Abstract. Introduction. ª Federation of European Neuroscience Societies. European Journal of Neuroscience, Vol. 20, pp.

Transcription

1 European Journal of Neuroscience, Vol. 20, pp , 2004 ª Federation of European Neuroscience Societies SHORT COMMUNICATION Prolonged cannabinoid treatment results in spatial working memory deficits and impaired long-term potentiation in the CA1 region of the hippocampus in vivo Matthew N. Hill,* David J. Froc,* Christopher J. Fox, Boris B. Gorzalka and Brian R. Christie Department of Psychology and the Brain Research Centre, 2136 West Mall, University of British Columbia, Vancouver, BC, Canada V6T 1Z4 Keywords: CA1, cannabinoid, hippocampus, long-term potentiation, spatial memory Abstract Adult male Long-Evans rats were administered the potent cannabinoid 1 receptor agonist HU-210 (100 lg kg, i.p.) for 15 days continuously and their performance on a matching-to-place version of the Morris water maze was subsequently evaluated. Overall, experimental animals performed significantly worse initially on the reference memory component of this task, but their performance improved over 5 days until it was indistinguishable from that of control animals. Animals given HU-210 did not exhibit working memory impairments at short intertrial delays (30 s); however, significant impairments were observed in learning performance with longer intertrial delays (300 s). In vivo electrophysiological analyses revealed that long-term potentiation in the CA1 region of the hippocampus was significantly impaired following the administration of HU-210 for 15 days. These results indicate that long-term cannabinoid exposure can produce marked deficits in reference and working memory performance, and also impair hippocampal synaptic plasticity in vivo. Introduction Cannabinoid intoxication has been shown to impede cognitive functioning and impair short-term memory in humans (Miller & Branconnier, 1983). These effects appear to be mediated by the activity of delta-9-tetrahydrocannabinol, the psychoactive constituent of cannabis (Gaoni & Mechoulam, 1964), at specific neural cannabinoid receptors (CB 1 and CB 2 ) located throughout the central nervous system (Ameri, 1999; Hampson & Deadwyler, 1999; Davies et al., 2002; Howlett, 2002). Given the effects of cannabinoid on mnemonic processes, it is not surprising that high densities of these receptors have been shown in the hippocampus, a structure known to be involved in human memory processes (Herkenham et al., 1991). Immunohistochemical staining has demonstrated that CB 1 receptors are found primarily on hippocampal GABAergic interneurons (Katona et al., 1999; Marsicano & Lutz, 1999; Tsou et al., 1999); however, the exogenous application of cannabinoids can also reduce glutamatergic signalling and excitatory activity (Shen et al., 1996; Misner & Sullivan, 1999). The CB 1 receptors are found primarily on axon terminals and are negatively coupled to adenylate cyclase and through activation of G i o protein subunits can regulate camp generation, intracellular calcium concentrations and ultimately neurotransmitter release (Schlicker & Kathmann, 2001; Ameri, 1999; Sullivan, 1999; Hoffman & Lupica, 2000; Howlett, 2002). Correspondence: Dr Brian R. Christie, as above. bchristie@psych.ubc.ca *M.N.H. and D.J.F. contributed equally to this study Received 13 April 2004, revised 27 May 2004, accepted 28 May 2004 Functionally, CB 1 receptors in the hippocampus are believed to mediate the cognitive deficits characteristic of cannabis intoxication (Hampson & Deadwyler, 1999; Davies et al., 2002). Administration of CB 1 receptor ligands has been shown to produce dose-dependent impairments in short-term memory as measured using the radial arm maze (Lichtman et al., 1995), the Morris water maze (Varvel et al., 2001) and delayed nonmatching-to-sample tasks (Hampson & Deadwyler, 1999). These findings suggest that CB 1 receptor activity has a crucial role in regulating synaptic plasticity in the hippocampus and that high levels of receptor activity, resulting from acute cannabinoid exposure, may compromise this process. In vitro studies have demonstrated that acute cannabinoid administration reduces field excitatory postsynaptic