Cannabinoid-induced reduction in antral pacemaker frequency: a telemetric study in the ferret


Address for Correspondence
Nathalie Percie du Sert, St George’s University of London, Cranmer Terrace, London SW17 0RE, UK.
Tel: +020 8725 5356


Background  The gastric myoelectric activity (GMA) is the electrical pacesetter potential, which drives gastric motility. Cannabinoids have broad-spectrum antiemetic and antinauseant activity. Paradoxically, they inhibit intestinal peristalsis and reduce gastric motility but their effect on GMA remains unknown.

Methods  Ferrets were surgically implanted with radiotelemetry transmitters to record GMA, body temperature and heart rate. The effect of WIN 55,212–2 (1 mg kg−1, i.p.), an agonist at the cannabinoid type 1 and 2 receptors was examined in conscious, unrestrained ferrets. WIN 55,212–2 was also compared to the anandamide upregulator URB 597 (5 mg kg−1, i.p.) for a potential to modulate the emetic response and behavioral changes induced by apomorphine (0.25 mg kg−1, s.c.).

Key Results  WIN 55,212–2 decreased GMA frequency (8.1 ± 0.4 cpm, compared to 9.6 ± 0.1 cpm in vehicle-treated animals, n = 6, P < 0.01). Apomorphine induced 9.0 ± 1.6 emetic episodes, WIN 55,212–2 inhibited the emetic response (3.3 ± 1.0 episodes, n = 6, P < 0.05) but URB 597 had no effect (9.0 ± 1.5 episodes). Apomorphine-induced hyperactivity in vehicle-treated animals (6.5 ± 3.6–16.6 ± 4.9 active behavior counts, n = 6, P < 0.01), which was reduced by WIN 55,212–2 (5.0 ± 1.5 counts, n = 6, P < 0.05).

Conclusions & Inferences  WIN 55,212–2 demonstrated clear antiemetic efficacy, which extends the broad-spectrum antiemetic efficacy of cannabinoids to dopamine receptor agonists in the ferret. Our results, however, suggest a more limited spectrum of action for URB 597. WIN 55,212–2 decreased the frequency of the antral electrical pacemaker, which reveals new insights into the mechanism regulating the decrease in motility induced by cannabinoids.


Phytocannabinoids such as delta-9-tetrahydrocannabinol (THC) are found in the marijuana plant (Cannabis sativa L.). Cannabinoids are agonists for cannabinoid type 1 (CB1) and/or type 2 (CB2) receptors,1 the two cannabinoid receptors identified to date, and endogenous ligands for these receptors, termed endocannabinoids, have been identified.2 While CB1 receptors are widely distributed in all major organ systems, particularly in the central and autonomic divisions of the nervous system, CB2 receptors are largely restricted to immune cells.3 Both receptors regulate signal transduction through Gi/o proteins, inducing reduction in cAMP levels, activation of K+ channels, inhibition of Ca2+entry and activation of mitogen-activated protein kinases.4

In the gastrointestinal (GI) tract, CB1 receptors are found on neurons of the myenteric and submucosal plexuses, with the stomach (corpus) and proximal colon expressing the highest density.5 Activation of these receptors delays gastric emptying and decreases small intestinal transit time via an inhibition of peristalsis coupled with an inhibition of contractile activity (see,5 and6 for recent reviews). Activation of CB2 receptors may also modulate GI propulsion, through inhibiting the release of mediators released during disrupted motility caused by inflammation.7

The antiemetic potential of cannabinoids has been demonstrated in humans and animal species. In preclinical studies, phytocannabinoids and endocannabinoids—either exogenously administered or, for endocannabinoids, indirectly upregulated using enzyme inhibitors—showed broad-spectrum antiemetic properties against stimuli such as the cytotoxic anticancer drug cisplatin, analgesic opioids, or motion.8–11 In humans, THC and the synthetic cannabinoids, nabilone, dronabinol and levonantradol, have been shown to reduce both the incidence and intensity of nausea and vomiting induced by anticancer chemotherapy.12–14 Dronabinol also improves the control of nausea when added to standard antiemetic regimen.13 The latter finding is of major interest, since current antiemetic therapies (5-HT3 and NK1 receptor antagonists) reduce greatly the occurrence of emesis, but are less effective against nausea.15 Considering the effect of cannabinoids on GI transit, their effect on nausea seems counter-intuitive, as gastric stasis is classically associated with nausea.16 A tentative explanation points towards the alteration of visceral perception induced by cannabinoids.6

