Cannabinoids enhance susceptibility of immature brain to ethanol neurotoxicity.код для вставкиСкачать
Cannabinoids Enhance Susceptibility of Immature Brain to Ethanol Neurotoxicity Henrik H. Hansen, PhD,1,2 Birte Krutz, MD,1,3 Marco Sifringer, PhD,3 Vanya Stefovska, MD, PhD,3 Petra Bittigau, MD,1 Fritz Pragst, MD,4 Giovanni Marsicano, PhD,5–7 Beat Lutz, PhD,5,6 and Chrysanthy Ikonomidou, MD, PhD3 Objective: Marijuana and alcohol are most widely abused drugs among women of reproductive age. Neurocognitive deficits have been reported in children whose mothers used marijuana during pregnancy. Maternal consumption of ethanol is known to cause serious developmental deficits. Methods: Infant rats and mice received systemic injections of ⌬9-tetrahydrocannabinol (THC; 1–10mg/kg) or the synthetic cannabinoid WIN 55,212-2 (1–10mg/kg), alone or in combination with subtoxic and toxic ethanol doses, and apoptotic neurodegeneration was studied in the brains. Results: Acute administration of THC (1–10mg/kg), the principal psychoactive cannabinoid of marijuana, markedly enhanced proapoptotic properties of ethanol in the neonatal rat brain. THC did not induce neurodegeneration when administered alone. Neuronal degeneration became disseminated and severe when THC was combined with a mildly intoxicating ethanol dose (3gm/kg), with the effect of this drug combination resembling the massive apoptotic death observed when administering ethanol alone at much higher doses. The detrimental effect of THC was mimicked by the synthetic cannabinoid WIN 55,212-2 (1–10mg/kg) and counteracted by the CB1 receptor antagonist SR141716A (0.4mg/kg). THC enhanced the proapoptotic effect of the GABAA agonist phenobarbital and the N-methyl-D-aspartate receptor antagonist dizocilpine. Interestingly, infant CB1 receptor knock-out mice were less susceptible to the neurotoxic effect of ethanol. Furthermore, the CB1 receptor antagonist SR141716A ameliorated neurotoxicity of ethanol. Interpretation: These observations indicate that CB1 receptor activation modulates GABAergic and glutamatergic neurotransmission and primes the developing brain to suffer apoptotic neuronal death. Ann Neurol 2008;64:42–52 The psychoactive ingredients of the cannabis (marijuana) plant, Cannabis sativa, produce a broad spectrum of effects in the mammalian brain. Best known are influences on psychomotor function, memory, cognition, and pain perception. These effects have been studied intensively in adult populations and in mature animals. They are induced via activation of cerebral cannabinoid CB1 receptors,1–3 which are located in the forebrain and cerebellum.2– 4 The CB2 receptor subtype is abundantly expressed in peripheral organs and is largely found in immune tissues.5 The impact of developmental cannabinoid exposure is gaining increasing attention because there is growing social concern that marijuana abuse during pregnancy, alone or in combination with other drugs (ethanol), may have serious effects on fetal brain development. The concern is based on the epidemic of marijuana abuse in adolescents and young adults.6 It is estimated that 7.5 to 15% of all pregnant women contacting public and private prenatal care facilities use illicit substances during pregnancy, and that marijuana is among the most frequently used illicit drugs in women of childbearing years.7 There is strong evidence for transplacental delivery of the pharmacologically active principal psychoactive component of marijuana, ⌬9-tetrahydrocannabinol From the 1Department of Pediatric Neurology and Neuroscience Research Center, Humboldt University, Berlin, Germany; 2Department of Functional Neuroanatomy and Biomarkers, Neurosearch A/S, Ballerup, Denmark; 3Department of Pediatric Neurology, Children’s University Hospital Carl Gustav Carus, University of Technology Dresden, Dresden; 4Institute of Legal Medicine, Humboldt University, Berlin; 5Molecular Genetics of Behaviour, Max Planck Institute of Psychiatry, Munich; 6Department of Physiological Chemistry, Johannes Gutenberg-University Mainz, Mainz, Germany; and 7Equipe AVENIR 8, Centre de Recherche Institut National de la Sante et de la Recherche Médicale U862, Bordeaux, France. Received May 23, 2007, and in revised form Sep 19. Accepted for publication Sep 21, 2007. Additional Supporting Information may be found in the online version of the article. 42 H.H.H. and B.K. contributed equally to this work. Published online Dec 7, 2007, in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/ana.21287 Address correspondence to Dr Ikonomidou, Department of Pediatric Neurology, Children’s Hospital, University of Technology Dresden, Fetscherstrasse 74, 01307 Dresden, Germany. E-mail: email@example.com © 2008 American Neurological Association Published by Wiley-Liss, Inc., through Wiley Subscription Services (THC), to the fetus.8 It has been described that maternal marijuana abuse results in intrauterine growth retardation.9 Also, infants exposed to marijuana have been reported frequently to exhibit a transient neonatal syndrome of lethargy, hypotonia, tremors, and blunted response to visual stimuli. Intrauterine cannabinoid exposure can result in mental disturbances in preschool and school-age children, for example, reduced verbal and memory skills, attention deficits, and learning disabilities.10,11 Experimental studies link behavioral and cognitive deficits and emotional responsiveness to developmental exposure to cannabinoids.12 Physiological cell death, a process by which unsuccessful neurons are deleted by apoptosis (cell suicide) from the developing central nervous system (CNS), has been recognized as a physiological phenomenon in the developing brain. In recent studies, we have