potentials (fepsps) (Ameri & Simmet, 2000), excitatory postsynaptic conductances (Shen et al., 1996; Misner & Sullivan, 1999; Sullivan, 1999) and high frequency stimulation-induced long-term potentiation (LTP), in the hippocampal CA1 region (Nowicky et al., 1987; Collins et al., 1994; Misner & Sullivan, 1999). The reduction in excitatory transmission, and LTP induction, is believed to underlie the detrimental effects of cannabinoids on memory processes (Sullivan, 2000); however, it is unclear whether CB 1 receptors exhibit tolerance, and whether these deficits will be observed when they are chronically activated over a period of weeks. In this study we characterize the effects of prolonged cannabinoid administration on working and reference memory using a novel Morris water maze paradigm, and for the first time, also examine the effects of prolonged cannabinoid administration on LTP in vivo. doi: /j x

2 860 M. N. Hill et al. Materials and methods Animals Male Long-Evans rats (Charles River, Montreal) weighing between 300 and 350 g were used in this study. Animals were housed in groups of three in triple mesh wire caging with ad libitum access to Purina rat chow and water. Colony temperature was maintained at 21 ± 1 C, and lighting was regulated on a reverse light dark cycle with lights off at h. All behavioural and electrophysiological testing occurred in the middle third of the dark cycle, 18 h after the last drug administration, and were performed in compliance with the guidelines of the Canadian Council on Animal Care. Drugs 3-(1,1-dimethylheptyl)-( )-11-hydroxy-D 8 -tetrahydrocannabinol (HU- 210), a potent and selective CB 1 receptor agonist, was obtained from Tocris Cookson (Bristol, UK) and dissolved in a vehicle of 8 : 1 : 1 of 0.9% saline : dimethyl sulfoxide : Tween 80. HU-210 was administered at a dose of 100 lg kg at a volume of 1 ml kg (i.p.). Behavioural apparatus The Morris water maze used in this experiment was a large circular fiberglass pool (193 cm diameter) that was filled to a depth of 70 cm with water (22 C). The water was made opaque by the addition of nontoxic white paint (Washable Dry Temp, Palmer Paint Products, Troy, MI, USA). The platform was a cylindrical jar ( cm) that had a square wire mesh platform on top, and was submerged 3 cm below the surface of the water. The room also had distinctive distal visual cues that surrounded the pool, and remained in place for the duration of the experiment. A computer-based automated tracking system (HVS Image, Hampton, UK) was used to calculate the swim path and escape latencies (the time each subject required to locate the hidden platform after being released) of each subject. Behavioural testing procedure Animals were administered daily injections of either HU-210 (100 lg kg) or the vehicle solution for 15 consecutive days, with the injections occurring in the middle third of the dark cycle. Behavioural testing was conducted daily prior to drug or vehicle injections, and all animals received the same version of a matching-to-place version of the Morris water maze (Harker & Whishaw, 2002). Briefly, animals were tested in the matching-to-place task for 5 consecutive days, with both the location of the platform and the release point in the pool changing on the first trial for each day. Both the release point and the platform location were the same for the second trial each day. For analysis, four points along the perimeter of the maze were arbitrarily designated as N, S, E and W, and these points also served as release points. Once the platform was located by a rat, it was allowed to remain there for 10 s before being removed from the pool and placed back into the holding cage for the remainder of the intertrial interval. If a rat was unable to locate the platform within 300 s, it was manually guided to it and allowed to remain on the platform for 10 s. The second trial was then initiated either 30 or 300 s after the animal was removed from the platform. The ability of animals to remember the platform s location on the second trial each day is believed to involve a strong working memory component (Harker & Whishaw, 2002), and the two intertrial intervals (30 or 300 s) were used to determine the effects