It is also known that a disrupted gastric myoelectric activity (GMA), termed gastric dysrhythmia, occurs in nauseated subjects exposed to illusory self-motion,17 or during pregnancy sickness18 and in a number of other clinical settings associated with nausea, including gastroparesis19 and eating disorders.20 In animal models, a disrupted GMA also occurs following the administration of emetic treatments, including cisplatin, apomorphine, vasopressin, adrenaline, prostaglandin E2, met-enkephalin and glucagon.21–23 The GMA (also termed gastric sow waves) is generated by gastric pacemaker cells: the interstitial cells of Cajal (ICC), which are regulated by neurons of the enteric, sympathetic and parasympathetic nervous systems, and hormones. It has been proposed that gastric dysrhythmias arise from an imbalance in autonomic outflow, a withdrawal of vagal tone and a relative increase in sympathetic activity.24

Whilst cannabinoids are effective against nausea and emesis, their actions on GMA remain unknown. In the present study, we investigated the effect of the CB1 and CB2 receptor agonist, WIN 55, 212–2 on GMA in conscious, unrestrained ferrets. The effect of WIN 55,212 on core temperature and heart rate, which are known to be modulated by CB1 receptors in rodents,25,26 was also simultaneously determined. In addition, we also examined the potential of WIN 55, 212–2 and URB 597, an inhibitor of fatty acid amide hydrolase (FAAH; an enzyme responsible for the hydrolysis of the endo-cannabinoid anandamide), to prevent the emesis induced by apomorphine.

Materials and Methods


Six female fitch or albino ferrets (Mustela putorius furo L.) (0.90 ± 0.05 kg) were obtained from Southland Ferrets (Invercargill, New Zealand). Prior to the experiments, they were housed communally in a temperature-controlled room at 24 ± 1 °C, under artificial lighting with lights on between 06:00 and 18:00 h. Water and pelleted cat food (Tri Pro Feline Formula Cat Food, American Nutrition®, UT, USA) were available ad libitum until the start of the experiments. All animals were then housed individually from the day of surgery to the end of the experiment. The experiments were conducted under the authority of a license provided by the Government of the Hong Kong SAR and approval from the Animal Experimentation Ethics Committee, The Chinese University of Hong Kong.

Telemetry transmitter implantation

Anesthesia was induced with ketamine (20 mg kg−1, i.m.; Alfasan, Woerden, Holland) and maintained with isoflurane (Halocarbon Products Corporation, River Edge, NJ, USA) ∼1.5%, in 3 : 1 O2 to N2O ratio using a custom made face mask and an anesthetic machine (Narkomed 2C, Dräger, Telford, PA, USA). Animals were placed on a heating pad (UCI#390 Analog moist heating pad, Rebirth Medical & Design, Inc., Tapei, Taiwan) and the level of anesthesia was assessed and monitored throughout the surgery by the pedal withdrawal reflex. Following a midline abdominal incision, the antrum was exposed and the biopotential wires of the telemetry transmitter (PhysioTel® ETA-F20, DSI, St Paul, MN, USA) were inserted and sutured in the serosal membrane. The body of the transmitter was inserted in the peritoneal cavity and sutured to the internal abdominal muscle. The abdominal cavity was sutured closed in layers and covered with a permeable spray dressing (Opsite®, Smith and Nephew, London, UK). Prophylactic antibacterial cover was provided with 0.5 mL kg−1 i.m. amoxicillin 20% (Alfasan). Buprenorphine (0.05 mg kg−1, s.c.; Temgesic®, Schering Plough, Welwyn Garden City, UK) was given as a preoperative analgesic 15 min before the induction of anesthesia, and 12 h postsurgery. Recovery was unremarkable and wounds healed within a week.

