Cannabinoids and neuronal damage: Differential effects of THC, AEA and 2-AG on activated microglial cells and degenerating neurons in excitotoxically lesioned rat organotypic hippocampal slice cultures

Experimental Neurology 203 (2007) 246 – 257 www.elsevier.com/locate/yexnr

Cannabinoids and neuronal damage: Differential effects of THC, AEA and 2-AG on activated microglial cells and degenerating neurons in excitotoxically lesioned rat organotypic hippocampal slice cultures Susanne Kreutz, Marco Koch, Chalid Ghadban, Horst-Werner Korf, Faramarz Dehghani ⁎ Dr. Senckenbergische Anatomie, Institut für Anatomie 2, Johann Wolfgang Goethe-Universität Frankfurt, Theodor-Stern-Kai 7, 60590 Frankfurt/Main, Germany Received 22 June 2006; revised 7 August 2006; accepted 10 August 2006 Available online 27 September 2006

Abstract Cannabinoids (CBs) are attributed neuroprotective effects in vivo. Here, we determined the neuroprotective potential of CBs during neuronal damage in excitotoxically lesioned organotypic hippocampal slice cultures (OHSCs). OHSCs are the best characterized in vitro model to investigate the function of microglial cells in neuronal damage since blood-borne monocytes and T-lymphocytes are absent and microglial cells represent the only immunocompetent cell type. Excitotoxic neuronal damage was induced by NMDA (50 μM) application for 4 h. Neuroprotective properties of 9-carboxy-11-nor-delta-9-tetrahydrocannabinol (THC), N-arachidonoylethanolamide (AEA) or 2-arachidonoylglycerol (2-AG) in different concentrations were determined after co-application with NMDA by counting degenerating neurons identified by propidium iodide labeling (PI+) and microglial cells labeled by isolectin B4 (IB+4 ). All three CBs used significantly decreased the number of IB+4 microglial cells in the dentate gyrus but the number of PI+ neurons was reduced only after 2-AG treatment. Application of AM630, antagonizing CB2 receptors highly expressed by activated microglial cells, did not counteract neuroprotective effects of 2-AG, but affected THC-mediated reduction of IB+4 microglial cells. Our results indicate that (1) only 2-AG exerts neuroprotective effects in OHSCs; (2) reduction of IB+4 microglial cells is not a neuroprotective event per se and involves other CB receptors than the CB2 receptor; (3) the discrepancy in the neuroprotective effects of CBs observed in vivo and in our in vitro model system may underline the functional relevance of invading monocytes and T-lymphocytes that are absent in OHSCs. © 2006 Elsevier Inc. All rights reserved. Keywords: Neuronal damage; N-methyl-D-aspartate; Neuroprotection; 9-Carboxy-11-nor-delta-9-tetrahydrocannabinol; N-arachidonoylethanolamine; 2-Arachidonoylglycerol

Introduction In the course of CNS pathologies such as trauma, inflammation or excitotoxic injury a massive loss of neurons occurs, concomitantly with activation of astrocytes and proliferation and rapid invasion of microglial cells (Heppner et al., 1998). Accumulation and activation of microglial cells and astrocytes at sites of the initial injury play a crucial role for the further fate of neurons that survived the initial lesion, a process known as secondary neuronal damage. The secondary neuronal damage is the result of intertwined neurotoxic and neuroprotective events

⁎ Corresponding author. Fax: +49 69 6301 6017. E-mail address: [email protected] (F. Dehghani). 0014-4886/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.expneurol.2006.08.010

and apparently involves complex interactions between activated microglial cells, neurons and astrocytes. In recent years numerous studies have demonstrated the modulating effects of cannabinoids (CBs) in neuronal transmission, pain regulation and behavior. CBs are also believed to influence neuronal injury. Phytocannabinoids derived from Cannabis sativa and synthetically designed CBs led to the discovery of an endogenous CB system that comprises endogenously produced ligands specific for CB receptors and enzymes for endogenous ligand biosynthesis and inactivation (Piomelli, 2003). Endogenous CBs are actively secreted and inactivated by neurons, microglial cells and astrocytes and the CB system is involved in the communication between neurons (Stella, 2004). The effects of CBs are mediated by the CB1 receptor (R), expressed on neurons, astrocytes and microglial

S. Kreutz et al. / Experimental Neurology 203 (2007) 246–257

cells or by the CB2R highly expressed on immunocompetent cells. On activated microglial cells, CB2R expression is much higher than CB1R expression indicating a predominant function of the CB2R for microglial cell physiology (Carlisle et al., 2002; Facchinetti et al., 2003; Klegeris et al., 2003; Walter et al., 2003). In neurotransmission, CBs are able to decrease excitation by inhibiting glutamatergic transmission thus acting as neuroprotective agents (Marsicano et al., 2003). In contrast, CBs attenuate GABAergic transmission and may thus augment excitatory neuronal damage (Mechoulam and Lichtman, 2003). However, several in vivo studies (Marsicano et al., 2003; Mechoulam et al., 2002; Panikashvili et al., 2001) have shown that phytocannabinoids like 9-carboxy-11-nor-delta-9-tetrahydrocannabinol (THC) and endogenous CBs like N-arachidonoylethanolamide (AEA) and 2-arachidonoylglycerol (2-AG) exert neuroprotective effects. In these in vivo studies, authors outlined the functional relevance of CB1R in neuroprotection. Since the role of the CB1R in neuronal injury is well documented (Khaspekov et al. 2004; Marsicano et al., 2003), we compared neuroprotective properties of THC, AEA and 2-AG after excitotoxic damage in organotypic hippocampal slice cultures (OHSCs) with special focus on CB2R function. We used the specific CB2R antagonist AM630 to analyze the effects of THC, AEA and 2-AG on microglial cells labeled by isolectin B4 and degenerated neurons identified by propidium iodide staining. In the model system of OHSCs the morphology and functional neuronal circuits are preserved, neurons and glial cells are arranged in an organotypic form resembling the in vivo situation. Since blood-borne monocytes and T-lymphocytes are apparently absent, microglial cells appear as the only immunocompetent cell type in this system (Hailer et al., 1996). Thus, OHSCs are suitable to specifically analyze the effects of CBs on microglial cells in context of neuroprotection. Materials and methods All animal experimentation was carried out in accordance with the Policy on the Use of Animals in Neuroscience Research and the Policy on Ethics as approved by the Society for Neuroscience and by the European Communities Council Directive (89/609/EEC). Preparation of the organotypic hippocampal slice cultures (OHSCs) To prepare OHSCs, 8-day-old Wistar rats (Charles River, Sulzfeld, Germany) were killed by decapitation and their brains were quickly dissected under sterile conditions (Stoppini et al., 1991). After removal of the frontal pole and the cerebellum, the brains were placed in minimal essential medium (MEM, Gibco BRL Life Technologies, Eggenstein, Germany) containing 1% (v/v) glutamine (Gibco) at 4°C. These preparations were cut into 350 μm-thick slices using a sliding vibratome (Leica VT 1000 5, Leica Microsystems AG, Wetzlar, Germany). Six to eight OHSCs were obtained from