shown that ethanol and compounds that are used as sedatives, anesthetics, or anticonvulsants13–16 trigger widespread apoptotic neurodegeneration throughout the developing brain when administered to immature rodents during the period of rapid brain growth. Such compounds include drugs that alter physiological synaptic activity, that is, antagonists of N-methyl-D-aspartate (NMDA) receptors (ketamine, nitrous oxide), agonists of GABAA receptors (barbiturates, benzodiazepines, propofol), and/or sodium channel blockers (phenytoin, valproate). Initial studies in rats demonstrated neurotoxic effects of ethanol and anesthetic and anticonvulsant drugs at relatively high doses, leading to questioning of the clinical relevance of these finding. However, in subsequent studies, it could be shown that with the use of sensitive techniques, much lower neuroapoptosis thresholds than originally reported could be detected for ethanol in the mouse brain.17 In addition, experimental data show marked interspecies variability regarding sensitivity to neurotoxicity of ethanol; that is, mice are much more sensitive than rats.17 When taking this evidence into consideration, it remains unclear how sensitive the human brain may be to such effects. We postulated that cannabinoids may exert neurotoxic effects in the developing mammalian brain similar to those of ethanol and sedative and anticonvulsant drugs, and designed this study to explore such effects. The results show that cannabinoid receptor agonists do not have proapoptotic properties at clinically relevant doses but can markedly potentiate neurotoxicity of ethanol and compounds that either enhance GABAergic neurotransmission or inhibit glutamate-mediated excitation. Materials and Methods Animal Experiments All animal experiments were performed in accordance with the guidelines of the Humboldt University in Berlin, Germany. Drug Injections Wistar rat pups (Bundesinstitut für gesundheitlichen Verbraucherschutz und Veterinärmedizin BgVV, Berlin, Germany), 1 to 14 days old, weighing 5 to 25gm, received subcutaneous (SC) injections of ethanol (0.5–1.8gm/kg, total dose 1–3.6gm/kg), THC (1–10mg/kg), WIN 55,212-2 (1– 10mg/kg), and SR141716A (0.4mg/kg). The NMDA receptor antagonist dizocilpine (MK801; 0.5mg/kg at 0 hours) and the GABAA agonist phenobarbital (40mg/kg at 0 hours) were administered intraperitoneally. Experimental groups consisted of at least 6 to 13 rats. Ethanol was prepared as a 20% solution in sterile normal saline, and injection volume was calculated according to dose. All other drugs and vehicle were administered in a volume of 10ml/kg. In each set of experiments, animals from the same litter were distributed equally among treatment groups. CB1 null mutant mice were generated and genotyped as described elsewhere.18 Postnatal day 7 (P7) homozygous CB1-deficient mice (CB1⫺/⫺), heterozygous (CB1⫹/⫺), and wild-type littermates (CB1⫹/⫹) from heterozygous breedings were used for the experiments. Mutant mice and their control mice were backcrossed six times into C57BL/6N after the generation of the mutant allele. Thus, they are in a predominant C57BL/6N background. Histology For histological analysis of the brains, animals received an overdose of intraperitoneal chloral hydrate and were transcardially perfused with heparinized 0.1M phosphate-buffered saline, pH 7.4, followed by 4% paraformaldehyde in cacodylate buffer, pH 7.4. Brains were postfixed for 5 days at 4°C and processed for either DeOlmos cupric silver19 or TUNEL (terminal deoxynucleotide transferase-mediated dUTP nick end-labeling) staining. For DeOlmos cupric silver staining, brains were embedded in agar. Coronal sections of 70m thickness were cut serially on a vibratome (Leica VT 1000 S; Leica, Nu␤loch, Germany) and processed for staining with silver nitrate and cupric nitrate according to DeOlmos and Ingram’s method.19 Degenerating cells were identified by their distinct dark appearance caused by the silver impregnation. TUNEL staining was performed on 12m thick paraffin sections using the ApopTag Peroxidase kit (S 7100; Oncor Appligene, Heidelberg, Germany) according to the manufacturer’s instructions. In brief, after pretreatment with proteinase K and quenching of endogenous peroxidase, sections were incubated first in equilibration buffer followed by working strength TdT enzyme (incorporating digoxigeninlabeled dUTP nucleotides to free 3⬘-OH DNA termini) for 1 hour at 37°C. Sections were incubated in stop/wash buffer (30 minutes; 37°C), then with antidigoxigenin-peroxidase conjugate (30 minutes) followed by diaminobenzidine tetrahydrochloride substrate (Sigma, Deisenhofen, Germany) and lightly counterstained with methylgreen. DeOlmos staining was used to study most of the rat brains in this study. TUNEL staining was used only in the rat experiments illustrated in Figure 5 and to study the infant mice brains that were too fragile to be cut on a vibratome and be processed for DeOlmos staining. Hansen et al: Neurotoxicity of Cannabinoids 43 Quantification of Cell Death in Different Brain Regions Degenerating cells were determined in sections stained by the DeOlmos cupric silver method or TUNEL method in the frontal cortex layer II, parietal cortex layer II, cingulate cortex layer II, cingulate cortex layer IV, retrosplenial cortex layer II, retrosplenial cortex layer IV, mediodorsal thalamus, laterodorsal thalamus, subiculum, and CA1 hippocampus by means of stereological disector,20 estimating mean numeric cell densities (Nv) of degenerating cells (cells/mm3). In silverstained sections, only structures that could be identified as cell bodies according to their size (⬎8m; not smaller cell fragments and axons) were counted. An unbiased