2016 Volume 39 Issue 5 Pages 747-753
Cannabinoids are the active ingredients in marijuana, which is among the most widely used addictive drugs despite the well-documented harmfulness related to its abuse. The mechanism underlying cannabinoid addiction remains unclear, which is attributed partially to the difficulty in behavioral testing of high-dose cannabinoids using the conditioned place preference (CPP) model. Here, we optimized conditions for establishing CPP with the synthetic cannabinoid HU210 intraperitoneally administered at a high dose. We found that the natural place preference of rats could be exploited for establishing a biased CPP model, and that the adverse effect of HU210 could be ameliorated by adding four daily pre-injections before the conditioning program. Thus, 0.1 mg/kg HU210 induced CPP when pre-injections were administered before traditional conditioning with HU210 administration paired with the non-preferred compartment. The present study provides a useful CPP model for behavioral measurement of the rewarding effects of cannabinoids.
Psychotropic drugs are characterized by their ability to elicit a rewarding effect, which eventually leads to abuse and addiction.1) Behavioral testing of animals is the classic approach for monitoring the rewarding effects of these drugs and is indispensable for dissecting the mechanisms underlying drug addiction.2,3) The conditioned place preference (CPP) model detects the rewarding effect of drugs after a conditioning procedure that builds an enforced connection between the motivational effect of addictive drugs and the place where the drug is administered.4) The CPP model is among the most popular paradigms for drug reward research and has been used in numerous studies examining many addictive drugs.4–6)
Cannabinoids have been well-documented for both their addictive effects and the harmfulness of their abuse.7) Although the rewarding effect of cannabinoids, for example, Δ9-tetrahydrocannabinol (Δ9-THC), HU210, and other cannabinoid agonists, has been studied using the CPP paradigm, the results of these test are controversial.3,6,8) This controversy may reflect the multiple experimental factors that affect the outcome of drug-place coupling, and the involvement of different brain regions and neurotransmitters, including the ventral tegmental area dopamine (DA) neurons as well as the glutamatergic and γ-aminobutyric acid (GABA)ergic inputs to the nucleus accumbens and ventral tegmental area, in the neural circuitry of cannabinoid reward.9–11) In addition, the potential adverse effects on learning and memory and the reported anxiolytic property of cannabinoids may also underlie these discrepancies.12,13) In the present study, the potential factors that may affect cannabinoid CPP models, such as cannabinoid doses, animal strains, and conditioning protocols, were considered and analyzed. We found CPP in response to cannabinoid administration after drug pre-injections combined with a biased protocol design. The results of an elevated plus-maze (EPM) test confirmed that the place preference shift was not elicited by the anxiolytic but by the rewarding effect of a cannabinoid.
Male Long Evans (LE) and Sprague-Dawley (SD) rats (3 months old), weighing 200 to 250 g, were purchased from Charles River (Wilmington, MA, U.S.A.). Rats were housed two per cage and maintained under a 12 h light/dark cycle (lights on at 7:00 a.m.) at 22±1°C 1 week before the start of the experiment. Food and water were available ad libitum. The experimental procedures were approved by the Canadian Council on Animal Care and the Institutional Animal Care and Use Committee of Shaanxi Normal University.
HU210 was purchased from Tocris Bioscience (Bristol, U.K.) and dissolved by sequentially adding dimethyl sulfoxide (DMSO), Tween 80, and saline at a ratio of 1 : 1 : 8 to reach a final concentration of 0.1 mg/mL.
Classic CPP Test ProtocolThe CPP test apparatus and recording software were products of TSE Systems Inc. (Midland, MI, U.S.A.). The apparatus was a black rectangular polyvinyl chloride chamber (75×22×30 cm) divided into three chambers separated by guillotine doors. The two end chambers (30×22×30 cm) used for conditioning were connected by a smaller center chamber (15×22×30 cm). The end chambers were distinguished by the styles of walls and floor, and both chambers were equipped with dimmable lights. Fifteen infrared beams spaced 5 cm apart monitored the motion of the rat. The infrared sensors communicated with a computer every 100 ms through an interface. All experimental events were controlled and recorded automatically by the computer.