247

each brain and immediately transferred into cell culture inserts (pore size 0.4 μm, Falcon, BD Biosciences Discovery Labware, Bedford, MA, USA). These cell culture inserts were then placed in 6-well culture dishes (Falcon) that contained 1 ml culture medium per well. The culture medium consisted of 50% (v/v) MEM, 25% (v/v) Hanks' balanced salt solution (HBSS, Gibco), 25% (v/v) normal horse serum (NHS, Gibco), 2% (v/v) glutamine, 1 μg/ml insulin (Boehringer, Mannheim, Germany), 1.2 mg/ml glucose (Braun, Melsungen, Germany), 0.1 mg/ml streptomycin (Sigma Chemicals, Deisenhofen, Germany), 100 U/ml penicilline (Sigma) and 0.8 μg/ ml vitamin C (Sigma), pH 7.4. The culture dishes were incubated at 35°C in a fully humidified atmosphere with 5% CO2 and the cell culture medium was changed every second day. THC (Sigma, Deisenhofen, Germany) and the CB2R antagonist AM630 (Tocris Cookson Inc., St. Louis, MO, USA) were dissolved in DMSO and stored at − 20°C. AEA and 2-AG (Tocris Cookson Inc., St. Louis, MO, USA) were dissolved in ethanol (EtOH) and stored at − 20°C. Prior to application of the CBs to OHSCs, the CBs were diluted in cell culture medium (for composition of cell culture medium, see above). Each CB was added to OHSCs with a maximal final solvent concentration of 0.1% (v/v) DMSO or 0.007% (v/v) EtOH in cell culture medium. Control experiments with equimolar concentrations of DMSO or EtOH alone did not show any significant effects on the parameters investigated in this study (data not shown). Treatment protocol The preparations were divided into different groups and treated according to the following protocols (Fig. 1). CTL: Unlesioned OHSCs (n = 14) were kept in culture medium for 9 days in vitro (div) without any treatment and served as control slices. NMDA: OHSCs were kept in culture medium for 6 days, lesioned with 50 μM NMDA (Sigma, Deisenhofen, Germany; n = 13) for 4 h and kept in culture medium for another 3 days. Cannabinoid or AM630: Unlesioned OHSCs were incubated in medium containing THC (0.03 μM, 0.3 μM, 3 μM or 15 μM; n = 6), AEA (0.001 μM, 0.01 μM, 0.1 μM or 1 μM; n = 6), 2-AG (0.001 μM, 0.01 μM, 0.1 μM or 1 μM; n = 6) or AM630 (10 μM; n = 6) from 6 div till 9 div. NMDA + Cannabinoid or AM630: OHSCs were treated on 6 div with 50 μM NMDA and THC (0.03 μM, 0.3 μM, 3 μM or 15 μM; n = 12), AEA (0.001 μM, 0.01 μM, 0.1 μM or 1 μM; n = 9), 2AG (0.001 μM, 0.01 μM, 0.1 μM or 1 μM; n = 10) or AM630 (0.1 μM, 1 μM or 10 μM; n = 10) for 4 h. Thereafter, the slices were incubated till 9 div with medium containing only THC, AEA, 2-AG or AM630 at the above-given concentrations. AM630 + NMDA + Cannabinoid: OHSCs were preincubated on 6 div with 10 μM of the CB2R antagonist AM630 15 min before AM630 + NMDA + THC (3 μM; n = 7), AM630 + NMDA + AEA (0.01 μM; n = 8) or AM630 + NMDA + 2-AG (0.001 μM; n = 7) was applied for 4 h, respectively. Thereafter, the slices were again preincubated with 10 μM AM630 for 15 min and then incubated in medium containing AM630 +

248

S. Kreutz et al. / Experimental Neurology 203 (2007) 246–257

Fig. 1. Overview of the experimental protocols. CTL: unlesioned OHSCs were cultured in control medium for 9 days in vitro (div) without any treatment. NMDA: OHSCs were cultured in control medium for 6 div, then lesioned with 50 μM NMDA for 4 h and incubated in control medium for another 3 div. Cannabinoid or AM630: unlesioned OHSCs were treated with THC (0.03 μM, 0.3 μM, 3 μM or 15 μM), AEA (0.001 μM, 0.01 μM, 0.1 μM or 1 μM), 2-AG (0.001 μM, 0.01 μM, 0.1 μM or 1 μM) or AM630 (10 μM) from day 6 till day 9. Culture medium was removed every second day. NMDA + Cannabinoid or AM630: simultaneous treatment with NMDA + THC (0.03 μM, 0.3 μM, 3 μM or 15 μM), AEA (0.001 μM, 0.01 μM, 0.1 μM or 1 μM), 2-AG (0.001 μM, 0.01 μM, 0.1 μM or 1 μM) or AM630 (10 μM) at 6 div for 4 h and THC (0.03 μM, 0.3 μM, 3 μM or 15 μM), AEA (0.001 μM, 0.01 μM, 0.1 μM or 1 μM), 2-AG (0.001 μM, 0.01 μM, 0.1 μM or 1 μM) or AM630 (10 μM) from day 6 till day 9. NMDA + Cannabinoid + AM630: pretreatment of slices with the CB2 receptor antagonist AM630 (10 μM) for 15 min before AM630 + NMDA + THC (3 μM), AM630 + NMDA + AEA (0.01 μM) or AM630 + NMDA + 2-AG (0.001 μM) exposure for 4 h. Thereafter the slices were again preincubated with AM630 for 15 min and then incubated in medium containing AM630 + THC, AM630 + AEA or AM630 + 2-AG from day 6 till day 9.