of cannabinoid ligands on delayed recall (Hampson & Deadwyler, 1998, 1999; Varvel et al., 2001). Electrophysiology procedures After completing the entire course of behavioural testing, animals were individually anaesthetized with sodium pentobarbital (65 mg kg, i.p.) and placed into a stereotaxic apparatus (Kopf Instruments). Supplemental doses of sodium pentobarbital were injected as needed to maintain a surgical level of anaesthesia. Rectal temperature was maintained at 37 ± 1.0 C throughout the course of the experiment with a grounded homeothermic temperature control unit (Harvard Instruments, MA, USA). A 75-lm stainless-steel recording electrode (A-M Systems, Inc., WA, USA) was directed through a trephine hole into the CA1 stratum radiatum region (3.0 mm posterior and 2.5 mm lateral to bregma). Similarly, a 75-lm monopolar stimulating electrode (A-M Systems, Inc.) was directed through a trephine hole immediately adjacent to the recording electrode (3.2 mm posterior and 2.3 mm lateral to bregma). The final depth of the stimulation and recording electrodes was determined by adjusting both electrodes to yield the maximal fepsp using a minimal degree of stimulation ( la). To record pyramidal cell responses to afferent stimulation, single-pulse stimuli, of a fixed amplitude and duration (120 ls), were delivered at 15-s intervals and recorded using a differential amplifier (Getting Instruments, San Diego, CA, USA). Baseline responses were standardized at 30% of maximal response size, and recordings were required to exhibit a stable baseline for a minimum of 30 min before conditioning stimulation was applied. Two different conditioning protocols were used to induce LTP in these experiments. The weak theta-patterned stimulus (TPS) protocol consisted of five trains of theta-patterned conditioning stimuli. The stimuli were delivered as 10 bursts of five pulses at 100 Hz with a 200-ms interburst interval. The strong TPS protocol consisted of five pulses delivered as 400-Hz bursts. There were 10 bursts per train, again administered at a 200-ms interburst interval. The administration of the protocol was repeated five times at 2-min intervals. The higher frequency stimuli typically produce significantly greater LTP (Farmer et al., 2004). Signals were acquired and analysed as described previously (Farmer et al., 2004), with the initial slope of the negative-going fepsp being used to determine alterations in the level of synaptic efficacy. All EPSP slope data are presented as the mean percent change from the preconditioning baseline ± SEM. Results Reference and working memory impairments following long-term cannabinoid administration? Animals were randomly assigned to either the control or drug groups, and then initiated on a series of injections with either the vehicle solution (n ¼ 23) or solution containing the CB 1 receptor agonist HU- 210 (n ¼ 23). The data from the first trial performed by the animals each day gives an indication of their reference memory for the requisites of the task, i.e. knowing there is an escape platform that they can find. As is evident in Fig. 1, animals from both groups learned this task over a 5-day period. Animals that received HU-210 were significantly impaired on this task initially (F 1,44 ¼ 5.13, P ¼ 0.03), taking significantly longer to become proficient in the task on day 1 (P <0.01), but by day 5 were able to perform the task as well as control animals (P ¼ 0.14). For the second trial each day, all animals were also randomly divided into two subgroups, reflecting the latency between trial 1 and trial 2. One group of control (n ¼ 11) and experimental (n ¼ 11) animals was tested on the task with only a 30-s delay, while the other groups (n ¼ 12 each) were given a 300-s intertrial interval. The longer