LE or SD rats were handled twice daily for 5 d before the CPP test. The classic CPP test was conducted in three phases, pre-test, conditioning, and test phases, as previously described.14) Briefly, rats were allowed to freely explore the entire apparatus for a 15 min session on the pre-test day (day 1). During each of the four 2 d conditioning sessions (days 2 to 9), rats were intraperitoneally (i.p.) injected with HU210 (treatment group, 0.05 or 0.1 mg/kg) or vehicle (control group) and placed 10 min later in the end chamber designated “drug” for 30 min. On the second morning, all rats received an i.p. injection of vehicle, and were placed in the other end chamber designated “non-drug” for 30 min. For the test phase (day 10), rats were tested under the conditions used for the pre-test but without an HU210 or vehicle injection. The amount of time spent in each chamber was recorded. The preference scores were calculated by subtracting the time spent in the drug conditioned chamber during the pre-test phase from the time spent in the same side during the test phase.
Optimization of the Classic CPP Procedure(1) Prolonged conditioning time: The classic CPP procedure was followed except that the time rats remained in the chambers during conditioning sessions was prolonged from 30 to 60 min. (2) Pre-injection(s): The classic CPP procedure was followed except that rats received a single or four daily i.p. injections of 0.1 mg/kg HU210 or vehicle 24 h before the conditioning sessions. (3) Rat strains: The classic CPP procedure was followed except that SD rats rather than LE rats were used. (4) Biased design: In the pre-test phase, the environmental parameters of the chambers, that is, wall colors, light intensities, and floor styles, were changed so that rats developed a natural preference for a certain end chamber. Rats that spent more time (over 150 s) in one end chamber than the other were included in this experiment and divided randomly into two groups. The selected rats were pre-injected i.p. with 0.1 mg/kg HU210 (group 2) or vehicle (group 1) once daily for 4 d. During each of the four conditioning sessions, rats were i.p. injected with 0.1 mg/kg HU210 (group 2) or vehicle (group 1) on the first morning and placed 10 min later in the non-preferred chamber for 30 min. On the second morning, each rat received an i.p. injection of vehicle and was placed 10 min later in the preferred chamber for 30 min. The rats conditioned in this manner then proceeded to the test phase, which followed the classic CPP procedure (Fig. 1).
All rats were subjected to the EPM test immediately after the CPP test. The EPM apparatus consisted of two open (50×10 cm) and two closed (50×10×40 cm) arms arranged such that the open arms were opposite one another. The maze was elevated 50 cm above the floor, and the tests were conducted under dim red light illumination (44 lx). Rats were placed in the central zone facing an open arm, and their behaviors were videotaped during the 5 min test period. The percentage of time spent in the open arms and percentage of entries into open arms were calculated as follows: % Open arm time=[(time in open arm/time in either arm)]×100; % Open arm entries=[(number of entries into the open arm/number of entries into either arm)]×100.
Western BlottingThe hippocampus was isolated and homogenized in radio immunoprecipitation assay (RIPA) lysis buffer containing 50 mM Tris (pH 7.4), 150 mM NaCl, 1% Triton X-100, 1% sodium deoxycholate, 0.1% sodium dodecyl sulfate (SDS), 1 mM phenylmethylsulfonyl fluoride (PMSF), 1 µg/mL leupeptin, and 10 µg/mL aprotinin. After being quantified via a bicinchoninic acid colorimetric assay kit (CWBIO, Beijing, China), the protein samples were separated using 10% SDS-polyacrylamide gel electrophoresis (PAGE), transferred onto polyvinylidene difluoride membranes (Millipore, Billerica, MA, U.S.A.), and subjected to sequential blotting with primary antibodies against the cannabinoid receptor type 1 (CB1R; Cat. # ab3559, Abcam, Cambridge, U.K.), brain-derived neurotrophic factor (BDNF; Clone # EPR1292, Abcam), and β-actin (Cat. # sc-130657, Santa Cruz, Dallas, TX, U.S.A.), and with a goat anti-rabbit secondary antibody (Kangcheng, Shanghai, China). The blotting bands were visualized by incubation with ECL reagent (Merck Millipore, Billerica, U.S.A.).