Experimental design

Following surgery, animals were housed individually in observation cages (W49 × L61 × H49.5 cm). They were allowed to recover for at least 7 days prior to further experimentation. Food was withdrawn 12 h before the start of the experiments, at t = 0 animals were then presented with 15 g of pelleted cat chow which was withdrawn at t = 30 min. At t = 60 min, baseline telemetry signals for GMA, intra-abdominal temperature and heart rate began, then 60 min later, animals were injected with WIN 55,212–2 (1 mg kg−1, i.p.) or vehicle (2% DMSO and 1% Tween 80 in 154 mM NaCl, 2.5 mL kg−1, i.p.) and recordings were continued for a further 3 h. In experiments where antiemetic effects of WIN 55,212–2 and URB 597 were tested, baseline behavior observation began at t = 90 min. At t = 120 min, animals were injected with WIN 55,212–2 (1 mg kg−1, i.p.) or URB 597 (5 mg kg−1, i.p.) or vehicle (2% DMSO and 1% Tween 80 in 54 mM NaCl, 2.5 mL kg−1, i.p.). At t = 150 min, the emesis-inducing, dopamine D2 receptor agonist, apomorphine (0.25 mg kg−1, s.c.) was injected and behavior observation was continued for a further 1 h. Animals received pretreatment with WIN 55,212–2 or URB 597 or vehicle, followed by apomorphine in a randomized order; each animal received the three different pretreatments with 4 days recovery between two experiments. The doses and route of administration used have been shown to be pharmacologically relevant in the ferret to induce (apomorphine27) or prevent emesis (WIN 55,212–2, URB 59710). Emesis was characterized by rhythmic abdominal contractions that were either associated with the oral expulsion of solid or liquid material from the GI tract (i.e., vomiting), or not associated with the passage of material (i.e., retching movements). An episode of retching and/or vomiting was considered separate when the animal changed its location in the observation cage, or when the interval between retches and/or vomits exceeded 5 s.28 The latency was defined as the time between the administration of the drug and the first emetic episode. Behaviors quantified as active behaviors were: walk (period of spontaneous locomotor activity lasting at least 1 s), rear (animal standing up on its hind legs) and jump; the frequency of such behaviors were cumulated and expressed as the active behavior count. Inactive behaviors were curl-up (usual position in which a ferret sleeps), lie down (either flat on their stomach, on their back or on their side), sit still (half recumbent posture, usually observed when the animal was observing something outside its cage); inactive behaviors were quantified as the total time during which such behaviors were observed.


WIN 55,212–2 mesylate (Sigma-Aldrich, St. Louis, USA) and URB 597 (Cayman Chemical, Ann Arbor, USA) were dissolved in 2% DMSO (Sigma-Aldrich) and 1% Tween 80 (Sigma-Aldrich) in 154 mmol L−1 NaCl, 2.5 mL kg−1. Apomorphine hydrochloride (Sigma-Aldrich) was dissolved in sodium disulfite (526 μmol L−1, Riedel-de Haën, Seelze, Germany) and injected in a volume of 0.5 mL kg−1, s.c.

Telemetry system and analysis of the data

A DSI Dataquest® A.R.T. telemetry system (Data Science International, St Paul, MN, USA) was used. The GMA and temperature were recorded using PhysioTel® ETA-F20 Mouse Transmitters. Telemetric signals were recorded via two receiver plates (PhysioTel® RPC-1) placed under the cages. The receivers were connected to a PC desktop computer via a matrix (Dataquest ART Data Exchange Matrix). An ambient pressure reference monitor (APR-1) was connected to the exchange matrix. Data was recorded with Dataquest Acquisition software (DQ ART 4.0, Data Science International). Analysis of telemetry recordings was carried out using Spike2® (version 6.11, Cambridge Electronic Design, Cambridge, UK).

Gastric myoelectric activity recordings

The GMA signal was recorded with a sampling frequency of 1000 Hz and analyzed as described previously.21 Briefly, two successive low pass FIR filters with a cut-off frequencies of 2.5 Hz and 0.3 Hz were applied, and the sampling frequency was reduced to 10.24 Hz. The following parameters were used to characterize the GMA: (i) the dominant frequency (DF, frequency bin with the highest power in the 0–15 cpm range); (ii) the repartition of power in the bradygastric, normal and tachygastric ranges (i.e., bradygastria, normogastria and tachygastria). The DF during a 1 h baseline was used to define the normal range in each animal, the limits of each range were then defined as follows: bradygastria: 0 to (DF−1) cpm, normogastria: DF ± 1 cpm, tachygastria: (DF + 1) to 15 cpm. To investigate the general effect of WIN 55,212–2 on the GMA, Fast Fourier Transforms were computed on successive 10 min epochs to construct the profiles of GMA repartition and the data were averaged in 30 min blocks for statistical analysis. Heart rate was also extracted from the raw GMA signal.

Statistical analysis

For the effect of WIN 55,212–2 on GMA, heart rate, behavior and intraperitoneal temperature, differences between treatment groups were compared using repeated measures two-way anovas (factors: treatment and time) followed by Bonferroni post-tests. The effect of WIN 55,212–2 and URB 597 on apomorphine-induced emesis were assessed by repeated measures one-way anova followed by Bonferroni post-tests. All statistical analysis were performed using Prism® version 5.0 (GraphPad software, San Diego, USA), P < 0.05 was taken as statistical significance.


