Introduction

Top Abstract Introduction Methods Results Discussion References

In anesthetized animals, the major psychoactive constituent of Cannabis sativa, THC, elicits a transient pressor response followed by hypotension and bradycardia (Dewey et al., 1970; Graham and Li, 1973; Estrada et al., 1987). In rats with genetic or surgically induced hypertension, THC significantly lowers mean arterial blood pressure to normotensive levels (Stark and Dews, 1980; Birmingham, 1973; Nahas et al., 1973), and it has also been shown to prevent immobilization stress-induced hypertension (Williams and Ng, 1973).

Cannabinoid receptors have been identified in the rat by radioligand binding and autoradiography (Devane et al., 1988; Herkenham et al., 1990). Subsequently, two cannabinoid receptors have been cloned: the CB₁ receptor located in the brain (Matsuda et al., 1990; Gerard et al., 1991), and in some peripheral organs (Shire et al., 1995; Ishac et al., 1996), and the CB₂ receptor identified in macrophages (Munro et al., 1993; Galiegue et al., 1995). Additionally, a splice variant of the CB₁ receptor, the CB_1A receptor has also been described (Shire et al., 1995).

In 1992, an endogenous cannabinoid receptor ligand, anandamide (arachidonyl-2-ethanolamide), was extracted and purified from porcine brain (Devane et al., 1992). As with THC, anandamide binds to cannabinoid receptors (Vogel et al., 1993; Felder et al., 1993), inhibits adenylate cyclase via an inhibitory G-protein (Vogel et al., 1993), and inhibits voltage-gated N-type calcium channels (Felder et al., 1993). In neurobehavioral assays, anandamide has been shown to mimic THC in terms of inducing catalepsy, hypomotility, hypothermia and analgesia (Fride et al., 1993; Smith, et al., 1994). We previously demonstrated that in anesthetized rats, anandamide causes a pressor response followed by hypotension and mild bradycardia (Varga et al., 1995), and similar effects have been observed in both conscious and anesthetized, spontaneously hypertensive rats (Lake et al., 1997). The hypotensive and bradycardic responses to anandamide and THC are inhibited by the selective CB₁ receptor antagonist, SR141716A (Rinaldi-Carmona et al., 1994), which suggests the involvement of CB₁ receptors (Varga et al., 1995). Activation of these receptors was found to inhibit sympathetic tone via a presynaptic mechanism at peripheral sympathetic nerve terminals (Varga et al., 1996; Ishac et al., 1996). CB₁ receptors in the brain that mediate the neurobehavioral effects of cannabinoids have been characterized not only by their susceptibility to inhibition by SR141716A (Rinaldi Carmona et al., 1994; Compton et al., 1996), but also by their selective interaction with cannabinoid enantiomers (Little et al., 1989). Furthermore, the rank order of potency of various cannabinoid analogs for eliciting neurobehavioral effects correlates well with their binding affinities for the brain cannabinoid receptor (Compton et al., 1992a, 1993). To further test whether the CB₁ receptors mediating neurobehavioral and cardiovascular effects are pharmacologically similar, we analyzed the cardiovascular effects of a series of cannabinoid analogs, including enantiomeric pairs, in urethane-anesthetized rats. The results indicate that the pronounced hypotension and bradycardia induced by structurally different cannabinoids are mediated by receptors that are pharmacologically similar to brain CB₁ receptors mediating neurobehavioral effects. In contrast, the mechanism of the brief pressor response elicited by some cannabinoids remains unclear.

	Methods

Top Abstract Introduction Methods Results Discussion References

Animals. Adult male Sprague-Dawley rats weighing 280 to 400 g were obtained from Harlan (Indianapolis, IN) and were housed in suspension cages with food and water ad libitum. The animals were maintained at 24 to 26°C under a 14:10 hr light/dark cycle and were allowed to acclimate for at least 1 wk before surgery.

Surgical preparation. Anesthesia was induced with diethyl ether and a femoral vein was cannulated for i.v. drug administration. Ether anesthesia was then discontinued and urethane was administered (0.7 g/kg, i.v. + 0.3 g/kg, s.c.). Urethane administered according to this protocol does not depress basal blood pressure and does not interfere with cardiovascular regulatory mechanisms (Maggi and Meli, 1986). The femoral artery was cannulated and the catheter connected to a pressure transducer (Abbott, North Chicago, IL) for continuous monitoring of BP with a physiograph (Astromed, Cortland, NY). HR was monitored via a tachograph preamplifier driven by the pressure wave. The trachea was cannulated with PE-160 tubing to maintain an open airway. Body temperature was maintained at 37 to 38°C throughout the experiments by using a water circulating heating pad (Gaymar Industries, Orchard Park, NY) and rectal thermometer.

