseline data between groups. Within-group outcomes were compared with baseline using paired t-test analysis. Between-group outcome analysis was performed using percent change from baseline and unpaired t-test. Linear correlation was used to compare infarct size and ejection fraction (EF).
Fourteen CBD- or vehicle-treated animals completed the experimental design. Four control animals and three CBD-treated animals died before day 7. Weight, LVEDD, LVESD, and SF of the animals that survived to day 7 are presented in Table 1.
All animals showed significant weight loss 7 days after the procedure: from 326.8 ± 3.4 g at baseline to 314.6 ± 8.7 g (CBD) and from 326.1 ± 4.3 g at baseline to 319.7 ± 8.1 g (controls; P < 0.05). As expected, a significant reduction in SF was observed in all animals after LAD ligation: from 48 ± 8% at baseline to 39 ± 8% (CBD) and from 44 ± 6% at baseline to 32 ± 9% (controls; P < 0.05). The LV area at risk and infarct size as a percentage of area at risk are presented in Fig. 1. The area at risk was similar in CBD and control animals. However, the mean infarct size was significantly and remarkably reduced in CBD-treated animals (9.6 ± 3.9% vs. 28.2 ± 7.0% in CBD and control arms, respectively, P < 0.001). Thus a relative reduction of 66% in the infarcted zone was observed in the CBD-treated animals. The ejection fraction on day 7 negatively correlated with infarct size (R2 = 0.45, β = −0.9; Fig. 2). To further determine whether the reduction in infarct size is related to an improved remodeling process, LV function of animals with the largest infarct in the CBD group was compared with that of animals at the lower end of infarct size range in the control groups. In these animals, the area of the infarct ranged from 16 to 20% in both groups. There was no difference in LVEDD on day 7 (7.4 ± 0.8 vs. 7 ± 0.6 mm respectively) or in SF (33% vs. 37% respectively) between CBD-treated animals and controls. Although this comparison is not statistically powerful, it might suggest that the CBD effect is mainly due to the reduction in infarct size, rather than major changes in myocardial contractility or remodeling processes.
Hearts from each group were stained with hematoxylin and eosin and Masson's trichrome (Fig. 3). The inflammatory infiltrate was graded on a scale of 1–3 by an expert pathologist on the basis of the number of leukocytes infiltrating the border of the infarcted zone. The inflammatory response in the control animals was graded 2–3 compared with 1+ or 0 in the CBD-treated animals (Fig. 3, A and B), indicating elimination of the inflammatory process in the treated animals. However, granulation tissue formation with early collagen deposition was similar in the treated and control animals (Fig. 3, D and E).
To understand whether the CBD cardioprotective effect is due to a direct effect on the myocardium or, rather, is mediated via secondary mechanisms, in vitro isolated heart studies were performed. Infarct size, area at risk, LVDP, and coronary flow (CF) after coronary ligation, presented as percentage of the baseline values after stabilization, are shown in Table 2. No differences in any of the parameters were observed between CBD-treated and control animals.
Infarct size and LVDP as percentage of baseline during ischemia and after reperfusion in CBD and control animals in an isolated heart preparation
In view of the lack of CBD effect on isolated heart, IL-6, CRP, and TNF-α were measured on day 7 (Table 3). IL-6 was significantly lower in the sera of CBD-treated than control animals (254 ± 22 vs. 2,812 ± 500 pg/ml, P < 0.001). CRP and TNF-α did not differ significantly between the two groups on day 7.
Our study demonstrates that CBD has a significant in vivo cardioprotective effect against myocardial ischemia induced by LAD ligation, with reduction of infarct size measured as percentage of the area at risk by TTC staining. This was accompanied by a significant reduction in the decrease in LV function, which is observed after coronary occlusion, measured by echocardiography (Fig. 4). The EF after 7 days correlated pretty well with infarct size, indicating that the preservation of EF in treated animals is mainly due to reduced infarct size. This fact is further supported by comparing animals with similar infarct sizes from both groups.