counting frame (0.05 ⫻ 0.05mm; dissector height, 0.07 or 0.01mm) and a high aperture objective were used for the sampling. The Nv for each brain region was determined with 8 to 10 dissectors. To assess overall severity of damage and enable comparisons among treatment groups, we determined numeric densities of degenerated cells in 10 regions, and these values were added to give a score for cell density of degenerating cells within each brain. The cell density scores were used to compare severity of brain damage between groups. In previous studies, we have compared quantitative results obtained using the DeOlmos silver staining and TUNEL staining techniques in the same brains by processing one hemisphere for silver staining and one for TUNEL staining. This was done to assure that these techniques give comparable results. Western Blot For Western blotting analysis, snap frozen tissue was homogenized (900g, 4°C) in lysis buffer (pH 7.6, 50mM tris[hydroxymethyl]aminomethane [Tris], 166mM KCl, 5mM ethylene diamine tetraacetic acid, 1% Triton X-100 [Sigma, St. Louis, MO]) containing a mixture of protease and phosphatase inhibitors (1mM phenylmethane sulfonyl fluoride, 0.5g/ml leupeptin, 1g/ml pepstatin, 2g/ml aprotinin, 1mM sodium orthovanadate). The homogenate was centrifuged at 1,050g (4°C) for 10 minutes, the microsomal fraction was subsequently centrifuged at 17,000g (4°C) for 20 minutes, and the resulting supernatant was collected. Protein concentrations were determined using the bicinchoninic acid kit according to the manufacturer’s instructions (Interchim, Montluçon, France). Protein extracts (20g/sample) and a biotinylated molecular weight marker (Cell Signaling Technology, Beverly, MA) were denaturated in Laemmli sample loading buffer (pH 6.8, 62.5mM Tris, 50% glycerol, 2% sodium dodecyl sulfate, 5% ␤-mercaptoethanol, 0.1mg/ml bromophenol blue) at 95°C, separated by 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis, and electrotransferred in transfer buffer (80mM Tris, 39mM glycine, 20% methanol) to a nitrocellulose membrane (0.2m pore; Protran, Schleicher & Schuell, Dassel, Germany). The membrane was rinsed with Tween 20 –containing Tris-buffered saline (10mM Tris, 150mM NaCl, 0.1% Tween 20, pH 7.4) and treated with blocking solution (5% nonfat dry milk in Tris-buffered saline) for 2 hours at room temperature to prevent nonspecific antibody binding. Equal loading and transfer of proteins was confirmed by staining the membranes with Ponceau S solution (Fluka, Buchs, Switzerland). 44 Annals of Neurology Vol 64 No 1 July 2008 The membrane was incubated overnight at 4°C with rabbit anti–CB1 receptor polyclonal antibody (1:500; Cayman Chemical, Ann Arbor, MI) or anti–CB2 receptor antibody (1:250; Santa Cruz, Santa Cruz, CA). After incubation with horseradish peroxidase–labeled secondary antibody (anti–rabbit 1:1,000; Amersham Pharmacia Biosciences, Buckinghamshire, United Kingdom) at room temperature, the immunoreactive protein was detected by the enhanced chemiluminescence system (ECL; Amersham Pharmacia Biosciences) and serial exposures were made on autoradiographic film (Hyperfilm ECL; Amersham Pharmacia Biosciences). Densitometric analysis of the blots was performed with the image analysis program TINA 2.09g (Raytest Isotopenmessgeräte, Straubenhardt, Germany). For stripping of the CB1 or CB2 receptor antibody, the same membrane was incubated for 45 minutes at 50°C in stripping buffer (pH 6.8, 62.5mM Tris, 2% sodium dodecyl sulfate, 0.7% ␤-mercaptoethanol). After washing with Trisbuffered saline and blocking, the membrane was reprobed overnight at 4°C with mouse anti–␤-actin monoclonal antibody (1:10,000; Sigma). Statistical Analyses Values are presented as mean ⫾ standard error of the mean. Comparisons among groups were made using one-way analysis of variance with Bonferroni’s multiple-comparisons test. Results Cannabinoids Do Not Induce Neurodegeneration but Do Potentiate the Neurotoxic Effect of Ethanol in the Neonatal Rat Brain P7 rats received two consecutive doses (at 0 and 2 hours) of THC (1–20mg/kg SC), WIN 55,212-2 (WIN, 5–10mg/kg SC), or vehicle (VEH, 8% Chremophor in phosphate-buffered saline, pH 7.4). Numeric densities of degenerating neurons (per cubic millimeter) of 10 brain regions of vehicle- and drugtreated P8 rats were assessed 24 hours after the first dose. No significant neurodegenerative response was noted in the brains of P7 pups treated with the highest dose of THC (20mg/kg) or WIN 55,212-2 compared with the vehicle-treated group in the frontal, parietal, cingulate, retrosplenial cortex, caudate nucleus (mediodorsal part), thalamus (laterodorsal, mediodorsal, ventral nuclei), and subiculum (Fig 1A, Supplemental Figure). Ethanol is known to cause apoptotic neurodegeneration in the developing rat brain. P7 rats received two consecutive (at 0 and 2 hours) doses of ethanol (EtOH; 1.5gm/kg SC, total dose 3gm/kg) alone and in combination with THC (1–10mg/kg SC). Ethanol alone produced a small apoptotic response in the neonatal rat brain, which was markedly potentiated in a dose-dependent manner by THC (1–10mg/kg) and WIN 55,212-2 (Table; see Figs 1B, C, Supplemental Figure). Measurement of ethanol blood levels 3 and 5 hours Fig 1. Cannabinoids do not induce neurodegeneration in the neonatal rat brain but enhance neurotoxicity of ethanol (EtOH). (A) Postnatal day 7 (P7) rats received two consecutive doses of ⌬9-tetrahydrocannabinol (THC; 1–20mg/kg subcutaneously [SC]), WIN 55,212-2 (WIN; 5–10mg/kg SC), or vehicle (VEH). Numeric densities of degenerating neurons (per mm3) were assessed 24 hours after the first dose (B) Coadministration of THC and incrementing doses of EtOH demonstrates that THC can