Effect of WIN 55,212–2 (1 mg kg−1, i.p.) on GMA, heart rate and intra-abdominal temperature

During the 1 h prior to the injection of vehicle, baseline DF was 8.9 ± 0.3 cpm (n = 6), the percentage of bradygastria, normogastria and tachygastria was 11.43 ± 3.83%, 65.2 ± 6.1% and 15.7 ± 2.1%, respectively (Fig. 1A–D). Following the intra-peritoneal injection of vehicle, a slow increase in DF was observed, which was significant from 30 min postinjection until the end of the observation period; 3 h postinjection the DF was 9.7 ± 0.2 cpm (P < 0.01; Fig. 1D). No changes were observed in the percentage of bradygastria and normogastria (P > 0.05), but compared to baseline, an increase in tachygastria was detected in the 3rd hour following the injection of vehicle (35.3 ± 10.6%, P < 0.05; Fig. 1C). Prior to the injection of WIN 55,212–2, the baseline DF was 8.7 ± 0.3 cpm (n = 6), and the GMA was characterized as follows: 8.5 ± 2.3% bradygastria, 60.5 ± 6.1% normogastria and 21.1 ± 3.7% tachygastria. Following the injection of WIN 55,212–2, the DF was initially increased to 9.2 ± 0. 2 cpm (P < 0.05 compared to baseline) and then decreased to 8.1 ± 0.4 cpm at 1.5 h postinjection; the decrease in DF was not statistically significant compared to baseline but it was significantly reduced compared to the corresponding values in vehicle-treated animals from 30 min to 2 h postinjection (P < 0.05; Fig. 1D). WIN 55,212–2 did not induce any changes in the repartition of power in the normogastric and tachygastric ranges (P > 0.05) but significantly increased the percentage of bradygastria to 30.2 ± 12.3% during the second hour postinjection (P < 0.01 compared to baseline; Fig. 1A–C). An example of the effect of WIN 55,212–2 (1 mg kg−1, i.p.) on the GMA is shown in Fig. 2. No differences in any of the GMA parameters were observed at baseline (P > 0.05).

Figure 1.

 Effect of a single injection of WIN 55,212–2 (1 mg kg−1, i.p.) and vehicle (2.5 mL kg−1, i.p.) on gastric myoelectric activity (A, B, C and D), intra-abdominal temperature (E) and heart rate (F) in the ferret. The graphs in the top panel show the percentage repartition of power in the bradygastric (A), normogastric (B) and tachygastric (C) ranges following a spectral analysis, graph (D) shows the antral dominant frequency. On all graphs, the vertical dotted line represents the time of injection. Veh: vehicle, WIN2: WIN 55,212–2. Results are reported as mean ± SEM (n = 6), statistical significance was assessed with repeated measures two-way anovas (factors: time and treatment) followed by Bonferroni post-tests. Statistically significant differences identified by the Bonferroni post-tests are indicated as follow: **P < 0.01 difference between the WIN 55,212–2 and vehicle treatments; +P < 0.05, ++P < 0.01, +++P < 0.001 difference with baseline in the WIN 55,212–2-treated animals; °P < 0.05, °°P < 0.01 differences with baseline in the vehicle-treated animals.

Figure 2.

 Running spectral analysis showing the effect of a single injection of WIN 55,212–2 (1 mg kg−1, i.p.) on the gastric myoelectric activity repartition in the ferret. Data obtained from one animal, each line represents 10 min of data and 5 new minutes are added between two consecutive lines (50% overlap). The black horizontal line indicates the time of injection of WIN 55,212–2 (WIN2), the plain vertical line indicates the dominant frequency (DF) during baseline and the two dotted line on either side of the DF represent the limits of the normogastric range.

The baseline intra-abdominal temperature was 38.6 ± 0.1 °C (n = 6) prior to the administration of vehicle, and 38.6 ± 0.1 °C (n = 6) prior to injection of WIN 55,212–2. No changes were observed following the injection of vehicle (P > 0.05) but a 2 °C reduction in temperature was observed in the WIN 55,212–2-treated animals (Fig. 1E). The temperature was statistically lower than during baseline from 30 min postinjection until the end of the observation period (P < 0.01, 36.4 ± 0.7 °C during the third hour postinjection) and differences between the WIN 55,212–2 and vehicle treatments were observed from 1 h postinjection (p < 0.001, see Fig. 1E).

Baseline heart rate was 242.8 ± 5.6 cpm (n = 6) prior to the injection of vehicle and 245.7 ± 11.5 cpm (n = 6) prior to the injection of WIN 55,212–2. The injection of vehicle did not induce any change in heart rate (P > 0.05 compared to baseline) and no differences were detected between the vehicle and WIN 55,212 treatments. However, the injection of WIN 55,212–2 induced a 20% reduction of the baseline heart rate; this effect was statistically significant immediately after the injection until 2.5 h postinjection (P < 0.01 compared to baseline, see Fig. 1F).