Experimental protocols. After a 30-min stabilization period, the animals received either vehicle or SR14716A (3 mg/kg, i.v.). Twenty min later, a single dose of an agonist was administered, and the changes in BP and HR were monitored for 60 min. As agonist effects on BP and HR were long lasting for most of the drugs and doses tested, each animal was tested with a single dose of an agonist, after either vehicle or SR141716A.

Drugs. The chemical structure of the cannabinoid agonists used is illustrated in Figure 1. Anandamide (arachidonyl-2-ethanolamide) and 4-(R)- and 4-(S)-OH-diadduct-⁹-THC (O-502 and O-522, respectively) were synthesized by Dr. Raj Razdan (Organix Inc., Woburn, MA). HU-210 and HU-211 were synthesized by Dr. Raphael Mechoulam (Hebrew University, Jerusalem, Israel). CP-55940 and SR141716A were provided by Dr. John Lowe at Pfizer Central Research. JWH-015 was synthesized by Dr. John Huffman (Clemson University). ⁹-THC was obtained from the National Institute on Drug Abuse. WIN-55212-2 was purchased from Research Biochemicals International (Natick, MA). Palmitoylethanolamine was purchased from BIOMOL Research Laboratories, Inc. (Plymouth Meeting, PA). All drugs were dissolved in 1:1:18 or 1:1:8 (emulphor-ethanol-saline). Emulphor (EL-620, a polyoxyethylated vegetable oil, GAF Corporation, Linden, NJ) is currently available as Alkmulphor. The dosing volume was 0.5 to 1 ml/kg i.v., followed by a catheter line flush of 0.2 ml saline. Injection of the same volume of vehicle had no effect on blood pressure or heart rate. All drugs were injected i.v. as bolus doses over a 10- to 15-sec period.

View larger version (22K): [in this window] [in a new window]   Fig. 1.   Structure of cannabinoid agonists tested. Classical cannabinoids (9-THC, HU-210 and enantiomer HU-211), nonclassical compounds (CP-55940, WIN-55212-2, JWH-015, O-502 and stereoisomer O-522) and endogenous ethanolamide conjugated fatty acids (anandamide and palmitoylethanolamide).

Data analysis. MAP was calculated as 1/3(systolic-diastolic BP) + diastolic BP. Basal MAP and HR for all groups was 102 ± 4.8 mmHg and 325 ± 5.4 bpm, respectively (n = 96). Time-dependent, agonist-induced changes in MAP and HR in the absence or presence of SR141716A were compared using analysis of variance followed by Tukey's post hoc test. The ED₅₀ for each agonist was calculated using ALLFIT, a nonlinear sigmoidal curve-fitting program (DeLean et al., 1977). For anandamide and THC, which produce both a pressor and a subsequent depressor response, ED₅₀s for both components were calculated, using the peak response minus predrug baseline value in both cases. The ED₅₀ for the depressor response to THC was calculated based on the descending portion of the biphasic curve. Correlation analysis and generation of the Pearson product-moment coefficient was performed using the StatView statistical package (Brainpower, Inc., Agoura Hills, CA). Data are presented as mean ± S.E. of the mean.

	Results

Top Abstract Introduction Methods Results Discussion References

THC (0.02-10 mg/kg, fig. 2A) elicited an initial brief pressor response followed by prolonged and marked hypotension and bradycardia. Dose-response relationships for these latter effects were biphasic with maximal hypotension (-62 ± 9 mmHg) observed at 2 mg/kg and maximal bradycardia (-140 ± 24 bpm) observed at 8 mg/kg of THC, beyond which doses the hypotensive and bradycardic effects were less pronounced (fig. 3). ED₅₀ values were 0.27 ± 0.09 mg/kg for the hypotension and 0.62 ± 0.10 mg/kg for the bradycardic effect. Maximal decreases in BP and HR took 15 to 25 min to develop after the lower doses and 4 min after the 2 mg/kg dose, and each lasted more than 60 min. Pretreatment with SR141716A (3 mg/kg) blocked the hypotension and bradycardia elicited by the 4.0 mg/kg dose (fig. 2A). The initial pressor response (ED₅₀ = 2.47 ± 0.85 mg/kg, data not shown) was observed at doses of THC  1 mg/kg, and was not antagonized by SR141716A (fig. 2A). No animal died following any dose of THC.