THC, AM630 + AEA or AM630 + 2-AG in the above indicated concentrations from 6 div till 9 div. The concentration of 10 μM AM630 was chosen, because it is the most effective and commonly used dosage reported in the literature (Mukherjee et al., 2004; Shoemaker et al., 2005). On 9 div, OHSCs were incubated with 5 μg/ml propidium iodide (PI) for 2 h prior to fixation to stain the nuclei of degenerating neurons intensely red. Uptake of PI and its use for identifying degenerating neurons in OHSCs has been established previously (Hailer et al., 2005; Pozzo Miller et al., 1994). All slices were then fixed with a 4% (w/v) solution of paraformaldehyde for 4 h. Lectin histochemistry and confocal laser scanning microscopy For IB4 staining, OHSCs were removed from the cell culture inserts, placed into a 24-well-plate (Greiner Bio-One, Essen, Germany) and washed with PBS containing 0.03% (v/ v) Triton-X-100 (PBS-T) for 10 min (one time). The slices were then incubated with normal goat serum (NGS, diluted 1:20 in PBS-T) for 1 h, stained with FITC-conjugated IB4 (Vector laboratories, Burlingame, CA, USA) diluted 1:50 in

PBS-T containing 0.25% (w/v) bovine serum albumine (BSA) for 3 h, washed with PBS-T for 10 min (three times) and finally coverslipped with DAKO fluorescent mounting medium (DAKO Diagnostika GmbH, Hamburg, Germany). OHSCs were analyzed with a Zeiss confocal laser scanning microscope (LSM 510, Zeiss, Göttingen, Germany). For visualization of IB4-labeled (IB4+) microglial cells monochromatic light at 488 nm with a dichroic beam splitter (FT 488/543) and an emission bandpass filter of 505–530 nm was used. For detection of PI-labeled (PI+) degenerating neurons, monochromatic light at 543 nm and an emission bandpass filter of 585– 615 nm were used. Confocal images were obtained at 160-fold magnification. Using the Z-mode of the LSM 510, the 350 μm thick OHSCs were optically cut into 2 μm-thick sections. Images were then converted into a binary image, segmented and measured semi-automatically by the Zeiss KS 400 system (Zeiss). Numbers of IB4+ or PI+ cells were counted in every third optical section through the granule cell layer (GCL) of the dentate gyrus (cells/GCL) as described previously (Dehghani et al., 2004; Hailer et al., 2001; Kohl et al., 2003). Statistical analysis The average number of IB4+ microglial cells and PI+ degenerating neurons in OHSCs was calculated for each experimental group derived from at least 3 different animals per concentration investigated. The one-way ANOVA test was used to determine whether the effect of the treatment on the number of IB4+ microglial cells or PI+ degenerating neurons was statistically significant (p < 0.05). If significant differences were detected, the one-way ANOVA test was followed by the Dunnett's test for multiple comparisons. Unlesioned OHSCs treated with THC, AEA or 2-AG (mean values) were compared to control OHSCs (mean values); lesioned OHSCs treated with THC, AEA or 2-AG (mean values) were compared to OHSCs treated with NMDA alone (mean values). Furthermore, lesioned OHSCs treated with THC, AEA or 2-AG (mean values) were compared to lesioned OHSCs treated with the CB2R antagonist AM630 plus THC, AEA or 2-AG (mean values). Results Preparations of unlesioned control OHSCs showed a good neuronal preservation (Fig. 2). Confocal laser scanning microscopy confirmed that the microglial cells were mainly localized in the molecular and plexiform layers and displayed a ramified phenotype. In the granule cell layer (GCL) of the dentate gyrus (DG) of control OHSCs, only very few IB4+ microglial cells (8.6/GCL) and virtually no degenerating, PI+ neurons (2.1/GCL) were found (Fig. 3). In OHSCs lesioned with NMDA, a massive accumulation of IB4+ microglial cells occurred at the site of neuronal injury throughout all layers of the DG including the GCL (Fig. 2). Quantitative analysis revealed that the number of IB4+ microglial cells in the GCL was significantly higher in lesioned OHSCs (116.3/GCL, p < 0.01) than in unlesioned control OHSCs. Also the number of PI+ degenerating neurons in the GCL was

S. Kreutz et al. / Experimental Neurology 203 (2007) 246–257

249

Fig. 2. Effects of THC, AEA and 2-AG on organotypic hippocampal slice cultures. Confocal laser scanning microscopy images in overview and in higher magnification, double labeled with IB4 (microglial cells, green) and PI (degenerating neurons, red). Untreated control OHSCs showed a very good preservation. Almost no PI+ cells and few ramified IB+4 microglial cells were found. Slices treated with NMDA (50 μM for 4 h) showed a massive accumulation of amoeboid IB+4 microglial cells and a dramatic increase in PI+ neurons. Slices treated with THC (3 μM) only showed almost no PI+ cells and very few IB+4 microglial cells. Simultaneous treatment of slices with NMDA + THC (3 μM) resulted in a strong reduction in the number of IB+4 microglial cells. However, numerous PI+ cells are still visible. Slices treated with AEA (0.01 μM) only showed virtually no PI+ cells and only very few IB+4 microglial cells. Simultaneous treatment of slices with NMDA + AEA (0.01 μM) resulted in a strong reduction in the number of IB+4 microglial cells, but the number of PI+ neurons remained high. Slices treated with 2-AG (0.001 μM) only showed almost no PI+ cells and only very few IB+4 microglial cells. Simultaneous treatment of slices with NMDA + 2-AG (0.001 μM) resulted in a strong reduction of IB+4 microglial cells as well as of PI+ neurons. Arrows point to microglial cells, arrowheads indicate condensed nuclei of degenerating neurons. GCL = granular cell layer, HI = Hilus. Scale bars = 50 μm.