Effect of WIN 55,212–2 (1 mg kg−1, i.p.) and URB 597 (5 mg kg−1, i.p.) on apomorphine-induced emesis and GMA disruption

Following pretreatment with the vehicle for URB 597 and WIN 55,212–2, apomorphine (0.25 mg kg−1, s.c) induced 9.0 ± 1.6 episodes of emesis, comprising 46.2 ± 6.5 retches and 5.3 ± 1.0 vomits, after a latency of 2.6 ± 0.4 min (n = 6, see Fig. 3). Pretreatment with URB 597 (5 mg kg−1, i.p.) had no effect on the emetic response; apomorphine induced 9.0 ± 1.5 episodes comprising 40.4 ± 8.7 retches and 4.7 ± 1.3 vomits following a latency of 3.1 ± 0.5 min (n = 6). After pretreatment with WIN 55,212 (1 mg kg−1, i.p.), the latency to apomorphine-induced emesis was unchanged (3.1 ± 0.5 min) but the number of episodes was reduced to 3.3 ± 1.0 (P < 0.05) and the number of retches and vomits to 20.0 ± 5.0 and 1.0 ± 0.8, respectively (n = 6, P < 0.05). A detailed analysis of the episodes revealed that vehicle-treated animals had 0.6 ± 0.1 vomits per episode. This number was unchanged following treatment with URB 597 (0.5 ± 0.1 vomits per episode) but reduced following WIN 55,212 (0.1 ± 0.1 vomits per episode, P < 0.05). The number of retches per episode, however, was not reduced following treatment with URB 597 (4.5 ± 0.6 retches per episode) or WIN 55,212 (6.4 ± 0.9 retches per episode) compared to vehicle-treated animals (5.4 ± 0.4 retches per episodes, P > 0.05). Thus, WIN 55,212 reduced the total number of episodes, and differentially reduced the number of vomits but not the number of retches per episode.

Figure 3.

 Effect of a single injection of WIN 55,212–2 (1 mg kg−1, i.p.) or URB 597 (5 mg kg−1, i.p.) on the emetic response induced by apomorphine (0.25 mg kg−1, s.c.) in the ferret. Veh: vehicle, WIN2: WIN 55,212–2, URB: URB 597. The latency to the onset of emesis (top panel) and the number and retches and vomits (lower panel) induced by apomorphine as reported as mean ± SEM (n = 6), statistical significance was assessed with repeated measures one-way anovas followed by Bonferroni post-tests and is indicated as *< 0.05.

Gastric myoelectric activity characteristics and significant differences with baseline and between groups are reported in Table 1; no differences in the GMA repartition were observed between the three treatment groups but the DF was significantly reduced following the administration of WIN 55,212–2 (P < 0.01).

Table 1.   Effect of a single injection of WIN 55 212–2 (1 mg kg−1, i.p.), URB 597 (5 mg kg−1, i.p.) or vehicle (2.5 mL kg−1, i.p.), followed by apomorphine on the GMA in the ferret
 BaselineCB/vehApomorphine 0–30 minApomorphine 30–60 min
  1. Results are reported as mean ± SEM (n = 6), statistical significance was assessed with repeated measures two-way anovas (factors: time and treatment) followed by Bonferroni post-tests. Statistically significant differences identified by the Bonferroni post-tests are indicated as follow: *< 0.05, **< 0.01, ***< 0.001 difference with baseline; < 0.01 difference with the vehicle treatment. GMA, gastric myoelectric activity; DF, dominant frequency; CB/veh, cannabinoids or vehicle.

 Vehicle71.0 ± 6.147.3 ± 10.8*39.4 ± 9.4**65.3 ± 8.0
 URB 59768.0 ± 7.146.1 ± 5.237.0 ± 6.4***65.2 ± 10.2
 WIN 55,21263.3 ± 7.851.7 ± 8.029.9 ± 9.6**44.9 ± 10.5
 Vehicle9.2 ± 0.39.8 ± 0.39.5 ± 0.39.4 ± 0.2
 URB 5979.0 ± 0.39.7 ± 0.29.1 ± 0.49.4 ± 0.2
 WIN 55,2129.4 ± 0.19.6 ± 0.18.8 ± 0.58.1 ± 0.4***

Behavioral effect of WIN 55,212–2 (1 mg kg−1, i.p.), URB 597 (5 mg kg−1, i.p.) and apomorphine (0.25 mg kg−1, s.c.)