View larger version (30K): [in this window] [in a new window]   Fig. 2.   Effects of classical cannabinoids on MAP and HR in anesthetized rats. A, Effects of THC (4 mg/kg, i.v.) after pretreatment with vehicle () or SR141716A (3 mg/kg, i.v., ). B, Effects of HU-210 (0.01 mg/kg i.v.) after vehicle () or SR141716A (3 mg/kg, i.v., ) or HU-211 (1 mg/kg i.v.) after vehicle (). Data are presented as mean ± S.E. (n = 5-9; *P < .05 from baseline, #P < .05 between groups).

View larger version (17K): [in this window] [in a new window]   Fig. 3.   Dose-dependent hypotension (A) and bradycardia (B) elicited by HU-210 (), CP-55940 (), WIN-55212-2 (), 9-THC (), anandamide () and JWH-015 (). Each dose represents the mean of three to six experiments. The average S.E. was less than 10% of the mean.

The stereoselectivity of cannabinoid-induced hypotension and bradycardia was evaluated using the enantiomers HU-210 and HU-211. At a dose of 0.01 mg/kg, HU-210 caused pronounced and long lasting hypotension and bradycardia without an initial pressor response (fig. 2B). In dose-response studies, HU-210 appeared to be more potent in causing hypotension than in eliciting bradycardia, the ED₅₀s for these being 0.0020 ± 0.0004 and 0.09 ± 0.01 mg/kg, respectively (fig. 3). Like THC, the effects of HU-210 were long lasting (fig. 2B), and the maximal decrease in MAP and HR exceeded those of THC (fig. 3). Pretreatment with SR141716A (3 mg/kg) blocked the hypotension and bradycardia elicited by the 0.01 mg/kg dose of HU-210 (fig. 2B). One in four animals died following the 0.3 mg/kg dose of HU-210, but none after lower doses. In contrast, the behaviorally inactive isomer HU-211 caused no changes in BP at doses of 0.1 and 1 mg/kg but, consistent with published observations (Mechoulam et al., 1992), caused a slowly developing increase in HR (fig. 2B).

At a dose of 0.3 mg/kg, CP-55940 elicited pronounced hypotension and bradycardia without an initial pressor effect (fig. 4A), which was similar to the effects of HU-210. The ED₅₀ was 0.011 ± 0.002 mg/kg for the hypotension and 0.11 ± 0.05 mg/kg for the bradycardia. Maximal hypotension (-83 ± 3 mmHg) and bradycardia (-218 ± 10 bpm) developed within 4 to 10 min. Pretreatment with SR141716A (3 mg/kg) blocked both the hypotension and the bradycardia elicited by 0.3 mg/kg CP-55,940 (fig. 4A). At a dose of 1 mg/kg, WIN-55212-2 also elicited long lasting hypotension and bradycardia, which developed within 1 to 5 min (fig. 4B). Similarly, WIN-55212-2 caused no initial pressor effect at any of the doses tested and had an ED₅₀ of 0.10 ± 0.02 mg/kg for hypotension and 0.29 ± 0.13 mg/kg for bradycardia (fig. 3). Pretreatment with SR141716A (3 mg/kg) blocked the hypotension, but only partially antagonized the bradycardia elicited by the 0.3 mg/kg dose of WIN-55212-2 (fig. 4B). One in three animals died after the highest dose tested (10 mg/kg), but none after lower doses. The time-response curve for both CP-55,940 and WIN-55212-2 was clearly biphasic, the reasons for which are not clear. The third compound in this class, JWH-015, was less potent than the first two, having an ED₅₀ of 10.1 ± 1.6 mg/kg for hypotension and 16.4 ± 2.2 mg/kg for bradycardia (fig. 3). Pretreatment with SR141716A (3 mg/kg) blocked the long lasting hypotension and bradycardia elicited by 10 mg/kg of JWH-015 (fig. 4C). One in four animals died after the 20- and 30-mg/kg doses. As with CP-55,940 and WIN-55212-2, no initial pressor response was observed at any dose.