significantly higher in NMDA lesioned slices (200/GCL, p < 0.01) than in unlesioned control OHSCs (Fig. 3). Unlesioned OHSCs treated with different concentrations of THC, AEA or 2-AG (Fig. 2) from 6 div until 9 div contained only very few, ramified IB4+ microglial cells which mainly appeared in the noncellular layers of the DG, thus, resembling

preparations of unlesioned control OHSCs (THC: 0.03 μM, 18.6/GCL; 0.3 μM, 20.6/GCL; 3 μM, 17.3/GCL; 15 μM, 6.3/ GCL. AEA: 0.001 μM, 19/GCL; 0.01 μM, 16.8/GCL; 0.1 μM, 13.6/GCL; 1 μM, 12.7/GCL. 2-AG: 0.001 μM, 16.6/GCL; 0.01 μM, 17.3/GCL; 0.1 μM, 12.6/GCL; 1 μM, 10.6/GCL, for all groups p > 0.05 versus CTL; Figs. 3A, C, E). Also the

250

S. Kreutz et al. / Experimental Neurology 203 (2007) 246–257

Fig. 3. Statistical analysis of the mean number of IB+4 microglial cells and PI+ neurons in unlesioned slices treated with THC, AEA or 2-AG and in lesioned slices simultaneously treated with THC, AEA or 2-AG. (A) Unlesioned slices treated with different concentrations of THC contained only very few IB+4 microglial cells. In lesioned slices simultaneously treated with THC, the mean number of IB+4 microglial cells was significantly reduced when compared to OHSCs treated with NMDA alone. This reduction occurred in a concentration-dependent manner. Maximal reduction of microglial cells (67.3%) was observed at a THC concentration of 15 μM. (B) THC-treated unlesioned slices contained almost no PI+ degenerating neurons. Treatment of lesioned slices with different concentrations of THC did not affect the mean number of PI+ nuclei. (C) Treatment of unlesioned slices with different concentrations of AEA resulted in a very low number of IB+4 microglial cells. In AEAtreated lesioned OHSCs, the mean number of IB+4 microglial cells was significantly reduced when compared to OHSCs treated with NMDA alone. The significant reduction of microglial cells was observed at all used AEA concentrations. (D) AEA-treated unlesioned slices contained only very few PI+ degenerating neurons. Treatment of lesioned slices with different concentrations of AEA did not affect the mean number of PI+ nuclei. (E) 2-AG treatment of unlesioned slices resulted in a very low number of IB+4 microglial cells. In 2-AG-treated lesioned OHSCs, the mean number of IB+4 microglial cells was significantly reduced when compared to OHSCs treated with NMDA alone. The significant reduction of microglial cells was observed at all used 2-AG concentrations. (F) Treatment of unlesioned slices with 2-AG at different concentrations resulted in a very low number of PI+ degenerating neurons. Simultaneous treatment of lesioned slices with 2-AG at concentrations of 0.001 μM, 0.1 μM and 1 μM significantly reduced the mean number of PI+ nuclei. **p < 0.01, *p < 0.05 versus NMDA. Error bars represent SEM.

S. Kreutz et al. / Experimental Neurology 203 (2007) 246–257

numbers of PI+ neurons in THC, AEA or 2-AG-treated slices resembled untreated control slices (THC: 0.03 μM, 0.8/GCL; 0.3 μM, 0.6/GCL; 3 μM, 1.6/GCL; 15 μM, 1.3/GCL. AEA: 0.001 μM, 6.6/GCL; 0.01 μM, 3.4/GCL; 0.1 μM, 2.6/GCL; 1 μM, 1.2/GCL. 2-AG: 0.001 μM, 5.3/GCL; 0.01 μM, 1.3/GCL; 0.1 μM, 2/GCL; 1 μM, 6.8/GCL, for all groups p > 0.05 versus CTL, Figs. 3B, D, F). Treatment of lesioned slices with THC resulted in a strong reduction of IB4+ microglial cells (Fig. 2). This effect was concentration-dependent and microglial cell numbers were maximally reduced at THC concentrations of 3 μM (52.9/ GCL; p < 0.01 versus NMDA) and 15 μM (33.3/GCL; p < 0.01 versus NMDA; Fig. 3A). The remaining microglial cells were mainly located in the GCL of the DG and they had an amoeboid morphology similar to those observed in OHSCs treated with

251

NMDA alone. Treatment with THC did not prevent neuronal damage (Fig. 2): the number of PI+ neurons in OHSCs treated with four different concentrations of THC was not significantly different from the number in OHSCs treated with NMDA alone (THC: 0.03 μM, 171.8/GCL; 0.3 μM, 201.6/GCL; 3 μM, 245.1/ GCL; 15 μM, 241.8/GCL, for all groups p > 0.05 versus NMDA, Fig. 3B). Treatment of lesioned slices with AEA resulted in a significant reduction of IB4+ microglial cells (Fig. 2). This effect occurred at all AEA concentrations used (0.001 μM, 101.3/ GCL, p < 0.01 versus NMDA; 0.01 μM, 92.7/GCL, p < 0.01 versus NMDA; 0.1 μM, 108/GCL, p < 0.05 versus NMDA; 1 μM, 96.6/GCL, p < 0.01 versus NMDA; Fig. 3C). Interestingly, the remaining microglial cells were located in the molecular layer of the DG. They had an amoeboid morphology similar to the microglial cells observed in OHSCs treated with NMDA alone. Treatment with AEA did not prevent neuronal damage (Fig. 2): the number of PI+ nuclei in OHSCs treated with four different concentrations of AEA did not significantly differ from the number in OHSCs treated with NMDA alone (0.001 μM, 205.1/GCL; 0.01 μM, 150.5/GCL; 0.1 μM, 178.2/GCL; 1 μM, 179.9/GCL, for all groups p > 0.05 versus NMDA, Fig. 3D). Treatment of lesioned OHSCs with 2-AG significantly reduced the number of IB4+ microglial cells (Fig. 2) at all concentrations used (0.001 μM, 91.4/GCL; 0.01 μM, 90.5/ GCL; 0.1 μM, 92.8/GCL; 1 μM, 87.5/GCL, for all groups p < 0.01 versus NMDA, Fig. 3E). The remaining IB4+ microglial cells were mainly located in the GCL of the DG and showed a round or amoeboid morphology. In contrast to THC and AEA, 2-AG significantly reduced the number of PI+ neurons (Fig. 2) compared to slices treated with NMDA alone (0.001 μM, 123.4/GCL, p < 0.01; 0.01 μM, 158.6/GCL, p < 0.05; 0.1 μM, 150.7/GCL, p < 0.05; 1 μM, 134.2/GCL, p < 0.01 for all groups versus NMDA, Fig. 3F). Pretreatment of lesioned slices with the CB2R antagonist AM630, followed by concomitant application of AM630 and THC (3 μM) attenuated the THC-mediated reduction of IB4+ microglial cells in the GCL of the DG (69.9/GCL, p < 0.01 versus NMDA + THC (3 μM); Figs. 4, 5A). Moreover the application of AM630 led to a significant reduction of PI+ Fig. 4. Effects of the CB2R antagonist AM630 (10 μM) on THC + NMDA, AEA + NMDA and 2-AG + NMDA-treated OHSCs. Confocal laser scanning microscopy images in overview and in higher magnification, double labeled with IB4 (microglial cells, green) and PI (degenerating neurons, red). Slices treated with AM630 + THC (3 μM) + NMDA showed an increased number of IB+4 microglial cells as compared to slices treated with THC (3 μM) + NMDA. Numbers of PI+ cells were reduced in AM630 + THC (3 μM) + NMDA treated slices, as compared to slices treated with THC (3 μM)+ NMDA. Incubation of OHSCs with AM630 + AEA (0.01 μM) + NMDA had no effect on the number of IB+4 microglial cells but led to a strong reduction of PI+ neurons, as compared to OHSCs incubated with AEA (0.01 μM)+ NMDA. Slices treated with AM630 +2AG (0.001 μM)+ NMDA showed even fewer IB+4 microglial cells when compared to slices treated with 2-AG (0.001 μM) + NMDA. However, incubation of OHSCs with AM630 + 2-AG (0.001 μM)+ NMDA resulted in a strong reduction of PI+ neurons, which is comparable to the reduction followed treatment with 2-AG (0001 μM)+ NMDA. OHSCs treated with AM630(10 μM)+ NMDA showed a strong accumulation of IB+4 microglial cells and a massive increase in the number of PI+ neurons. Arrows point to microglial cells, arrowheads indicate condensed nuclei of degenerating neurons. Scale bars= 50 μm.