The reduction of the infarct size was associated with a qualitative reduction in inflammatory infiltration of the ischemic and necrotic myocardial zones. CBD did not affect early collagen formation as detected by Masson's trichrome staining. These findings were associated with a significant reduction in serum IL-6 levels. We did not detect a significant effect of CBD on infarct size, cardiac contractility, and coronary blood flow in the in vitro isolated rat heart model. Because CBD has known anti-inflammatory effects (5, 6), it is plausible that the in vivo cardioprotective effect of CBD may be mediated by a systemic immunomodulatory effect, such as the reduction in IL-6 levels. A similar concept was recently proposed as the mechanism of action of a newly synthesized molecule, 1,6-bis(phosphocholine)-hexane, which binds to CRP and prevents its interaction with various inflammatory ligands. Complement activation was also noted (23). Infarct size is reduced from 24% in CRP-treated rats 5 days after LAD ligation to 17% by 1,6-bis(phosphocholine)-hexane (23). CRP is assumed to play a role in the innate immunity and immune activation in damaged myocardium (14). Thus anti-inflammatory agents such as CBD and 1,6-bis(phosphocholine)-hexane might have a role in preventing immune-related myocardial injury. It should be stressed that the in vivo and in vitro models were not fully comparable. The CBD treatment duration and the reperfusion period were shorter in the isolated heart model. Also, the ischemic period was longer. These experimental differences may explain the difference in infarct size between the in vivo and ex vivo experiments. However, the CBD effect in vivo was very robust, whereas it was essentially nonexistent in the ex vivo experiment (Table 2), indicating that CBD did not have an equivalent effect in vivo and ex vivo.
Previous studies have shown that cannabinoids, which act on cannabinoid receptors, reduce infarct size and protect the heart from ischemia-related arrhythmia (15, 27, 28). Also endogenous cannabinoids were found to be involved in preconditioning induced by heat and ischemia. Furthermore, such cannabinoids have vasodilatory effects and can improve endothelial function (3, 13). In all these studies, however, the effects of these cannabinoids, either endogenous or synthetic, were shown to be mediated by activation of the CB1 or CB2 receptor. Inasmuch as CBD does not act via the known cannabinoid receptors and does not have direct hemodynamic effects, its mechanism of action presumably differs from that of the cannabinoid receptor agonists. Evidence of transactivation of adenosine receptors by CBD was recently published (5). This effect was associated with the anti-inflammatory properties of CBD.
Adenosine receptors seem to play a role in protecting the heart (34). Adenosine A1 and A3 receptors have been most intensively studied, and the A2A receptor may also play a role. The A3 receptor seems to have multiple beneficial effects on ischemia-reperfusion injury, including modulation of necrotic and apoptotic cell death and enhancement of contractile function (11). A1 receptor overexpression increases myocardial resistance to ischemia (12). The A2A receptor may also play a role in protection from ischemia (34–36). In view of the recently proposed mechanism of action of CBD associating it with enhancement of adenosine signaling (5), one might speculate that the effect on the myocardium may be mediated via adenosine receptor effects.
In the past, corticosteroids were studied as a potential therapy for myocardial infarction because of their anti-inflammatory effect (10). However, this therapy has ultimately turned out to be harmful, inasmuch as corticosteroids delay and interfere with myocardial scar formation. This leads to thinner scars and increases susceptibility to rupture (30). In our experiments, we observed the treated animals for 7 days only. Thus it is possible that the beneficial effect observed after 7 days may become harmful over a longer time period. It might cause a delay in scar healing and also might reduce scar strength, leading to unfavorable LV remodeling in a manner similar to that of corticosteroids. Therefore, in CBD-treated animals, a longer period of follow-up is warranted to assess the long-term recovery and safety. However, we did not find any difference between the treated and control animals in relation to myofibroblast activity. This might suggest that CBD did not adversely affect the scarring process.
In our trial, we have shown that CBD induces a reduction in infarct size that is correlated with EF, is caused by reduced myocardial inflammation, and is not associated with favorable remodeling processes. We therefore assume that the reduced inflammatory response is mediated by an attenuated immunologic response. This is in agreement with other trials that have shown beneficial effects of CBD in a number of inflammatory diseases (2, 4, 6, 26, 33). However, we have yet to establish the precise inflammatory mechanism affected by CBD.
In conclusion, we have shown that CBD causes reduced infarct size in an in vivo rat model of ischemia and reperfusion. Furthermore, it seems that this effect is not direct and may be mediated by a reduced inflammatory response. Thus CBD may be a promising novel treatment for myocardial ischemia.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
We thank Dr. Eli Pikarsky (Pathology Department, Hadassah Hebrew University Medical Center) and Dr. Marielle Scherrer-Crosbie (Massachusetts General Hospital Echo Lab) for advice and Dr. Oren Shibolet (Gastroenterology Department, Hadassah Hebrew University Medical Center) for technical assistance.