3 Cannabinoid treatment, spatial memory and LTP 861 Fig. 1. Chronic HU-210 (100 lg kg) treatment impaired both reference and working memory in the Morris water maze. (A1) Animals administered HU-210 for 15 days showed marked impairment in learning that they needed to find a hidden platform, despite the fact that swim speed did not differ significantly between the groups (A2). (B1,B2) When a 30-s intertrial interval was used, both control and experimental animals performed identically, and both groups easily found the hidden platform on the second trial. (C1) When the intertrial interval was increased to 300 s, animals in the experimental group were significantly impaired in their ability to remember where the platform was located. (C2) Swim speeds remain identical between groups on this task. delay places a greater demand on working memory (Harker & Whishaw, 2002). As is shown in Fig. 1, animals in both groups were again able to learn the task over the 5-day testing period; however, animals that were administered cannabinoids did significantly worse with the longer intertrial delay period. Animals retested at 30 s were able to learn the task as quickly as control animals, and no differences in swim speed (F 4,20 ¼ 0.39, P > 0.05) or escape latency (F 4,20 ¼ 0.23, P > 0.05) were observed between the two groups (Fig. 1B). In contrast, when a 300-s intertrial interval was used, animals in the experimental group were severely impaired. Their initial escape latency was significantly longer than those of animals in the control group (F 4,22 ¼ 2.63, P < 0.05), despite the fact that swim speed was not significantly different between the two groups (F 4,22 ¼ 0.91, P > 0.05; Fig. 1C. To ensure that the difference between the two groups was not simply due to differences between the two testing procedures, we compared the two control groups in their trial 2 scores for both the 30-s and the 300-s delay. There was no significant difference between the control groups for the two intertrial intervals (F 4,20 ¼ 2.05, P > 0.05). Thus, the impairment at the long intertrial interval was specific to the HU-210-treated animals. Long-term HU-210 administration impaired hippocampal LTP Following behavioural testing, animals from all groups underwent electrophysiological experimentation. As shown in Fig. 2, stimulation of Schaeffer collateral inputs elicited negative-going fepsps with a mean amplitude of 0.92 ± 0.23 mv and a mean latency-to-peak of 8.41 ± 1.76 ms. These potentials could be reliably maintained for periods of >2 h (data not shown), and did not differ significantly between the two experimental groups. Paired-pulse analyses (50-ms interpulse interval) did not reveal significant differences in pairedpulse facilitation between animals in the control (31.1 ± 6.8%; n ¼ 7) and cannabinoid (31.0 ± 16.5%; n ¼ 6) groups, indicating that transmitter release was not affected in these studies. Our initial experiments examined the effects of HU-210 on the induction of LTP using weak TPS. When these conditioning stimuli

4 862 M. N. Hill et al. Fig. 2. Chronic HU-210 (100 lg kg) treatment impaired the induction of LTP in the CA1 region of the hippocampus in vivo. (A) Administration of five trains of 100-Hz stimuli as theta-patterned bursts produced significant LTP in both control and experimental animals; however, experimental animals administered HU-210 show significantly less LTP overall. (B) Attempts to induce more LTP with stronger stimuli (five trains of 400-Hz stimuli delivered as theta-patterned bursts) only produced significant LTP in control animals. Experimental animals failed to show a significant degree of LTP in this instance. Insets in A and B show examples of evoked waveforms recorded prior to (dashed) and following (solid) the application of conditioning stimulation. Scale bars, 0.5 mv, 10 ms. were administered, there was a significant group time interaction (F 74,666 ¼ 1.98, P < 0.001) for the postconditioning recordings. While both the control (29.6 ± 8.9%, t 1,5 ¼ 3.32, P < 0.05) and cannabinoid (8.4 ± 2.2%, t 4 ¼ 3.75, P < 0.05) groups showed significant LTP, as measured at 60-min postconditioning, the cannabinoid-treated group showed significantly less LTP (t 4 ¼ 3.23, P < 0.05) than control animals. To determine whether we could induce a larger degree of LTP in the animals that had been administered HU-210, in separate animals we applied stronger conditioning stimuli (strong TPS), which normally induces larger, more robust, LTP. In these animals, there was again a significant group time interaction F 74,740 ¼ 4.83, P < 0.001) following the application of the strong TPS (Fig. 2). However, when LTP was