252

S. Kreutz et al. / Experimental Neurology 203 (2007) 246–257

S. Kreutz et al. / Experimental Neurology 203 (2007) 246–257

neurons when compared to slices treated with NMDA + THC (3 μM) alone (162.8/GCL, p < 0.05 versus NMDA + THC (3 μM); Figs. 4, 5B). Pretreatment of slices with AM630 prior to application of NMDA + AEA (0.01 μM) did not significantly affect the number of IB4+ microglial cells when compared to slices treated with NMDA + AEA (0.01 μM) alone (95.7/GCL, p > 0.05 versus NMDA + AEA (0.01 μM); Figs. 4, 5C), but it led to a significant reduction of PI+ degenerating neurons when compared to slices treated with NMDA + AEA (0.01 μM) alone (79.8/GCL, p < 0.05 versus NMDA + AEA (0.01 μM); Figs. 4, 5D). Pretreatment of slices with AM630 prior to application of NMDA + 2-AG (0.001 μM) led to a significant reduction in the number of IB4+ microglial cells as compared to slices treated with NMDA + 2-AG (0.001 μM) alone (58.8/GCL, p < 0.01 versus NMDA + 2-AG (0.001 μM); Figs. 4, 5E). With regard to number of PI+ neurons, AM630 did not lead to a significant reduction of PI+ degenerating neurons when compared to slices treated with NMDA + 2-AG (0.001 μM) alone (110.8/GCL, p > 0.05 versus NMDA + 2-AG (0.001 μM); Figs. 4, 5F). Unlesioned OHSCs treated with AM630 (10 μM) from 6 div until 9 div contained only very few, ramified IB4+ microglial cells which mainly appeared in the noncellular layers of the DG, thus, resembling preparations of unlesioned control OHSCs (AM630: 10 μM, 2.4/GCL, versus CTL, Fig. 5G). Also the numbers of PI+ neurons in AM630 (10 μM)-treated slices resembled untreated control slices (AM630: 10 μM, 4.6/GCL, versus CTL; Fig. 5H). Treatment of lesioned slices with AM630 resulted in a significant reduction of IB4+ microglial cells (Figs. 4, 5G). Microglial cells located in the GCL of the DG displayed an amoeboid morphology similar to those observed in OHSCs treated with NMDA alone. Treatment with AM630 did not prevent neuronal damage (Fig. 4): the number of PI+ neurons in OHSCs treated with AM630 was not significantly different from the number in OHSCs treated with NMDA alone (AM630: 10 μM, 174.2/GCL; p > 0.05 versus NMDA; Fig. 5H). Discussion In the present study, we compared the effects of the phytocannabinoid 9-carboxy-11-nor-delta-9-tetrahydrocannabinol (THC) and the endogenous cannabinoids (CBs) N-arachidonoylethanolamine (AEA) and 2-arachidonoylglycerol (2-AG) on neuronal damage in excitotoxically lesioned organotypic hippocampal slice cultures (OHSCs). OHSCs represent a suitable model to examine neuronal damage since they comprise