Ferret behavior is summarized in Fig. 4. In the 30 min baseline prior to injection of the vehicle, ferrets had 6.5 ± 3.6 counts of active behaviors and spent 21.6 ± 2.1 min inactive; no differences were observed during baseline between any of the three groups and no changes were detected following the injection of the vehicle (compared to baseline, Fig. 4). However, in the 30 min that followed the injection of apomorphine, the active behavior count was increased to 16.6 ± 4.9 (P < 0.01 compared to baseline) and the time spent inactive decreased to 9.5 ± 3.0 min (P < 0.01 compared to baseline). No differences were detected between the URB 597 and vehicle treatments; in the 30 min baseline ferrets had 3.3 ± 1.7 counts of active behavior and spent 23.8 ± 3.9 min inactive. No changes were observed following treatment with URB 597. However, following the injection with apomorphine the active behavior count increased to 13.3 ± 2.6 (P < 0.05 compared to baseline) and the time spent inactive reduced to 8.6 ± 1.7 min (P < 0.05 compared to baseline). In the 30 min preceding the injection of WIN 55,212–2, ferrets had 8.3 ± 3.4 counts of active behavior and spent 15.2 ± 3.6 min inactive. Following the injection of WIN 55,212–2 the activity count was unchanged (P > 0.05 compared to baseline) but the amount of time spent inactive was increased to 26.1 ± 1.5 min (P < 0.01 compared to baseline). Following the injection of apomorphine, the activity count and the time spent inactive were not different from baseline (Fig. 4B,C). No differences were observed between the treatment groups during baseline or in the 30 min preceding the administration of apomorphine but in the 30 min following apomorphine the activity count was reduced in WIN 55,212–2-treated (5.0 ± 1.5 counts) compared to vehicle-treated ferrets (P < 0.05); no other differences between the treatment groups were observed.

Figure 4.

 Effect of a single injection of WIN 55,212 (1 mg kg−1, i.p.) or URB 597 (5 mg kg−1, i.p.) followed by apomorphine (0.25 mg kg−1, s.c.) on the ferret behavior and intra-abdominal temperature. A: cumulative count of active behaviors (i.e., walk, jump, rear) per 30 min, B: total time spent immobile (i.e., curl-up, lying down, sitting still) per 30 min, C: intra-abdominal temperature. The dotted lines represent the time of injection of the cannabinoids or vehicle (CB) and apomorphine (apo). Results are reported as mean ± SEM (n = 6), statistical significance was assessed with repeated measures two-way anovas (factors: time and treatment) followed by Bonferroni post-tests. Statistically significant differences identified by the Bonferroni post-tests are indicated as follow: *< 0.05, **< 0.01, ***< 0.001 difference between the WIN 55,212–2 and vehicle treatments; ++< 0.01, +++< 0.001 difference with baseline in the WIN 55,212–2-treated animals; ψP < 0.05, ψψP < 0.01, §§< 0.01, §§§< 0.0001 difference with baseline in the URB 597-treated animals.

The baseline intra-abdominal temperature was 38.8 ± 0.1 °C prior to the injection of vehicle, and 38.7 ± 0.2 °C and 38.5 ± 0.2 °C prior to the administration of WIN 55,212–2 and URB 597, respectively (Fig. 4C; P > 0.05). No changes were observed immediately following the cannabinoid or vehicle treatment (P > 0.05) but in the second 30-min period following the administration of apomorphine, a significant reduction in temperature was observed in the vehicle-treated animals (38.2 ± 0.2 °C, P < 0.05) and in the URB 597-treated animals (37.6 ± 0.5 °C, P < 0.01) compared to baseline values. In the WIN 55,212–2-treated group, following apomorphine injection, the temperature was statistically lower than during baseline (P < 0.001) and lower than the vehicle-treated group (P < 0.05), during the entire observation period (36.2 ± 0.7 °C during the second 30 min period post apomorphine administration, see Fig. 4C).