View larger version (36K): [in this window] [in a new window]   Fig. 4.   Effects of nonclassical cannabinoids on MAP and HR in anesthetized rats. A, Effects of CP-55940 (0.3 mg/kg, i.v.) after pretreatment with vehicle () or SR141716A (3 mg/kg, i.v., ). B, Effects of WIN-55212-2 (1 mg/kg, i.v.) after vehicle () or SR141716A (3 mg/kg, i.v., ). C, Effects of JWH-015 (10 mg/kg, i.v.) after vehicle () or SR141716A (3 mg/kg, i.v., ). Points represent means ± S.E. from four experiments for each treatment group; *P < .05 from baseline and #P < .05 between groups.

The stereoisomers O-502 and O-522 did not elicit any change in BP or HR at doses of 0.5 or 5 mg/kg. At a dose of 30 mg/kg, a brief (<4 min) pressor response was observed after either O-502 (+30 ± 6 mmHg) or O-522 (+41 ± 4 mm Hg). Both agents also caused moderate, but long lasting tachycardia (O-502: +50 ± 4 bpm; O-522: +45 ± 8 bpm) at the 30-mg/kg dose. Neither of these two cannabinoid isomers binds to the CB₁ receptor or elicits neurobehavioral effects in mice (table 1).

As described earlier (Varga et al., 1995), anandamide elicited an initial transient bradycardia followed by a brief pressor and a more prolonged depressor response that lasted less than 15 min. During this latter hypotensive phase there was also moderate bradycardia (fig. 5). Dose-response studies yielded an ED₅₀ of 2.5 ± 0.3 mg/kg for the hypotension with a peak change of -46 ± 3 mmHg, and an ED₅₀ of 2.5 ± 0.2 mg/kg for the bradycardia with a peak of -43 ± 15 bpm (fig. 3). Pretreatment with SR141716A (3 mg/kg) inhibited the hypotension and parallel bradycardia elicited by the 4.0-mg/kg dose (fig. 5). The initial transient vagal bradycardia/hypotension and the subsequent brief pressor response that preceded the hypotension were not blocked by SR141716A (fig. 5). No animals died following up to 30 mg/kg of anandamide. In contrast, palmitoylethanolamine did not influence BP at doses of 1 to 20 mg/kg and caused a moderate delayed tachycardia at the 4 and 20 mg/kg doses.

View larger version (22K): [in this window] [in a new window]   Fig. 5.   Effects of anandamide (4 mg/kg i.v.) on MAP and HR in anesthetized rats pretreated with vehicle () or SR141716A (3 mg/kg, i.v., ). Points represent mean ± SE (n = 5-9); *P < .05 from baseline and #P < .05 between groups.

As illustrated in figure 3, the rank order of potency of the six analogs that elicited hypotension and bradycardia was the same for these two effects: HU-210  CP-55940 > WIN-55212-2 > THC > anandamide > JWH-015. In earlier studies, similar rank orders of potency have been determined for CB₁ receptor binding in rat brain membranes and for various behavioral effects in mice (table 1). Therefore, we determined the correlation between rat brain CB₁ receptor binding affinity (K_i) and the hypotensive and bradycardic ED₅₀ values of the above six analogs. A significant positive correlation was observed between hypotension (r = 0.97) and bradycardia (r = 0.96) to the binding affinity (fig. 6A). A similar, significant positive correlation could be established between antinociceptive potency in mice and either the hypotensive (r = 0.99) or the bradycardic effects (r = 0.96) in rats (fig. 6B). Table 1 includes the absolute potencies of the various analogs tested for eliciting hypotension and bradycardia as determined in our study, and their potencies in eliciting hypothermia, antinociception, ring-immobility and hypoactivity, as well as their receptor binding affinities.

View larger version (16K): [in this window] [in a new window]   Fig. 6.   A, Linear correlation between log Ki values for displacement of binding of 3H-CP-55,940 to rat brain membranes and log ED50 values for maximal hypotension () or bradycardia (), as shown on the abscissa. The correlation coefficient for hypotension was 0.97 (P < .05) and for bradycardia was 0.96 (P < .05). B, Linear correlation between log ED50 values for antinociception and log ED50s for each compound for hypotension () or bradycardia (). The correlation coefficient for hypotension was 0.99 (P < .05) and for bradycardia was 0.95 (P < .05).