assessed at 60 min postconditioning, only the control group showed a significant degree of LTP (47.51 ± 13.01%, t 5 ¼ 3.65, P < 0.05). Surprisingly, the strong TPS did not result in a significant facilitation of the evoked fepsp in the cannabinoid group (9.78 ± 4.79%, t 5 ¼ 2.04, P > 0.05), and overall significantly less LTP was induced in experimental animals when compared to controls (t 5 ¼ 3.96, P < 0.05). Discussion The data obtained here demonstrate that animals that are exposed to high levels of the CB 1 receptor agonist HU-210 over a 15-day period display significant deficits in both reference and working memory. The deficits in working memory were evident when longer intertrial delays (300 s) were used on a matching-to-place version of the Morris water maze (Harker & Whishaw, 2002). It is important to note that, while these deficits were apparent initially, all animals learned the task and there was no difference between the performance of control and experimental animals by day 5. Interestingly, it has been suggested that cannabinoids selectively disrupt the process of mnemonic encoding during memory tasks (Hampson & Deadwyler, 2000) and that the ability of cannabinoids to impair recall is delay-dependent (Hampson & Deadwyler, 1998) and more apparent in tasks that engage working memory processes (Varvel et al., 2001). Our data corroborate these findings, as experimental and control animals performed equivalently at short delays but experimental animals exhibited a significant impairment when the intertrial delay was increased. This delay-dependent effect offers strong support for the concept that cannabinoids impair the consolidation of memories, but not mnemonic processes in general. This implies that long-term cannabinoid exposure leads to deficits that compromise hippocampal functioning and this may account for the cognitive deficits seen in habitual heavy cannabis users (Solowij et al., 2002). Our results are also the first to show that long-term administration of the cannabinoid agonist HU-210 inhibits LTP induction in the CA1 region of the hippocampus in vivo. These findings are in agreement with previous in vitro work that demonstrates that acute administration of cannabinoid ligands can reduce LTP induction (Nowicky et al., 1987; Collins et al., 1994; Misner & Sullivan, 1999). This study adds to the existing body of research in two important ways: first, it demonstrates that deficits in LTP induction do not develop tolerance after prolonged treatment and, second, that these deficits can be observed up to 18 h following the final drug administration, indicating that these alterations are probably due to a neuroadaptive change from chronic drug exposure and not due to the effects of drug exposure itself. It has previously been shown that treatment with another CB 1 receptor ligand (WIN 55,212-2) for 21 days resulted in shorter and disjointed dendrites in the CA1 region (Lawston et al., 2000). This dendritic atrophy could induce the observed decrements in LTP; however, the possibility also exists that prolonged cannabinoid treatment may alter postsynaptic glutamate receptor subunit expression (Sullivan, 2000). Further research is necessary to fully characterize the neurochemical and morphological changes that occur in the hippocampus following long-term cannabinoid administration. Acknowledgements The authors would like to thank Stephanie Lieblich and Indy Gill for technical support. This research was supported by NSERC, CIHR, and HELP awards to B.R.C., and an NSERC award to B.B.G. M.N.H. was supported by MSFHR and NSERC postgraduate scholarships. B.R.C. is a BMO Young Investigator. Abbreviations CB 1, cannabinoid 1 receptors; CB 2, cannabinoid 2 receptors; fepsps, field excitatory postsynaptic potentials; HU-210, 3-(1,1-dimethylheptyl)-( )-11- hydroxy-d8-tetrahydrocannabinol; LTP, long-term potentiation; TPS, thetapatterned stimulus.