Statistical AnalysisAll data are presented as the mean±standard error of the mean (S.E.M.) and were analyzed using SPSS 10.0 software. Independent sample t-tests and least significant difference (LSD) tests were used for intra- or inter-group comparison, respectively. Comparisons of multiple groups were conducted using one-way ANOVA. The accepted level of statistical significance is p<0.05.
LE rats, which are reportedly sensitive to developing a cannabinoid reward response, were first used in the classic CPP paradigm.6,15) However, we observed an unexpected significant adverse effect of HU210 on rats; the animals displayed piloerection and increased repose time and defecation, especially after the first HU210 injection (data not shown). In addition, both control and HU210-treated (0.05 and 0.1 mg/kg) rats spent the same amount of time in the two end chambers during the test phase (Fig. 2A). The failure to develop place preference was not due to a decrease in the activity of the treated rats because rats showed similar travel distances (Fig. 2B) and numbers of entries into the end chambers during the pre-test and test periods (Fig. 2C). Thus, CPP elicited by high-dose cannabinoids cannot be observed by following the classic testing procedure.
A. Rats (n=8) received an i.p. injection of vehicle or HU210 (0.05 or 0.1 mg/kg body weight). The time spent in each chamber was recorded, and the preference scores were calculated as described in Materials and Methods. The p values were 0.216 (HU210/0.05) and 0.097 (HU210/0.1) compared with the vehicle group. B, C. Average distance traveled (B, p=0.370) and total number of entries (C, p=0.406) for rats receiving 0.1 mg/kg HU210 were recorded in the pre-test (before conditioning) and test (after conditioning) periods.
The duration of time that rats are forced to remain in the designated compartment during conditioning may be a factor that affects the coupling of the drug reward effect with the environment of the compartment. We thus prolonged the time rats stayed in the end chambers from 30 to 60 min in all conditioning sessions. However, rats failed to develop a preference for the end chamber in which they received an HU210 injection (Fig. 3A). Therefore, prolonged conditioning is insufficient for CPP development in response to high-dose cannabinoids in our test system.
A. The classic CPP procedure was followed except that the conditioning/training time of rats (n=8) was prolonged from 30 to 60 min. The preference scores were calculated and plotted (p=0.232). B. Rats received either a single i.p. injection (1P) or one i.p. injection daily for 4 d (4P) of 0.1 mg/kg HU210 before the conditioning sessions and the test phase, and the preference scores were calculated and plotted (p=0.089 for 1P and p=0.245 for 4P compared with the vehicle group, respectively; p=0.042 for 4P compared with 1P pre-injection group). * p<0.05. C. Rats were killed at the time indicated after the first pre-injection, and hippocampal protein samples were prepared and subjected to Western blotting analysis. Data shown are representative of three independent experiments. * p<0.05, ** p<0.01, as indicated or compared with the group on day 0.
Given the reported influence of cannabinoids on learning and memory, and the adverse effect of HU210 observed in our experiment,12) we next sought to minimize these effects by adding either a single or multiple days of pre-injections of HU210 before conditioning sessions. When a single HU210 pre-injection was administered before conditioning, no significant difference was detected between the time rats spent in the drug- and non-drug-paired chambers (Fig. 3B). However, despite a statistically insignificant difference between the pre-test and test, rats receiving once daily HU210 pre-injections for 4 d remained in the drug chamber longer than those receiving the single pre-injection (Fig. 3B).
Depending on the dosages and time courses of administration, cannabinoids reportedly either cause neuronal impairment or play a neuroprotective role by increasing the production of proinflammatory cytokines or neurotropic growth factors, respectively.16,17) Our previous study revealed that HU210 exerts an adverse effect on the hippocampal CA3-CA1 synapses.12) Furthermore, the administration of cannabinoid receptor agonists downregulates cannabinoid receptors, which changes neuronal activity and sensitivity in critical brain regions.18) Thus, we next examined the expression of CB1R and BDNF in the hippocampus, a brain region critically involved in drug reward.19) Although HU210 had no effect on CB1R expression, multiple HU210 pre-injections facilitated the production of BDNF, which is in contrast to the decrease observed in BDNF levels shortly after the first HU210 administration (Fig. 3C). These data suggest a role for multiple pre-injections in relieving the neurotoxicity of acute high-dose cannabinoid administration, likely via promoting the secretion of BDNF.