promote neurodegeneration when combined with EtOH. P7 rats received two consecutive doses of EtOH (total dose of 3gm/kg) and THC (total dose of 10 or 20mg/kg). THC significantly increased neurotoxic effect of EtOH. ***p ⬍ 0.001 in comparison with VEH; ⫹⫹⫹ p ⬍ 0.001 in comparison with EtOH group. (C) The synthetic cannabinoid WIN 55,212-2 has neurotoxic properties when coadministered with EtOH. P7 rats received two consecutive doses of EtOH (total doses of 2.2 or 3gm/kg SC) and WIN 55,212-2 (WIN, 5mg/kg SC). Neurodegeneration was assessed 24 hrs after the first dose. **p ⬍ 0.01; ***p ⬍ 0.001 in comparison with VEH; ⫹⫹⫹p ⬍ 0.001 in comparison with the corresponding EtOH dose group. (D) THC has no significant effect on plasma EtOH levels in neonatal rats. Rat pups received two consecutive doses of THC (total dose 10 mg/kg SC) in combination with EtOH (total dose 2.2gm/kg) or VEH. Cardiac plasma EtOH levels were assessed. No statistical differences between groups were found at 3 or 5 hours. (E) Neurodegenerative effect of THC and EtOH coexposure is reversed by the cannabinoid receptor antagonist SR141716A. P7 rats received a single dose of SR141617A (0.4mg/kg SC) before two consecutive doses of EtOH (total dose 3gm/kg) and THC (total dose 10mg/kg). The neurotoxic effect of EtOH ⫹ THC was reversed by SR141716A. ***p ⬍ 0.001 compared with VEH; ⫹⫹⫹p ⬍ 0.001 compared with EtOH and ###p ⬍ 0.001 as indicated. The x-axis shows the total dose of compounds administered. Data are expressed as means ⫾ SEM and have been analyzed by means of one way ANOVA with Bonferroni’s multiple comparison test. after the first drug injection showed that THC had no influence on cardiac plasma ethanol levels (see Fig 1D). The neurodegenerative effect of THC and ethanol coexposure could be reversed by the cannabinoid receptor antagonist SR141716A (see Fig 1E). SR141716A was administered at the dose of 0.4mg/kg SC before two consecutive (at 0 and 2 hours) doses of ethanol (1.5gm/kg SC) and THC (5mg/kg SC). Neurotoxic Effect of Ethanol and ⌬9Tetrahydrocannabinol Coexposure Is Age Dependent P2 to P14 rat pups received two consecutive (at 0 and 2 hours) doses of ethanol (1.1gm/kg SC, total dose of 2.2gm/kg) in combination with THC (5mg/kg SC). Numeric densities of degenerating neurons (per cubic millimeter) in 10 brain regions of vehicle (8% Chremophor in phosphate-buffered saline, pH 7.4) and Hansen et al: Neurotoxicity of Cannabinoids 45 Table. ⌬9-Tetrahydrocannabinol Treatment Enhances Neurotoxic Effect of Ethanol in the Developing Forebrain of Neonatal Rats Vehicle Ethanol 3gm/kg Ethanol 3gm/kg ⴙ THC 2mg/kg Ethanol 3gm/kg ⴙ THC 10mg/kg Ethanol 3gm/kg ⴙ THC 20mg/kg Frontal cortex II 771 ⫾ 138 228 ⫾ 73 285 ⫾ 83 253 ⫾ 83 285 ⫾ 143 Parietal cortex II 1,296 ⫾ 499 457 ⫾ 105 476 ⫾ 208 631 ⫾ 203 500 ⫾ 358 Cingulate cortex II 1,071 ⫾ 408 857 ⫾ 275 1,571 ⫾ 297 2,044 ⫾ 454 1,571 ⫾ 143 Cingulate cortex IV 457 ⫾ 67 628 ⫾ 22 5,079 ⫾ 801 10,428 ⫾ 3,143a Retrosplenial cortex II 400 ⫾ 97 799 ⫾ 233 571 ⫾ 143 568 ⫾ 127 1,857 ⫾ 1,000 Retrosplenial cortex IV 399 ⫾ 76 1,199 ⫾ 266 4,380 ⫾ 537 5,968 ⫾ 1,537 10,143 ⫾ 4,572b Mediolateral thalamus 143 ⫾ 45 840 ⫾ 204 3,194 ⫾ 465 4,301 ⫾ 1,136 9,414 ⫾ 1,872a Laterodorsal thalamus 68 ⫾ 14 805 ⫾ 221 5,295 ⫾ 862a 4,672 ⫾ 1,204a 16,778 ⫾ 64a 4,371 ⫾ 992 9,790 ⫾ 714a 24,396 ⫾ 2,881a 37,857 ⫾ 715a Brain Region 1,142 ⫾ 226 Subiculum 668 ⫾ 167 CA1 285 ⫾ 80 971 ⫾ 291 7,381 ⫾ 938 8,603 ⫾ 2,285 15,286 ⫾ 1,429a 5,558 ⫾ 774 11,670 ⫾ 2,280 34,291 ⫾ 3,420a 56,515 ⫾ 8,592a 104,117 ⫾ 10,721a Total score c a Postnatal day 7 (P7) rats received vehicle, ethanol (3gm/kg given in two separate doses of 1.5gm/kg subcutaneously 2 hours apart) with or without ⌬9-tetrahydrocannabinol (THC; 2-20mg/kg, given in two separate doses of 1, 5, or 10mg/kg at 0 and 2 hours). Using a stereological dissector method, we assessed the numeric densities of degenerating neurons (per mm3) in transverse 70m sections of 10 brain regions of vehicle (saline) and drug-treated P8 rats (24 hours after the first dose). Data are expressed as mean numeric densities of degenerating neurons ⫾ standard error of the mean (n ⫽ 6-13 per group). a p ⬍ 0.001, bp ⬍ 0.05, cp ⬍ 0.01, one-way analysis of variance, Bonferroni’s multiple-comparisons test; refers to comparisons between the ethanol and ethanol ⫹ THC groups. drug-treated P8 rats (24 hours after the first dose) were assessed. Quantification of neurodegeneration demonstrated that the combination of THC and ethanol was neurotoxic beginning on P4 with a maximum on P7, declined by P10, and was absent by P14 (Fig 2). ⌬9-Tetrahydrocannabinol Enhances the Neurotoxic Effect of a GABAA Agonist and an N-methyl-Daspartate Antagonist in the Neonatal Rat Brain P7 rats received a single toxic dose of phenobarbital (at 0 hours; 40mg/kg intraperitoneally), a GABAA receptor agonist, and two consecutive (at 0 and 2 hours) doses of THC (1–10mg/kg SC) to assess whether cannabinoid receptor activation influences toxicity of GABAA agonists. THC dose-dependently potentiated proapoptotic effects of phenobarbital in neonatal rats (Fig 3A). P7 rats received a single suprathreshold toxic dose of MK801 (at 0 hours; 0.5mg/kg intraperitoneally), a noncompetitive NMDA receptor antagonist, and two consecutive (at 0 and 2 hours) doses of THC (5 or 10mg/kg SC) to assess whether cannabinoid receptor activation influences toxicity of NMDA antagonists. THC dose-dependently potentiated proapoptotic effects of MK801 in neonatal rats (see Fig 3B). ⌬9-Tetrahydrocannabinol and Ethanol Coexposure Upregulates Cannabinoid CB1 but Not CB2 Receptor Levels Rat pups received two consecutive (at 0 and 2 hours) doses of THC (5mg/kg SC) in combination with eth- 46 Annals of Neurology Vol 64 No 