The antiemetic effects of WIN 55,212–2 reported herein are consistent with similar effects of WIN 55,212–2 and other cannabinoid receptor agonists against opioid and cisplatin-induced emesis in the ferret.9,10,29 Consistently, cannabinoids have also been shown to suppress motion, cisplatin, and lithium chloride-induced emesis in other commonly used emetic species such as the house musk shrew (Suncus Murinus11,30,31). The present study is, however, the first to report antiemetic effects of a cannabinoid against a dopamine receptor agonist in the ferret, extending the broad-spectrum antiemetic activity of cannabinoids in this species, and consistent with findings in the least shrew (Cryptotis Parva32). Surprisingly, URB 597, which is reported to inhibit endogenous anandamide degradation, did not antagonize the emetic effect of apomorphine. This finding contrasts with reports of the antiemetic activity of URB 597 against morphine-6-glucoronide in the ferret10 and nicotine and cisplatin in S. murinus.33 In the least shrew, however, URB 597 proved ineffective against either apomorphine or cisplatin.34 Taken together, our findings might therefore support a more limited spectrum of action for URB 597, particularly as the same dose and mode of administration was used in both ferret studies (this study10). In our study, WIN 55,212–2 had a differential effect on the number of retches and vomits per episode. Following vehicle treatment 60% of the episodes culminated in a vomit, whereas vomiting was observed in only 10% of the episodes following WIN 55,212–2. However, the number of retches per episode was non-significantly increased by ∼18%. This finding corroborates a study in cats by McCarthy and Borison, who reported that vomiting induced by apomorphine was selectively blocked by the cannabinoid levonantradol.35 This requires further study to investigate the dose-response relationship of the effect of cannabinoid receptor agonists on retching and vomiting as cannabinoids may provide a useful tool to investigate the central mechanism involved in the transition between retching and vomiting; a likely effect would be to prevent the inhibition of crural diaphragm contraction which marks the transition between retching and vomiting.36 Crural diaphragm inhibition also occurs during gastro-esophageal reflux, which is also reduced by cannabinoids.37

Modulation of intestinal and gastric motility by cannabinoids is well characterized and is evidenced in vivo by a delay in gastric emptying and increased transit time observed in humans38 and rodents.39,40 A major finding of this study was that WIN 55,212–2 reduced the frequency of the GMA. To the best of our knowledge, this is the first report of a negative chronotropic effect of a cannabinoid agonist on the gastric slow waves, which sheds further light on the mechanism underlying the inhibition of gastric motility. Our results are in line with the decrease in antral contraction frequency observed in rodents,41 suggesting that the effect of cannabinoids on the motility is, at least partly, mediated via pacing of the gastric electrical rhythm. It, however, remains possible that cannabinoids act by disrupting the coupling between electrical activity of ICCs and the mechanical activity of the smooth muscle cells. Certainly, cannabinoids, via prejunctional CB1 receptors, have been shown to reduce intestinal smooth muscle cell contractility by inhibiting cholinergic, as well as non-adrenergic and non-cholinergic transmission, in the enteric nervous system.42 However, the findings of the present study suggest an additional component to the proposed mechanism of regulation by cannabinoids of gastric motility.

Studies on the mechanism regulating the frequency of the gastric slow waves have revealed that although acetylcholine and other muscarinic agonists have a positive chronotropic effect on the slow waves via activation of muscarinic M3 receptors, blockade of these receptors does not alter the spontaneous frequency of the slow waves.43–45 These findings imply that ICCs are not tonically active and a reduction of acetylcholine transmission is not sufficient to reduce the frequency of the slow waves. There is, however, evidence that WIN 55,212 and other CB1 receptor agonists, independently of adenylate cyclase, inhibit voltage-activated inward calcium currents via ϖ-conotoxin-sensitive (N, P or Q-type) calcium channels.46 In the ICCs, blockade of calcium entry though voltage-dependant calcium channels (excluding L-type) would be expected to have a negative chronotropic effect.47 Following this reasoning, we postulate that the effect of WIN 55,212–2 on GMA frequency is mediated by a direct action on ICC activity. Cannabinoid type 1 receptors have been identified throughout the GI tract5 and although CB1 receptors were not found to be colocalized with the ICC marker c-kit the in the mouse colon,48 to the best of our knowledge there is no evidence either for, or against, the presence of CB receptors on gastric ICCs, in any species. An alternative explanation is that WIN 55,212–2 acts via an inhibitory neurotransmitter such as nitric oxide (NO). Cannabinoid type 1 activation has been shown to initiate inhibitory motor pathways of the enteric nervous system, resulting in the release of NO,49 which has been identified as a regulator of slow wave frequency.50,51