SD Rats Display Natural Preference for One of the CompartmentsAnimals often exhibit a natural preference for certain elements in an environment, and this preference should be considered and even exploited in the establishment of a CPP model.20) Given that LE rats rarely showed a natural place preference in our previous experiment, we used SD rats for further studies after first conducting a pilot experiment to assess their exploratory behavior in a novel environment. Similar to LE rats, we failed to observe CPP of these rats when following the classic test procedure or a protocol including 4 pre-injections prior to the conditioning sessions (Fig. 4A). Nevertheless, we found that 77.78% (14/18) of the SD rats tested spent significantly more time (over 150 s) in one rather than the other chamber in the pre-test stage, suggesting a natural place preference for these rats (Fig. 4B). Their chamber preference remained stable, as revealed by continuous independent tests in the subsequent 2 d (Fig. 4C). Therefore, SD rats tend to develop natural place preference, which will potentially affect the establishment of CPP for addictive drugs.
A. Rats received no injection (classic) or one i.p. injection daily for 4 d (4P) of 0.1 mg/kg HU210 before the conditioning sessions (vehicle or 0.1 mg/kg HU210) and the test phase. The preference scores were calculated and plotted (p=0.137 for classic and p=0.106 for 4P compared with the vehicle group, respectively). B. Sprague Dawley rats (n=18) were tested for their natural preference for 15 min, and the chambers in which a rat spent less and more time were designated non-preferred and preferred chambers, respectively. The next day, the time spent in each chamber was plotted. C. The percentages of time spent in preferred or non-preferred chamber to the total time spent in either of the end chambers were calculated during the 2 d pre-test. ** p<0.01 compared with the non-preferred group.
The SD rats showing natural chamber preference in the CPP apparatus were randomly grouped and used for further study. After receiving once daily pre-injections for 4 d, these rats were then subjected to conditioning via intentionally choosing the non-preferred chamber for HU210 administration. As a result, rats receiving HU210 in the non-preferred chamber exhibited significantly more time spent in this chamber compared with control rats injected with vehicle (Fig. 5A). The extended stay of rats in the drug-paired chamber was not due to a decrease in activity because the total distance traveled and number of entries into the end chambers were comparable in the pre-test and test phases (Figs. 5B, C). The results of the EPM test showed no significant difference in the anxiety level between HU210-treated and control rats on the test day, suggesting that the preference shift from the naturally preferred to non-preferred compartment was not attributable to a potential anxiolytic effect of HU210 (Figs. 5D, E). Consistent with the above results, multiple pre-injections allowed a better recovery of rats from an HU210-induced loss of body weight during the conditioning sessions (Fig. 5F). Taken together, biased selection of the naturally non-preferred chamber for drug administration, combined with multiple pre-injections to minimize the adverse effects of cannabinoids, is sufficient to support a significant place preference shift induced by high-dose cannabinoids.
A. Rats received daily pre-injections of HU210 (0.1 mg/kg, i.p.) for 4 d prior to classic CPP conditioning. The chamber naturally non-preferred by each rat was used as the chamber for HU210 administration (0.1 mg/kg) in the CPP test. The preference scores were calculated and plotted after the test phase (n=7; p=0.0069 for HU210 compared with the vehicle group). ** p<0.01. B, C. No significant difference in the distance traveled (B; vehicle, p=0.354; HU210, p=0.201) or in the total number of entries (C; vehicle, p=0.410; HU210, p=0.500) in either chamber was observed during pre-test and test periods in the HU210-injected group compared with the group of rats receiving vehicle injections (n=7). D, E. Rats described in A were subjected to the elevated plus-maze test immediately after the CPP test. The percentage of time spent (D; p=0.805) and number of entries (E; p=0.193) into the open arms are plotted. F. Body weights were recorded for rats subjected to four pre-injections and conditioning sessions with vehicle or HU210 during the experimental period, starting on the day of the first pre-injection.