5 Cannabinoid treatment, spatial memory and LTP 863 References Ameri, A. (1999) The effects of cannabinoids on the brain. Prog. Neurobiol., 58, Ameri, A. & Simmet, T. (2000) Effects of 2-Arachidonylglycerol, an endogenous cannabinoid, on neuronal activity in rat hippocampal slices. Naunyn Scmiedeberg s Arch. Pharmacol., 361, Collins, D.R., Pertwee, R.G. & Davies, S.N. (1994) The actions of synthetic cannabinoids on the induction of long term potentiation in the rat hippocampal slice. Eur. J. Pharmacol., 259, R7 R8. Davies, S.N., Pertwee, R.G. & Riedel, G. (2002) Functions of cannabinoid receptors in the hippocampus. Neuropharmacology, 42, Farmer, J., Zhao, X., van Praag, H., Wodtke, K., Gage, F.H. & Christie, B.R. (2004) Effects of voluntary exercise on synaptic plasticity and gene expression in the dentate gyrus of adult male sprague-dawley rats in vivo. Neuroscience, 124, Gaoni, Y. & Mechoulam, R. (1964) Isolation, structure, and partial synthesis of an active constituent of hashish. J. Am. Chem. Soc., 86, Hampson, R.E. & Deadwyler, S.A. (1998) Role of cannabinoid receptors in memory storage. Neurobiol. Dis., 5, Hampson, R.E. & Deadwyler, S.A. (1999) Cannabinoids, hippocampal function and memory. Life Sci., 65, Hampson, R.E. & Deadwyler, S.A. (2000) Cannabinoids reveal the necessity of hippocampal neural encoding for short-term memory in rats. J. Neurosci., 20, Harker, K.T. & Whishaw, I.Q. (2002) Place and matching-to-place spatial learning affected by rat inbreeding (Dark-Agouti, Fischer 344) and albinism (Wistar, Sprague-Dawley) but not domestication (wild rat vs. Long-Evans, Fischer-Norway). Behav. Brain Res., 134, Herkenham, M., Lynn, A.B., Johnson, M.R., Melvin, L.S., de Costa, B.R. & Rice, K.C. (1991) Characterization and localization of cannabinoid receptors in the rat brain: a quantitative in vitro autoradiographic study. J. Neurosci., 11, Hoffman, A.F. & Lupica, C.R. (2000) Mechanisms of cannabinoid inhibition of GABA (A) synaptic transmission in the hippocampus. J. Neurosci., 20, Howlett, A.C. (2002) The cannabinoid receptors. Prostaglandins Other Lipid Med., 68 69, Katona, I., Sperlagh, B., Sik, A., Kafalvi, A., Vizi, E.S., Mackie, K. & Freund, T.F. (1999) Presynaptically located CB1 cannabinoid receptors regulate GABA release from axon terminals of specific hippocampal interneurons. J. Neurosci., 19, Lawston, J., Borella, A., Robinson, J.K. & Whitaker-Azmitia, P.M. (2000) Changes in hippocampal morphology following chronic treatment with the synthetic cannabinoid WIN 55, Brain Res., 877, Lichtman, A.H., Dimen, K.R. & Martin, B.R. (1995) Systemic or intrahippocampal cannabinoid administration impairs spatial memory in rats. Psychopharmacology, 119, Marsicano, G. & Lutz, B. (1999) Expression of the cannabinoid receptor CB1 in distinct neuronal subpopulations in the adult mouse forebrain. Eur. J. Neurosci., 11, Miller, L.L. & Branconnier, R.J. (1983) Cannabis: Effects on memory and the limbic cholinergic system. Psychol. Bull., 93, Misner, D.L. & Sullivan, J.M. (1999) Mechanism of cannabinoid effects on long term potentiation and depression in hippocampal CA1 neurons. J. Neurosci., 19, Nowicky, A.V., Teyler, T.J. & Vardaris, R.M. (1987) The modulation of long term potentiation by delta-9-tetrahydrocannabinol in the rat hippocampus, in vitro. Brain Res. Bull., 19, Schlicker, E. & Kathmann, M. (2001) Modulation of transmitter release via presynaptic cannabinoid receptors. Trends Pharmacol. Sci., 22, Shen, M., Piser, T.M., Seybold, V.S. & Thayer, S.A. (1996) Cannabinoid receptor agonists inhibit glutamatergic synaptic transmission in rat hippocampal cultures. J. Neurosci., 16, Solowij, N., Stephens, R.S., Roffman, R.A., Babor, T., Kadden, R., Miller, M., Christiansen, K., McRee, B. & Vendetti, J. (2002) Cognitive functioning of long-term heavy cannabis users seeking treatment. JAMA, 287, Sullivan, J.M. (1999) Mechanisms of cannabinoid-receptor mediated inhibition of synaptic transmission in cultured hippocampal pyramidal neurons. J. Neurophysiol., 82, Sullivan, J.M. (2000) Cellular and molecular mechanisms underlying learning and memory impairments produced by cannabinoids. Learn. Mem., 7, Tsou, K., Mackie, K., Sanudo-Pena, M.C. & Walker, J.M. (1999) Cannabinoid CB1 receptors are localized primarily on cholecystokinin-containing GABAergic interneurons in the rat hippocampal formation. Neuroscience, 93, Varvel, S.A., Hamm, R.J., Martin, B.R. & Lichtman, A.H. (2001) Differential effects of delta 9-THC on spatial reference and working memory in mice. Psychopharmacology, 157,