253

functional cellular networks involving neurons and glial cells (Adamchik et al., 2000; Adembri et al., 2004; Hailer et al., 2001; Wolf et al., 2002). Thus, the OHSCs have proven an excellent model to analyze whether and how putative neuroprotective substances influence resident cells of the central nervous system and to determine the impact of the interaction between resident cells on neurodegenerative and neuroprotective phenomena. Our analyses were performed in the dentate gyrus (DG) because this region possesses the highest density of NMDA receptors in the hippocampal formation as demonstrated by [3H] MK-801 binding studies (Bekenstein et al., 1990). Granular cell layers were considered more favorable over the CA1 region for quantitative analyses. A neuronal demise in the dentate gyrus regularly leads to degeneration of CA3 neurons which send their axons via Schaffer collaterals to the CA1 region where they can subsequently induce an anterograde degeneration. As PI-labeling is not specific enough to distinguish between excitotoxically damaged and anterogradely degenerating neurons, quantitative analyses were performed in granular cell layer of the hippocampus formation only. Staining with Griffonia simplicifolia isolectin B4 (IB4), a reliable marker for microglial cells, was performed to visualize microglial cell numbers in the DG. IB4 binds to galactose containing glucoconjugates and labels microglial cells independently of their activation state (Lyons et al., 2000; Martinez-Contreras et al., 2002; Matsuura et al., 1997; Streit and Kreutzberg, 1987). Using neuron- or gliaspecific markers in combination with PI-labeling, we recently demonstrated that the PI-labeling was restricted to degenerating neurons. Triple labeling with NeuN, PI or IB4 in lesioned OHSCs revealed phagocytosis of degenerating PI+ and NeuN+ granule cells by microglial cells (Hailer et al., 2005). Several studies have revealed that THC, AEA and 2-AG elicit neuroprotective effects in different brain areas in vivo as well as in neuronal single cell cultures (Hampson et al., 1998; Mechoulam, 2004; Sinor et al., 2000; van der Stelt et al., 2001). Endocannabinoids are only produced “on demand” by specific enzymes and are rapidly inactivated, for instance AEA by the fatty acid amid hydrolase (FAAH) and 2-AG by the monoacylglycerollipase (MGL; for review see Piomelli, 2003). Under basal conditions, only very low amounts of endogenous CBs can be detected, e.g. 1.5 ± 0.7 pmol AEA and 8.6 ± 3.5 pmol 2-AG per unstimulated hippocampal slice. Whereas glutamatergic stimulation had only a moderate or no effect at all on AEA levels (Stella et al., 1997; Stella and Piomelli, 2001; Walter and Stella, 2003), glutamatergic treatment caused a fourfold increase in 2-AG levels (32.0 ± 11.2 pmol per hippocampal slice; Stella et al., 1997). Due to

Fig. 5. Statistical analysis of the mean number of IB+4 microglial cells and PI+ neurons in slices, treated with 10 μM AM630 +THC (3 μM) + NMDA, +AEA (0.01 μM) + NMDA or +2-AG (0.001 μM) + NMDA. (A) The microglial cell-reducing effect of THC was significantly blocked by the CB2R antagonist AM630; (B) Incubation of slices with AM630 led to a significant reduction of PI+ degenerating neurons when compared to slices treated with THC (3 μM) + NMDA, but no difference was found when compared to the NMDA group. (C) AM630 did not significantly antagonize the AEA-induced reduction of IB+4 microglial cells in lesioned slices. (D) The number of PI+ degenerating neurons was significantly reduced when slices were treated with AM630 in addition to AEA (0.01 μM) + NMDA. (E) Incubation of 2-AG (0.001 μM) + NMDA-treated slices with AM630 significantly reduced the number of IB+4 microglial cells when compared to slices treated with 2-AG (0.001 μM) + NMDA. (F) AM630 did not significantly antagonize the 2-AG-induced PI+ cell reducing effect in lesioned slices. **p < 0.01, *p < 0.05 versus THC (3 μM) + NMDA (A, B); **p < 0.01, *p < 0.05 versus AEA (0.01 μM) + NMDA (C, D); **p < 0.01, *p < 0.05 versus 2-AG (0.001 μM) + NMDA (E, F). (G) AM630 significantly reduced the number of IB+4 microglial cells (*p < 0.05 versus NMDA). (H) The mean number of and PI+ neurons in lesioned slices treated with AM630 (10 μM) was not significantly different from slices treated with NMDA only (p > 0.05 versus NMDA). Error bars represent the SEM.

254

S. Kreutz et al. / Experimental Neurology 203 (2007) 246–257

this “on demand” upregulation, a neuroprotective role for endocannabinoids has been proposed. In the present study, only 2-AG, the most abundant CB in the CNS (Mechoulam et al., 1995; Sugiura et al., 1995), exerted neuroprotective effects on granule cells in the dentate gyrus (DG) of the hippocampus after NMDA-treatment. Neuroprotective effects of 2-AG have also been found in the in vivo model of closed head injury (Panikashvili et al., 2001): 2-AG led to a better clinical recovery and reduced brain edema, the infarct volume and the neuronal cell loss in the CA3 field of the hippocampus. Because of the absence of blood-derived leukocytes in the OHSCs, it is reasonable to assume that 2-AG elicits its neuroprotective effects by actions on resident CNS cells. In contrast to 2-AG, AEA, the second endogenous CB applied exogenously, did not exert neuroprotective effects on dentate gyrus neurons in the OHSCs, whereas, in different in vivo model systems (van der Stelt et al., 2001; Veldhuis et al., 2003), AEA and its metabolites were reported to attenuate cytotoxic edema after intracranial injection of the Na+/K+ATPase inhibitor ouabain. AEA also protected neurons from cell death in neuronal single cell cultures (Sinor et al., 2000). Like AEA, THC did not elicit neuroprotective effects in OHSCs even though THC has been shown to protect neurons from excitotoxicity in vivo (van der Stelt et al., 2001) as well as in single cell culture (Hampson et al., 1998). Taken together, only 2-AG exerted neuroprotection in the OHSCs whereas AEA and THC failed to do so. In this context, it should be kept in mind that it is the 2-AG level that is strongly increased after pathological overstimulation of neurons (see above). Since the three CBs investigated differed with regard to their neuroprotective effects, it appeared interesting to analyze whether they also exerted different effects on microglial cells. The model system used here appears very suitable for this type of analysis since it does not contain blood-derived leukocytes like monocytes and T-lymphocytes, which are known to invade the injured brain parenchyma through the disrupted blood– brain barrier and, as demonstrated recently, transform their morphology to microglial like cells in zones of acute, anterograde axonal degeneration after entorhinal cortex lesion (Bechmann et al., 2005). Microglial cells are of particular interest because they appear to be involved in the process of secondary damage (Blight, 1992; Marty et al., 1991) either facilitating neuronal survival (Nagata et al., 1993) or releasing neurotoxic substances and killing neurons (Colton and Gilbert, 1987; Giulian et al., 1994). Because of the latter effect, microglial cells are thought to contribute to the process of secondary damage occurring after ischemic and traumatic lesions of the CNS (Kim and Ko, 1998; Schwab and Bartholdi, 1996). After excitotoxic brain lesions, microglial cells migrate towards the sites of injury and produce large amounts of oxygen radicals leading to secondary damage and further loss of brain function. Ullrich and colleagues reported that this microglial migration is strongly controlled by the integrin CD11a and PARP-1. Downregulation of PARP or CD11a strongly abrogated microglial migration and prevented secondary neuronal damage (Ullrich et al., 2001a,b). Furthermore, the directed migration of leukocytes within the denervated hippocampus in