Among psychotropic drugs, cannabinoids show a leading incidence of addiction.7,21) The rewarding effect of cannabinoids has been extensively studied using behavioral models, but the results of CPP tests with cannabinoids are conflicting.3,6,22) In particular, Δ9-THC elicited CPP, conditioned place aversion (CPA), or no place preference dependent on dose and rat strain used in the paradigms.6,8,22) Cheer et al. observed CPA following i.p. injection of HU210 at doses of 0.02, 0.06, and 0.1 mg/kg.3) By contrast, we detected a CPP shift elicited by high-dose HU210 (0.1 mg/kg) after optimizing the test protocol. Although a standardized CPP paradigm, especially of the conditioning sessions, is needed for cannabinoids, the discrepancies in these CPP results are likely attributable to differences in both the drug and the test subjects.11,20)
Behavioral studies rely on the responsive performance of the animals.10) In CPP models, the natural preference of animals, which is common in rodents, can be utilized for a biased design of the paradigm.23,24) In our study, LE rats were originally tested because of their relatively high sensitivity to cannabinoids, as documented by other researchers.6,15) However, LE rats failed to develop cannabinoid-related place preference despite the use of pre-injections. We observed frequent and stable natural compartment preference in SD but not LE rats. Thus, SD rats displaying a natural place preference were tested in our CPP protocol, with the non-preferred compartment of each rat used for cannabinoid administration. Consequently, these SD rats exhibited a significant shift of place preference, as revealed by the ratios of time spent in the drug-paired compartment pre- and post-drug conditioning.
Most abused drugs including cannabinoids activate the brain’s reward circuitry by enhancing dopamine release in reward relevant projection loci such as the medial forebrain bundle.25) This action of the synthetic cannabinoid, HU210, might be veiled by its widely reported neurotoxicity, e.g. impairment of spatial memory which is required for the development of CPP.25) Our recent study also confirmed a cannabinoid-induced impairment in learning and memory dependent on astroglial cannabinoid receptors by acting on the hippocampal CA3-CA1 synapses.12) Using a Morris water maze paradigm, Hill et al. found that consecutive i.p. administration of the same dose of HU210 resulted in significant impairment of rat memory; however, the performance of tested rats improved over 5 d and was indistinguishable from that of control animals despite HU210 injection continued.26) Consistently, we observed here that the abnormalities in general health or performance caused by HU210 were relieved in parallel with restored BDNF production in the hippocampus 3 to 4 d after first administration, leading to our hypothesis that HU210 pre-injections help overcome the adverse effects of HU210 that interfere with the development of CPP behavior. Therefore, we modified the classic CPP paradigm by adding pre-injection(s) preceding the conditioning procedures. Indeed, the daily administration of HU210 for 4 d prior to conditioning allowed rats to recover from loss of body weight and decreased activity, increasing the potential for a drug reward response. Consequently, multiple pre-injections of HU210 prior to the classic conditioning sessions facilitated the development of CPP in rats. Although it is unclear why HU210-elicited CPA was observed by several other groups, our data suggest that a rewarding behavior by HU210 conditioning can be detected given that the interfering adverse effects of HU210 are alleviated. Whereas here we primarily used a pre-injection dose identical to that used during the conditioning and test procedures, exploring whether lower dose pre-injections are sufficient to dilute the effects of cannabinoids that impair CPP development is warranted.11,27)
In addition to eliciting a reward response, cannabinoids reportedly exhibited other pharmacological activities such as an anxiolytic-like effect which may also affect the behavioral performance of animals.21) Thus, we used an EPM test to assess the anxiety-like behaviors of rats receiving programmed HU210 injections. As a result, we found that the preference shift of rats from the naturally preferred to non-preferred compartment did not originate from an anxiolytic action of the cannabinoid. Together, the results of our study demonstrated the feasibility of generating an effective CPP model for use with high-dose cannabinoids that minimizes the adverse actions of the drug and exploits the natural place preference of the test subjects.
This work was supported by the National Natural Sciences Foundation of China (No. 31371137) and the Research Project of Chinese Ministry of Education (113056A).
The authors declare no conflict of interest.