	Discussion

Top Abstract Introduction Methods Results Discussion References

It has long been recognized that the major psychoactive constituent of marijuana, THC, can cause pronounced and long lasting hypotension and bradycardia (Vollmer et al., 1974; Siqueira et al., 1979). More recently, analogous effects have been described for the endogenous cannabinoid ligand, anandamide (Varga et al., 1995; 1996). We have reported that these effects are inhibited by the CB₁ receptor antagonist SR141716A (Varga et al., 1995) and that the inhibitory potency of SR141716A against the hypotensive response to THC or anandamide (Lake et al., 1997) was similar to its inhibitory potency against the neurobehavioral effects of THC (Rinaldi-Carmona et al., 1994). This has suggested that the cannabinoid-induced hypotension and bradycardia are mediated by CB₁-like receptors. In our study, we sought to further test this hypothesis by analyzing the cardiovascular effects of a series of cannabinoid analogs including two pairs of enantiomers in urethane-anesthetized rats. The analogs selected for testing belong to three classes: classical cannabinoids (THC, HU-210 and HU-211), nonclassical cannabinoids (CP-55940, WIN-55212-2, JWH-015, O-502 and O-522), and endogenous ethanolamine-conjugated fatty acids (anandamide and palmitoylethanolamine). This should minimize the possibility that their effects may be related to a common structural feature unrelated to their interaction with cannabinoid receptors. Six of the 10 analogs tested elicited hypotension and bradycardia, and they include compounds from all three classes. With the possible exception of anandamide, the hypotension and bradycardia were very pronounced and of long duration, with a clearly established rank order of potency. The involvement of specific receptors in these effects is indicated by their marked stereoselectivity. The behaviorally active analog HU-210 (Little et al., 1989) displayed the same high potency for hypotension (ED₅₀: 2 µg/kg) as for eliciting neurobehavioral effects (ED₅₀s of 2-4 µg/kg, see table 1), although it was somewhat less potent in eliciting bradycardia. A significantly lower potency of HU-210 to elicit hypotension (ED₅₀  50 µg/kg) was noted in Wistar rats, which may be related to the use of pentobarbital as anesthetic (Vidrio et al., 1996). In contrast, the behaviorally inactive enantiomer, HU-211, did not elicit hypotension at doses up to 1 mg/kg, indicating a minimum of 500-fold stereoselectivity. At the 1-mg/kg dose, HU-211 elicited a moderate tachycardic effect, which is similar to earlier findings (Mechoulam et al., 1992), and thus is probably not mediated by cannabinoid receptors.

The ability of SR141716A to inhibit the hypotensive and bradycardic effects of all six analogs is in agreement with earlier observations with THC and anandamide (Varga et al., 1995; Lake et al., 1997). In a previous study, the AD₅₀ of SR141716A for inhibiting THC- or anandamide-induced hypotension and bradycardia in anesthetized rats was found to be in the range of 0.1 to 0.3 mg/kg iv. (Lake et al., 1997). Our finding that a dose of 3 mg/kg SR141716A completely blocked the effects of most analogs is compatible with the involvement of CB₁ receptors in these effects. This possibility is further supported by the significant positive correlation between the potencies of the six analogs in eliciting hypotension and bradycardia and the binding affinity of these compounds or their analgesic potency in mice (fig. 6, table 1). However, despite this strong correlation there are also some subtle differences. First, it is evident that the absolute potencies of the six analogs tested for causing hypotension are slightly but consistently higher than their potencies for eliciting neurobehavioral effects (table 1). Second, the four most potent hypotensive analogs were appreciably less potent in eliciting bradycardia than hypotension (table 1). Third, the hypotensive ED₅₀s of the three most potent analogs (HU-210 > CP-55940 > WIN-55212-2) vary over two orders of magnitude, whereas their binding K_is for CB₁ receptors are roughly equal (table 1). Finally, a dose of SR141716A that completely blocked the hypotensive response to 1 mg/kg WIN-55212-2, only partially inhibited its bradycardic effect (fig. 4B). Although the existence of a significant receptor reserve for the hypotensive action or differences in receptor coupling mechanisms may explain some of these discrepancies, we cannot exclude the possibility that they may reflect the existence of different isoforms of CB₁ receptors. Such a possibility is also suggested by reports that certain cannabinoid analogs devoid of neurobehavioral effects can elicit hypotension in experimental animals (Adams et al., 1976; Zaugg and Kyncl, 1983), which we were able to confirm and extend in the case of abnormal cannabidiol (J. L. Wiley, E. J. N. Ishac, R. K. Razdan, B. R. Martin, K. Varga and G. Kunos, unpublished results).