1 July 2008 anol (1.1 or 1.5gm/kg SC, total ethanol doses of 2.2 and 3gm/kg) or vehicle. Other age-matched rat pups received THC or ethanol alone. Brain samples from the thalamus and dorsal subiculum, two highly sus- Fig 2. The neurotoxic effect of ethanol (EtOH) and ⌬9tetrahydrocannabinol (THC) coexposure is age dependent. P2 to P14 rat pups received two consecutive doses of EtOH (total dose 2.2gm/kg) in combination with THC (total dose 10mg/kg; solid circles), EtOH (total dose of 2.2gm/kg; open circles) or THC (total dose 10mg/kg; open triangles). Neurodegeneration was assessed 24 hrs later. Depicted is the mean ⫾ standard error of the mean of percentage increase of degeneration scores in relation to vehicle treated littermates. The combination of THC and EtOH was neurotoxic beginning on P4 with a maximum on P7, declined by P10 and was absent by P14. ***p ⬍ 0.001 in relation to vehicle-treated rats (one-way analysis of variance with Bonferroni’s multiple comparison test). ceptible brain regions to the neurotoxic effect of THC and ethanol coexposure (see the Table), were collected 24 hours after drug injection, and the protein extracts were subsequently subjected to Western blot analysis for CB1 receptor immunoreactivity. A significant upregulation of CB1 receptor levels was evident in brain regions susceptible to the proapoptotic effect of ethanol and THC (Fig 4). No upregulation of CB2 receptor levels was seen in subiculum and thalamus by Western blot analysis after administration of THC, ethanol, or their combination (see Fig 4A). Fig 3. ⌬9-Tetrahydrocannabinol (THC) enhances the neurotoxic effect of a GABAA agonist and an N-methyl-D-aspartate (NMDA) receptor antagonist in the neonatal rat brain. Postnatal day 7 (P7) rats received either (A) a single toxic dose of phenobarbital (PHE; 40mg/kg intraperitoneally [IP]), and two consecutive doses of THC (total doses of 2–20mg/kg THC), or (B) a single suprathreshold toxic dose of MK801 (0.5mg/kg, IP), a non-competitive NMDA receptor antagonist, and two consecutive doses of THC (total doses 10 or 20mg/kg). Neurodegeneration was assessed in silver stained sections 24 hrs after the first dose. Data are expressed as mean cell densities of degenerating neurons ⫾ standard error of the mean (silver stained sections) and were compared with one-way analysis of variance with Bonferroni’s multiple-comparison test. ***,⫹⫹⫹p ⬍ 0.001; asterisks refer to comparisons with VEH and plus signs refer to comparison with the PHE or MK 801 group. CB1 Receptor Antagonist SR141716A Reduces Neurotoxicity of Ethanol in Infant Rat Brain To further assess contribution of CB1 receptors to the neurotoxic effect of ethanol, we coadministered ethanol (1.8gm/kg at 0 and 2 hours; total ethanol dose 3.6gm/ kg) with SR141716A (0.4mg/kg at 0 hours) to P7 rats and quantified apoptosis in their brains 24 hours later. The CB1 receptor antagonist slightly but significantly reduced proapoptotic effects of ethanol in infant rat brain (Fig 5). CB1 Receptor Knock-out Mice Are Less Susceptible to the Neurotoxic Effect of Ethanol P7 mice (CB1⫹/⫹, CB1⫹/⫺, and CB1⫺/⫺) received two consecutive doses (at 0 and 2 hours) of ethanol (1.8gm/kg; total dose 3.6gm/kg). Numeric densities of degenerating neurons (per cubic millimeter) of 10 individual brain regions of vehicle- and drug-treated mice were assessed 24 hours after the first dose on TUNELstained sections. CB1⫺/⫺ mice were significantly less susceptible to the neurotoxic effect of ethanol (*p ⬍ 0.05) than wild-type mice. Heterozygous CB1⫹/⫺ mice demonstrated a trend toward lower levels of neuronal death in susceptible brain regions, but this effect did not reach significance (Fig 6). Discussion In this article, we show that cannabinoid receptor agonists (THC and WIN 55,212-2) have proapoptotic properties in the neonatal rat brain when coadministered with subtoxic doses of ethanol without influencing plasma ethanol levels. THC and WIN 55,212-2, when administered alone, did not induce neurodegeneration that could be detected by our experimental approach. Neuronal degeneration became disseminated and severe when THC was combined with a mildly toxic ethanol dose. The detrimental effect of THC in combination with ethanol was completely counteracted by the CB1 receptor antagonist SR141716A, indicating that activation of CB1 receptors potentiates toxicity of ethanol. Both components of ethanol’s proapoptotic action, that is, agonism at GABAA receptors and antagonism at NMDA receptors, were enhanced by CB1 agonists. THC enhanced the proapoptotic effect of the GABAA agonist phenobarbital and the N-methyl-Daspartate receptor antagonist dizocilpine. In addition to the obvious modulatory effect of exogenous CB1 receptor agonists on ethanol’s neurotoxic action, endogenous cannabinoids appear to also modulate proapoptotic effect of ethanol. Support for such modulatory effect is provided by the finding that infant CB1 receptor knock-out mice were less susceptible to the neurotoxic effect of ethanol than were wild-type mice. This effect cannot be explained by altered metabolism of ethanol in CB1⫺/⫺ mice, as these mice have been shown to actually reach higher blood alcohol levels Hansen et al: Neurotoxicity of Cannabinoids 47 Fig 4. ⌬9-Tetrahydrocannabinol (THC) and ethanol (EtOH) coexposure upregulates cannabinoid CB1 but not CB2 receptor levels. Rat pups received two consecutive doses of THC (total dose 10mg/kg) in combination with EtOH (total doses of 2.2 and 3gm/kg) or vehicle (VEH). Other age-matched rat pups received THC or EtOH alone. Representative Western blots from thalamus (THA) and subiculum (Sub) are shown in (A). (B, C) Results of the analysis of the signal density ratio of CB1 