response to axonal injury caused by entorhinodentate lesions is dependent of innate glial production of chemokines. CCR2 was reported to be critical for migration of monocytes and T-cells to the site of injury, since denervated mice hippocampi lacking this receptor were not invaded by these cells (Babcock et al., 2003). Microglial cells are potential target cells for cannabinoids, which may act upon the CB2 receptor (R) or the abnormalcannabidiol (abn-CBD)R, both expressed on activated microglial cells (Carlisle et al., 2002; Facchinetti et al., 2003; Jarai et al., 1999; Walter et al., 2003). We therefore focused in our study on the functional relevance of the CB2R by use of the specific CB2R antagonist AM630. In contrast to the CB2R, the role of the CB1R in neuroprotection is well documented: in an in vivo model of kainic acid-induced excessive seizures, the involvement of CB1R in neuroprotection was proven by using a CB1Rnull mutant mice (CB1−/−; Marsicano et al., 2003). In an in vitro model system of OHSCs using kainic acid-induced excessive seizures, use of (CB1−/−) mice and pharmacological blockade of the CB1R with SR141716A again clearly demonstrated the role of the CB1R in the protection against excitotoxicity (Khaspekov et al., 2004). THC caused a concentration-dependent reduction in the number of microglial cells, which was significantly antagonized by the specific CB2R antagonist AM630 at 10 μM, the most effective and commonly used dosage (Mukherjee et al., 2004; Shoemaker et al., 2005). This finding indicates that the THCinduced reduction of microglial cell number involves the CB2R. With regard to the number of PI+ degenerating neurons, the application of the CB2R antagonist AM630 to THC + NMDAtreated slices reduced the number of PI+ degenerating neurons when compared to OHSC treated with NMDA + THC; however, no difference was found when AM630 + THC + NMDA-treated slices were compared to OHSCs treated with NMDA alone. The endogenous CB AEA also caused a significant reduction in the number of microglial cells. This AEA-induced reduction, however, does not seem to be mediated by the CB2R alone since the selective CB2R antagonist AM630 did not affect AEA-induced reduction of IB4+ microglial cells. This observation is supported by very recent findings demonstrating that AEA-mediated effects were not inhibited by AM630 in a model of neuropathic pain (Guindon and Beaulieu, 2006). AM630 significantly reduced the number of PI+ degenerating neurons when applied to AEA + NMDA-treated slices. This observation was very surprising since AEA alone did not reduce the number of PI+ neurons in lesioned slices. As recently shown, the involvement of several intracellular signal cascades, like the mitogen-activated protein kinase-phosphatase-1 (MKP-1) cascade or the mitogen-activated protein kinase (MAPK) cascade, regulates and controls the inflammatory response within the CNS. High levels of cannabinoids like WIN55,212-2 or AEA were reported to induce the MKP-1 and to switch off the MAPK pathway on microglial cells influencing their migration and NO-release (Eljaschewitsch et al., 2006). These findings may explain why blockade of the CB2R with AM630 leads to reduced neuronal damage in slices treated with AM630 + AEA + NMDA, since the reduced secretion of NO and of other proinflammatory cytokines from

S. Kreutz et al. / Experimental Neurology 203 (2007) 246–257

microglial cells is reported to be correlated with reduced neuronal damage. In contrast to THC and AEA, 2-AG significantly reduced the number of both, microglial cells and degenerated PI+ neurons in lesioned OHSCs. 2-AG is known to be a strong agonist on the CB2R, highly expressed by microglial cells. Interestingly, the application of the CB2R antagonist AM630 led to an even stronger reduction of IB4+ microglial cells. This phenomenon, however, clearly indicates that the 2-AGmediated reduction of IB4+ microglial cells is not exclusively driven by the CB2R. As mentioned above, microglial cells also express the abn-CBDR for which 2-AG is also a strong agonist. A possible explanation may be the assumption that 2AG-mediated reduction of IB4+ microglial cells is modulated by the CB2R and another CBR like the abn-CBDR which drives the 2-AG-mediated reduction of microglial cells. Ongoing studies with specific agonists/antagonists for the abn-CBDR will help to prove this hypothesis. Selective blockade of the CB2R by application of AM630 did not influence the 2-AGmediated neuroprotection implying that this effect is not associated with CB2R activation as widely accepted by the scientific community (Marsicano et al., 2003; Panikashvili et al., 2006; van der Stelt and Di Marzo, 2005). Notably, the number of microglial cells has been reduced by all three cannabinoids tested, but only 2-AG was able to protect neurons from excitotoxicity. These differential responses suggest that a reduction of microglial cells is not a neuroprotective event per se and that there is no general correlation between neuronal degeneration and the number of microglial cells. As already mentioned above, monocytes and activated Tcells are absent in the model of OHSCs and it is tempting to speculate that the absence of these cells in the OHSCs is related to the failure of AEA and THC to exert neuroprotective properties as observed in several in vivo systems. This assumption may be supported by the notion that 2-AG exhibits opposite effects on T-cells as compared to AEA and THC with regard to the general inflammatory response, migration and release of cytokines. In general, 2-AG stimulates immunological/inflammatory responses (Sugiura et al., 2004) while AEA and THC reduce them. Kaminski and colleagues reported that THC selectively inhibits T-cell-dependent humoral immune responses through direct inhibition of accessory T-cell functions (Kaminski, 1998; Schatz et al., 1997). Furthermore, 2-AG has been shown to induce migration of human peripheral blood monocytes and human peripheral blood natural killer cells while AEA and THC failed to do so (Kishimoto et al., 2003; Kishimoto et al., 2005). Using a chemotaxis chamber technique, Maestroni and colleagues reported that 2-AG exerted a powerful chemotactic activity on immature and mature dendritic cells. This effect was completely inhibited by the CB2R antagonist SR 144528 (Maestroni, 2004). 2-AG, AEA and THC also differ in modulating functions of immunocompetent cells, e.g. inhibiting cytokine production in macrophages/mononuclear cells (Berdyshev et al., 1997; Gallily et al., 2000). THC and AEA have been shown to diminish LPS-induced NO and IL-6 production in a concentration-dependent manner, whereas 2-