Apart from the possibility of CB₁ receptor heterogeneity, it is also likely that the effects of some of the analogs are complex and involve more than one mechanism via CB₁ receptors located in different tissues. The results of our previous studies suggest that the hypotensive action of anandamide is due to inhibition of sympathetic tone (Varga et al., 1995), and a similar mechanism has been proposed earlier for THC (Vollmer et al., 1974; Siqueira et al., 1979). Additional evidence has eliminated the central nervous system, sympathetic ganglia or postsynaptic sites in the vasculature or heart as possible sites of the sympathoinhibitory action of anandamide, and has suggested a presynaptic mechanism at sympathetic nerve terminals (Varga et al., 1996). Indeed, this possibility is strongly supported by the demonstration of a CB₁ receptor-mediated inhibition of exocytotic norepinephrine release induced by anandamide or THC in isolated cardiac tissue or vas deferens, and by the presence of CB₁ receptor mRNA in a sympathetic ganglion (Ishac et al., 1996). However, in our experiments the three most potent hypotensive analogs, CP-55,940, HU-210 and WIN-55212-2 lowered MAP by more than 80 mmHg (see table 1), which far exceeds the decrease in MAP caused by removing sympathetic tone. In a previous study done in the same strain of rats under identical conditions, the hypotensive response to anandamide was no longer present after alpha-adrenergic blockade with 2 mg/kg phentolamine, which lowered MAP from the same resting level by 51 ± 3 mmHg (Varga et al., 1995). Therefore, CB₁ receptors other than those located presynaptically on sympathetic nerve terminals must contribute to the hypotensive effect of the more potent and efficacious cannabinoids.

Anandamide has been shown to bind to CB₁ as well as CB₂ receptors in transfected cell lines (Showalter et al., 1996), whereas a saturated analog, palmitoylethanolamine, has been found to bind to the rat CB₂ receptor (Facci et al., 1995) but not to the CB₁ receptor (Devane et al., 1992; Felder et al., 1993) or to the human CB₂ receptor (Showalter et al., 1996). Our finding that palmitoylethanolamine did not elicit any change in blood pressure and caused moderate tachycardia argues against the possible role of CB₂ receptors in the hypotensive and bradycardic effects. The lack of CB₂ receptor involvement in these cannabinoid-induced effects is also suggested by their potent inhibition by SR141716A, an antagonist with 60-fold lower affinity for CB₂ (K_i: 702 nM) than for CB₁ receptors (K_i: 12.3 nM, Showalter et al., 1996). Conclusive evidence against CB₂ receptor involvement awaits the development of a selective and potent CB₂ receptor antagonist.

Of the 10 compounds evaluated, only four elicited a transient pressor response: anandamide, THC, O-502 and O-522. The anandamide- and THC-induced pressor response was not CB₁ receptor-mediated as it was not attenuated by SR141716A. The involvement of CB₂ receptors is also unlikely, because JWH and WIN-55212-2 are even more potent at CB₂ than at CB₁ receptors (Showalter et al., 1996), yet they produced no pressor effect. O-502 and its stereoisomer O-522, which are both devoid of cannabinoid-like neurobehavioral effects and do not bind to cannabinoid receptors (table 1), did not alter MAP or HR at the lower doses tested and caused a moderate pressor response and tachycardia only at a dose of 30 mg/kg. This suggests that the pressor and tachycardic effects are not mediated by cannabinoid receptors and may result from a nonreceptor- mediated direct effect on vascular smooth muscle, or an indirect effect via release of a vasoconstrictor agent. Sympathetic activation can be excluded as the underlying mechanism, in view of the earlier finding that spinal cord transection or alpha-adrenergic receptor blockade does not attenuate the pressor response to THC (Siqueira et al., 1979) or anandamide (Varga et al., 1995).

In summary, we have found that several cannabinoids, including anandamide, elicit hypotension and bradycardia. These effects are dose-dependent, antagonized by SR141716A, enantioselective and display a similar same rank order of potency as observed for binding to CB₁ receptors in rat brain preparations and for eliciting antinociception in mice. All this indicates that the hypotension and bradycardia elicited by cannabinoids is mediated by a CB₁-like cannabinoid receptor. However, due to subtle differences in agonist potencies for the different effects of cannabinoids and in the inhibitory effects of SR141617A, the possible existence of CB₁ receptor isoforms cannot be excluded. Further studies are aimed at identifying cannabinoid analogs that cause hypotension but are devoid of neurobehavioral effects, and thus may prove of interest in the management of hypertension.