receptor/␤-actin are shown in the thalamus and the subiculum (means ⫾ standard error of the mean). Ratios were normalized to VEH levels. **p ⬍ 0.01, ***p ⬍ 0.001 (compared to vehicle; one-way analysis of variance with Bonferroni’s multiple-comparisons test). than wild-type mice after receiving the same ethanol dose.21 In addition, the CB1 receptor antagonist SR141716A reduced neurotoxicity of ethanol in the rat brain. Finally, our results show that, in the developing brain, the endogenous cannabinoid system is modified in the context of exposure to ethanol; upregulation of CB1 receptors occurred in brain regions in which ethanol-induced apoptotic neurodegeneration takes place. CB1 receptor distribution in the rat brain has been well characterized among the different animal species. Autoradiographic,22 immunohistochemical,23 and in situ hybridization studies24 have provided CB1 localization in the rat CNS in great detail. Features of these receptors include their atypical location during developmental stages (being present mostly in fiber-enriched areas),4,25,26 together with their abundant and selective presence in discrete anatomic regions and neuronal cir- 48 Annals of Neurology Vol 64 No 1 July 2008 cuits within the CNS, such as the cortex, hippocampal formation, basal ganglia, and cerebellum.4,22 In the developing brain, CB1 receptors are abundantly expressed and undergo remarkable spatiotemporal distributional changes during prenatal and postnatal mammalian brain development.2,27 The areas of the brain where we observed a potentiating effect of CB1 receptor agonists on ethanol neurotoxicity are known to express functional CB1 receptors in the early postnatal period in the rat.27 However, the pattern of degeneration elicited by combining ethanol and THC is the same as the one elicited by higher doses of ethanol and only partly matches the distribution pattern of CB1 receptors. Areas of the brain, such as the brainstem, which express high levels of CB1 messenger RNA and functional CB1 receptors at this age do not display neuroapoptosis after administration of ethanol alone or in combination with Fig 5. CB1 receptors modulate ethanol (EtOH) neurotoxicity in infant rodent brain. Postnatal day 7 (P7) rats were injected with vehicle, EtOH (total dose 3.6gm/kg) or EtOH plus SR141716A (0.4mg/kg) and neurodegeneration was evaluated on TUNEL-stained sections 24 hours later. The CB1 receptor antagonist SR141716A had a small but significant inhibitory effect on EtOH neurotoxicity. Data are expressed as mean cell densities of degenerating neurons ⫾ standard error of the mean (n ⫽ 11–15 per group). *p ⬍ 0.05, compared with EtOH group, analysis of variance with Bonferroni’s multiplecomparisons test). THC. This finding, together with the results showing that CB1⫺/⫺ mice are less susceptible to the neurotoxicity of ethanol and that the CB1 receptor antagonist SR141716A reduces ethanol neurotoxicity, support a modulatory effect of cannabinoids on ethanol neurotoxicity. In the brain, CB1 receptors exhibit a presynaptic location, suggesting that the endogenous cannabinoid system could play a prominent role in synaptic neurotransmission.2 The endogenous cannabinoid system has been involved in the control of the “depolarizationinduced suppression of inhibition” and of the “depolarization-induced suppression of excitation.”28 Endocannabinoids are thought to act as retrograde signaling molecules in these forms of short-term synaptic plasticity by activating CB1 receptors, a fact that may have important consequences on reward or memory processes, or both.28,29 The endocannabinoid system modulates synaptic transmission and long-term regulation of synaptic plasticity throughout the brain. Cannabinoids alter coherent neuronal network activity in the brain.29 In the neocortex, pyramidal neurons synthesize and release the endogenous cannabinoid ligands anandamide and 2-arachidonylglycerol.30 In addition, there are high levels of expression of the type-1 cannabinoid receptor (CB1)23,31 and the endocannabinoid metabolizing enzymes fatty acid amide hydrolase and monoacyl glycerol lipase.32 Data from different groups indicate that endocannabinoids promote homosynaptic and heterosynaptic long-term depression in the hippocampus, striatum, amygdala, and nucleus accumbens.28 This action may be a crucial aspect of marijuana-induced alteration of memory, motor, and reward brain systems, respectively, and expands the physiological relevance of the endogenous cannabinoid system. Furthermore, experimental evidence supports involvement of endocannabinoid signaling in fundamental developmental processes such as cell proliferation, migration, differentiation, and survival during patterning of the CNS.4 Currently, the literature contains only sparse and inconclusive information on morphological and biochemical changes in the developing brain after cannabinoid exposure. In this context, a number of the neurobehavioral effects produced by maternal exposure to cannabinoids are believed to be linked to interference with neurotransmitter systems, as demonstrated in animal studies. Hence, persistent alterations in neurotransmitter homeostasis are apparent in adult rats that were prenatally exposed to THC.33 CB1 receptor activation modulates cortical activity. For example, administration of THC decreases sensory-evoked cortical responses in anesthetized animals.34 Furthermore, Wallace and colleagues35 demonstrated that THC or the cannabinoid receptor agonist WIN55,212-2 completely eliminated recurrent epileptic activity in a rat pilocarpine model of epilepsy. Depolarization-induced suppression of excitation and inhibition, which are important forms of short-term retrograde