On the other hand, it is noteworthy that the reduction of slow wave frequency observed in the present study was associated with a reduction of body temperature. A similar correlation between hypothermia and reduction of slow wave frequency has been reported in anaesthetized rats52 and the isolated canine antrum.53 In humans, the ingestion of a low temperature meal induces transient reduction in postprandial slow wave frequency.54 Thus, further studies on GMA using URB 597, which had no effect on body temperature and emesis, need to be conducted to explore the role of endogenous cannabinoids on GMA and its relationship with emesis. As part of the study, the effect of cannabinoids on the GMA disturbance induced by apomorphine was also tested. Unfortunately, in that series of experiments, administration of the vehicle for URB 597 and WIN 55,212–2 disrupted the GMA; it significantly reduced normogastria (see Table 1) and increased tachygastria (data not shown). The administration of apomorphine 30 min later further reduced normogastria, which is consistent with previous findings.21 However, the level of GMA disruption observed in all three groups prior to the administration of apomorphine compromised the interpretation of the effects of URB 597 and WIN 55,212–2 on apomorphine-induced dysrhythmia. In this study we chose to inject cannabinoids and vehicle intra-peritoneally because of the poor solubility of some of the compounds used (e.g., URB 597) and to match the design of other ferret emesis studies.10 This might, however, be a confounding factor for this type of study and other routes of administration (e.g., i.v. or s.c.) may be preferable. It is somewhat paradoxical that cannabinoids are potent inhibitors of the emetic reflex and yet they reduce gastric motility, which tends to be associated with nausea and vomiting.55 We showed here that the antiemetic effect of WIN 55,212–2 was associated with a reduction of antral DF. Although the significance of this finding is unknown, it is in agreement with a study in S. murinus showing that animals which did not develop emesis to a motion stimulus had a lower DF than animals which had an emetic response,56 suggesting that a lower DF may be protective against an emetic challenge.

It is noteworthy that WIN 55,212–2 is a non-specific cannabinoid agonist and has been reported to have an action on non-cannabinoid receptors. Indeed, in addition to activation of CB1 and CB2 receptors1 WIN 55,212–2 exhibits a non CB1/CB2 mediated CNS effect,57 activates the transcription of the peroxisome proliferator-activated receptors PPARα and PPARδ58 and functionally inhibits the vanilloid receptor TRPV1.59 It is therefore conceivable that the GMA effects observed in the present study are mediated by receptors other than cannabinoid. Further studies, using selective CB1 and CB2 receptor agonist, and selective CB1 and CB2 receptor antagonists are warranted to confirm the involvement of cannabinoid receptors. However, it should be pointed out that whilst we cannot exclude the involvement of CB2 receptors in our study, reductions in GMA and emetic episodes, induced by WIN 55212–2 were accompanied by hypothermia, bradycardia and a reduction of apomorphine-induced hyperactivity all of which are associated with CB1 activation in rodents.1,60 Administration of apomorphine decreased the time spent immobile and increased the frequency of active behaviors such as rearing and jumping. Indeed, apomorphine has been reported to increase locomotor activity and velocity of motion, as well as the frequency of rearing in the ferret.27 This effect relies on activation of dopamine D1 and D2 receptors in the nucleus accumbens61,62 and is independent of emesis, which is mediated by dopamine D2 receptors in the area postrema.27 Interestingly, WIN 55,212–2 abolished apomorphine-induced hyperactivity, consistent with findings in the rat in which the locomotor hyperactivity induced by the selective D2 receptor agonist quinpirole is antagonised by CB1 receptor activation.63,64 This effect could be due to the antagonistic interaction of CB1 and D2 receptors within CB1/D2 heterodimeric complexes in the striatum.64 It should be mentioned that, in parallel with its lack of effect on body temperature, URB 597 also had no effect on behavioral changes associated with apomorphine.

Apomorphine alone also induces hypothermia, which is well documented in rodents and thought to be mediated by dopamine D2, and possibly D1, receptors.65 Our findings in the ferret extend existing data obtained from the rat, both in terms of the magnitude change and onset of action.66 Further reduction in body temperature was seen in ferrets given WIN 55,212–2 prior to apomorphine, although partial recovery of temperature was observed, which was not seen in animals administered with WIN 55,212–2 alone (c.f. Fig. 4CvsFig. 1E). This might suggests an interaction of the cannabinoid and dopamine systems in the regulation of body temperature. Such an interaction has been observed in rodents, where dopamine receptor agonists and antagonists potentiate and antagonize cannabinoid-induced hypothermia, respectively.67 Further investigation of the interaction between the dopamine and cannabinoid systems in thermoregulation is thus warranted, especially in non-rodent species.

In conclusion, WIN 55,212–2 decreases the electrical pacemaker frequency in the antrum of ferrets, which provides new insights into the mechanism by which cannabinoids regulate gastric motility. Furthermore, WIN 55212–2, but not URB 597, is effective against emesis induced by a dopamine receptor agonist in ferrets, confirming the broad-spectrum, antiemetic efficacy of direct agonists of cannabinoid receptors.


The authors would like to thank Jack Wu and Man K. Wai for excellent technical assistance, and Cambridge Electronic Design, especially Steven Clifford for providing the Spike2 scripts. NPdS was sponsored by a Wellcome Trust Value In People award. NPdS, W-SVH, JAR and PLRA designed the research study; NPdS and W-SVH performed the research; NPdS analyzed the data; NPdS, W-SVH, JAR and PLRA wrote the paper.

Competing Interests

The authors have no competing interests.