neuronal plasticity, involve endocannabinoids and the activation of presynaptic cannabinoid CB1 receptors. It has been reported that CB1dependent, depolarization-induced suppression of excitation can be elicited from autaptic cultures of excitatory mouse hippocampal neurons.36 Together, these results suggest that an endogenous endocannabinoid tone is present, and that activation of CB1 receptors in the cortex reduces excitatory synaptic transmission. In vitro studies suggest that endocannabinoid signaling in the cortex also targets inhibitory synaptic transmission because CB1 receptor staining appears to be primarily localized on GABAergic interneurons. In layer 2/3 of the neocortex, depolarization of principal neurons re- Hansen et al: Neurotoxicity of Cannabinoids 49 Fig 6. CB1 receptor knock-out mice were less susceptible to the neurotoxic effect of ethanol (EtOH) than were wild-type mice. (A) Postnatal day 7 (P7) mice (wild type, CB1⫹/⫺, and CB1⫺/⫺) received two consecutive doses of ethanol (total dose 3.6gm/kg). Neurodegeneration was assessed 24 hours after the first dose on TUNEL stained sections. Data are expressed as mean cell densities of degenerating neurons ⫾ standard error of the mean; statistical analysis was done by means of one way analysis of variance (ANOVA). Wild-type mice are more susceptible to EtOH neurotoxicity than rats. CB1⫺/⫺ mice were significantly less susceptible to the neurotoxic effect of EtOH (*p ⬍ 0.05) than were wild-type mice. White bars represent levels of apoptosis in vehicle-treated mice. There was no difference in basal levels of apoptosis between wild-type and CB1⫺/⫺ mice (n ⫽ 6 –10 per group). (B) Light micrographs of TUNEL-stained sections taken from the brains of CB1⫹/⫹, CB1⫹/⫺, and CB1⫺/⫺ mice from subiculum and thalamus 24 hours following treatment with EtOH. Dark spots are TUNEL-positive cells which are more numerous in the CB1⫹/⫹ brains. Bar ⫽ 50m. sults in a retrograde endocannabinoid-mediated suppression of GABA release.37 Our observations indicate that CB1 receptor activation modulates GABAergic and glutamatergic neurotransmission and primes the developing rat brain to suffer apoptotic neuronal death when exposed to an NMDA antagonist, a GABAA agonist, or ethanol, which combines both mechanisms of action. The mechanism for this potentiating effect of CB1 receptor agonists on developmental drug neurotoxicity may relate to inhibition of glutamate release, inhibition of ion channels, or inhibition of NMDA receptors. Upregulation of CB1 receptors after exposure to THC alone or in combination with ethanol may constitute an additional modulatory component. Whether proapoptotic intracellular signaling cascades activated by ethanol, phenobarbital, or dizocilpine are modulated directly by CB1 receptor activation currently remains an open question. Our study does not address and 50 Annals of Neurology Vol 64 No 1 July 2008 does not exclude involvement of the CB2 receptor in the modulatory effect of THC on ethanol neurotoxicity. Interestingly, we did detect protein expression for the CB2 receptor in the infant rat brain but observed no effect on this expression after drug treatment. Studies using selective agonists and antagonists of the CB2 receptor are needed to address this aspect, which gains interest given recent findings on distribution of CB2 receptors in the mammalian brain.38 Our results do not contradict reports on neuroprotective effect of CB1 receptor agonists in developmental models of brain injury caused by asphyxia,39 because the excitotoxic mechanism of degeneration in these models is quite different from the mechanism that leads to apoptosis by ethanol. Ethanol, similar to NMDA antagonists and GABAA agonists, is inducing neuronal death in the developing brain by suppressing synaptic activity, which, at this age, elicits vital trophic signals for immature neurons.40 Our results on the acute modulatory role of endogenous and exogenous cannabinoids on developmental neurotoxicity of ethanol need to be further addressed in studies after longer survival periods. With the use of behavioral and stereological techniques, such studies would explore whether acute changes reflect permanent neuronal loss and lead to behavioral deficits. The results of the acute studies have interesting potential therapeutic implications including the use of CB1 receptor antagonists for preventing brain damage in fetuses and neonates exposed to ethanol, sedatives, and/or anticonvulsant drugs. This work was supported by The Alfred Benzon Foundation (H.H.H.) and Bundesministerium für Bildung und Forschung (BMBF) (01GZ0305, C.I.). References 1. Piomelli D. The molecular logic of endocannabinoid signalling. Nat Rev Neurosci 2003;4:873– 884. 2. Glass M, Dragunow M, Faull RLM. Cannabinoid receptors in the human brain: a detailed anatomical and quantitative autoradiographic study in the fetal, neonatal and adult human brain. Neuroscience 1997;77:299 –318. 3. Herkenham M, Lynn AB, Little MD, et al. Cannabinoid receptor localization in brain. Proc Natl Acad Sci U S A 1990; 87:1932–1936. 4. Harkany T, Guzman M, Galve-Roperh I, et al. The emerging functions of endocannabinoid signaling during CNS development. Trends Pharmacol Sci 2006;28:83–92. 5. Pertwee RG. Pharmacology of cannabinoid CB1 and CB2 receptors. Pharmacol Ther 1997;74:129 –180. 6. Johnston L, O’Malley P, Bachman J. Adolescent substance use. Epidemiology and implications for public policy. Pediatr Clin North Am 1995;42:241–260. 7. Adams ME, Gfroerer JC, Rouse BA. Epidemiology of substance abuse including alcohol and cigarette smoking. Ann N Y Acad Sci 1989;562:123–132. 8. Ganapathy VV, Prasad PV, Ganapathy ME, Leibach FH. Drugs of abuse and placental transport. Adv Drug Deliv Rev 1999;38:99 –110. 9. Zuckerman B, Frank DA, Hingson R, et al. Effects of maternal marijuana and cocaine on fetal growth. N Engl J Med 1989; 320:762–776. 10. Huizink AC, Mulder EJ. Maternal smoking, drinking or cannabis use during pregnancy and neurobehavioral and cognitive functioning in human offspring. Neurosci Biobehav Rev 2006; 30:24 – 41. 11. Fried PA, Watkinson B, Gray R. Differential effects on cognitive functioning in 13- to 16-year-olds prenatally exposed to cigarettes and marijuana. Neurotoxicol Teratol 2003;25: 427– 436. 12. Mereu G, Fa M, Ferraro L, et al. Prenatal exposure to a cannabinoid agonist produces memory deficits linked to dysfunction in hippocampal long term potentiation and glutamate release. Proc Natl Acad Sci U S A 2003;100:4915– 4920. 13. Bittigau P, Sifringer M, Genz K, et al. Antiepileptic drugs and apoptotic neurodegeneration in the developing brain. Proc Natl Acad Sci U S A 2002;99:15089 –15094. 14. Ikonomidou C, Bittigau P, Ishimaru MJ, et al. Ethanolinduced apoptotic neurodegeneration and fetal alcohol syndrome. Science 2000;287:1056 –1060. 15. Ikonomidou C, Bosch F, Miksa M, et al. Blockade of NMDA receptors and apoptotic neurodegeneration in the developing brain. Science 1999;283:70 –74. 16. Olney JW, Young C, Wozniak DF, et al. Do pediatric drugs cause developing neurons to commit suicide? Trends Pharmacol Sci 2004;25:135–139. 17. Young C, Olney JW. Neuroapoptosis in the infant mouse brain triggered by a transient small increase in blood alcohol concentration. Neurobiol Dis 2005;22:548 –554. 18. Marsicano G, Wojtak CT, Azad SC, et al. The endogenous cannabinoid system controls extinction of aversive memories. Nature 2002;418:530 –534. 19. DeOlmos JS, Ingram WR. An improved cupric-silver method for impregnation of axonal and terminal degeneration. Brain Res 1971;33:523–529. 20. Gundersen HJG, Bendtsen TF, West MJ. Some new, simple and efficient stereological methods and their use in pathological research and diagnosis. APMIS 1988;96:379 –394. 21. Lallemand F, de Witte P. Ethanol induces higher BEC in CB1 cannabinoid receptor knockout mice while decreasing ethanol preference. Alcohol Alcohol 2005;40:54 – 62. 22. Herkenham M, Lynn AB, Johnson MR, et al. Characterization and localization of cannabinoid receptors in rat brain: a quantitative in vitro autoradiographic study. J Neurosci 1991;11: 563–583. 23. Tsou K, Brown S, Sanudo-Pena MC, et al. Immunohistochemical distribution of cannabinoid CB1 receptors in the rat central nervous system. Neuroscience 1998;83:393– 411. 24. Mailleux P, Vanderhaeghen JJ. Distribution of neuronal cannabinoid receptor in the adult rat brain: a comparative receptor binding radioautography and in situ hybridization histochemistry. Neuroscience 1992;48:655– 668. 25. Berrendero F, Sepe N, Ramos JA, et al. Analysis of cannabinoid receptor binding and mRNA expression and endogenous cannabinoid contents in the developing rat brain during late gestation and early postnatal period. Synapse 1999;33:181– 191. 26. Berguis P, Rajnicek AM, Morozov YM, et al. Hardwiring the brain: endocannabinoids shape neuronal connectivity. Science 2007;316:1212–1216. 27. Berrendero F, Garcia-Gil L, Hernandez ML, et al. Localization of mRNA expression and activation of signal transduction mechanisms for cannabinoid receptor in rat brain during fetal development. Development 1998;125:3179 –3188. 28. Chevaleyre V, Takajashi KA, Castillo PE. Endocannabinoidmediated synaptic plasticity in the CNS. Annu Rev Neurosci 2006;29:37–76. 29. Bernard C, Milh M, Morozov YM, et al. Altering cannabinoid signaling during development disrupts neuronal activity. Proc Natl Acad Sci U S A 2005;102:9388 –9393. 30. Stella N, Piomelli D. Receptor-dependent formation of endogenous cannabinoids in cortical neurons. Eur J Pharmacol 2001; 425:189 –196. 31. Marsicano G, Lutz B. Expression of the cannabinoid receptor CB1 in distinct neuronal subpopulations in the adult mouse forebrain. Eur J Neurosci 1999;11:4213– 4225. 32. Gulyas AI, Cravatt BF, Bracey MH, et al. Segregation of two endocannabinoid-hydrolyzing enzymes into pre- and postsynaptic compartments in the rat hippocampus, cerebellum and amygdala. Eur J Neurosci 2004;20:441– 458. 33. Fernández-Ruiz JJ, Berrendero F, Hernández ML, Ramos JA. The endogenous cannabinoid system and brain development. Trends Neurosci 2000;23:14 –20. Hansen et al: Neurotoxicity of Cannabinoids 51 34. Wilkison DM, Pontzer N, Hosko MJ. Slowing of cortical somatosensory evoked activity by delta 9-tetrahydrocannabinol and dimethylheptylpyran in alpha-chloralose-anesthetized cats. Neuropharmacology 1982;21:705–709. 35. Wallace MJ, Blair RE, Falenski KW, Martin BR, DeLorenzo RJ. The endogenous cannabinoid system regulates seizure frequency and duration in a model of temporal lobe epilepsy. J Pharmacol Exp Ther 2003;307:129 –137. 36. Straiker A, Mackie K. Depolarization-induced suppression of excitation in murine autaptic hippocampal neurons. J Physiol 2005;569:501–517. 52 Annals of Neurology Vol 64 No 1 July 2008 37. Bodor AL, Katona I, Nyiri G, et al. Endocannabinoid signaling in rat somatosensory cortex: laminar differences and involvement of specific interneuron types. J Neurosci 2005;25:6845– 6856. 38. Van Sickle MD, Duncan M, Kinglsey PJ, et al. Identification and functional characterization of brainstem cannabinoid CB2 receptors. Science 2005;310:329 –332. 39. Martinez-Orgado J, Fernandez-Frutos B, González R, et al. Neuroprotection by the cannabinoid agonist WIN-55212 in an in vivo newborn rat model of acute severe asphyxia. Mol Brain Res 2003;114:132–139. 40. Hardingham GE, Bading H. The Yin and Yang of NMDA receptor signalling. Trends Neurosci 2003;26:81– 89.