255

AG inhibited the production of IL-6 but slightly increased iNOS-dependent NO production (Chang et al., 2001). Gallily et al. (2000) reported that 2-AG had no effect on NOproduction at all. Kishimoto et al. (2005) further demonstrated that 2-AG induced accelerated production of IL-8 in human promyelocytic leukemia HL-60 cells, whereas AEA failed to induce enhanced IL-8 production. Even though the effects of 2-AG, AEA and THC on neuroprotection on the one hand and on blood-derived monocytes and activated T-cells on the other hand are not directly comparable, they indicate that 2-AG exerts its effects through mechanisms that are distinct from those used by AEA and THC. One may conclude that 2-AG exerts neuroprotection without involvement of blood-derived monocytes and activated T-cells while AEA and THC utilize a mechanism of neuroprotection that depends on the availability of monocytes and activated T-cells. The results obtained in single cell cultures may be considered as being too artificial because the diverse neuron–glia interactions occurring in vivo and in the OHSCs are absent in such a system. In summary, two major conclusions can be drawn from the present study: (1) Reduction of microglial cells is not a neuroprotective event per se. (2) CBs differ with respect to their neuroprotective capacity suggesting that 2-AG, AEA and THC use different mechanisms to exert neuroprotection and affect different cell types in the CNS and/or invading monocytes and activated T-cells after cerebral injury. Acknowledgments This study was supported by the Deutsche Forschungsgemeinschaft (Graduiertenkolleg Neuronale Plastizität to S K) and the Dr. Paul and Cilli Weill-Stiftung. The authors gratefully acknowledge the expert technical assistance by Mrs. Iris Habazettl. References Adamchik, Y., Frantseva, M.V., Weisspapir, M., Carlen, P.L., Perez Velazquez, J.L., 2000. Methods to induce primary and secondary traumatic damage in organotypic hippocampal slice cultures. Brain Res. Brain Res. Protoc. 5, 153–158. Adembri, C., Bechi, A., Meli, E., Gramigni, E., Venturi, L., Moroni, F., De Gaudio, A.R., Pellegrini-Giampietro, D.E., 2004. Erythropoietin attenuates post-traumatic injury in organotypic hippocampal slices. J. Neurotrauma 21, 1103–1112. Babcock, A.A., Kuziel, W.A., Rivest, S., Owens, T., 2003. Chemokine expression by glial cells directs leukocytes to sites of axonal injury in the CNS. J. Neurosci. 23, 7922–7930. Bechmann, I., Goldmann, J., Kovac, A.D., Kwidzinski, E., Simburger, E., Naftolin, F., Dirnagl, U., Nitsch, R., Priller, J., 2005. Circulating monocytic cells infiltrate layers of anterograde axonal degeneration where they transform into microglia. FASEB J. 19, 647–649. Bekenstein, J.W., Bennett Jr., J.P., Wooten, G.F., Lothman, E.W., 1990. Autoradiographic evidence that NMDA receptor-coupled channels are located postsynaptically and not presynaptically in the perforant path-dentate granule cell system of the rat hippocampal formation. Brain Res. 514, 334–342. Berdyshev, E.V., Boichot, E., Germain, N., Allain, N., Anger, J.P., Lagente, V., 1997. Influence of fatty acid ethanolamides and delta9-tetrahydrocannabinol on cytokine and arachidonate release by mononuclear cells. Eur. J. Pharmacol. 330, 231–240.

256

S. Kreutz et al. / Experimental Neurology 203 (2007) 246–257

Blight, A.R., 1992. Macrophages and inflammatory damage in spinal cord injury. J. Neurotrauma 9 (Suppl. 1), S83–S91. Carlisle, S.J., Marciano-Cabral, F., Staab, A., Ludwick, C., Cabral, G.A., 2002. Differential expression of the CB2 cannabinoid receptor by rodent macrophages and macrophage-like cells in relation to cell activation. Int. Immunopharmacol. 2, 69–82. Chang, Y.H., Lee, S.T., Lin, W.W., 2001. Effects of cannabinoids on LPSstimulated inflammatory mediator release from macrophages: involvement of eicosanoids. J. Cell. Biochem. 81, 715–723. Colton, C.A., Gilbert, D.L., 1987. Production of superoxide anions by a CNS macrophage, the microglia. FEBS Lett. 223, 284–288. Dehghani, F., Conrad, A., Kohl, A., Korf, H.W., Hailer, N.P., 2004. Clodronate inhibits the secretion of proinflammatory cytokines and NO by isolated microglial cells and reduces the number of proliferating glial cells in excitotoxically injured organotypic hippocampal slice cultures. Exp. Neurol. 189, 241–251. Eljaschewitsch, E., Witting, A., Mawrin, C., Lee, T., Schmidt, P.M., Wolf, S., Hoertnagl, H., Raine, C.S., Schneider-Stock, R., Nitsch, R., Ullrich, O., Ullrich, O., Diestel, A., Eyupoglu, I.Y., Nitsch, R., Ullrich, O., Diestel, A., Bechmann, I., Homberg, M., Grune, T., Hass, R., Nitsch, R., 2006. The endocannabinoid anandamide protects neurons during CNS inflammation by induction of MKP-1 in microglial cells—Regulation of microglial expression of integrins by poly(ADP-ribose) polymerase-1—Turnover of oxidatively damaged nuclear proteins in BV-2 microglial cells is linked to their activation state by poly-ADP-ribose polymerase. Neuron 49, 67–79. Facchinetti, F., Del Giudice, E., Furegato, S., Passarotto, M., Leon, A., 2003. Cannabinoids ablate release of TNFalpha in rat microglial cells stimulated with lypopolysaccharide. Glia 41, 161–168. Gallily, R., Breuer, A., Mechoulam, R., 2000. 2-Arachidonylglycerol, an endogenous cannabinoid, inhibits tumor necrosis factor-alpha production in murine macrophages, and in mice. Eur. J. Pharmacol. 406, R5–R7. Giulian, D., Li, J., Leara, B., Keenen, C., 1994. Phagocytic microglia release cytokines and cytotoxins that regulate the survival of astrocytes and neurons in culture. Neurochem. Int. 25, 227–233. Guindon, J., Beaulieu, P., 2006. Antihyperalgesic effects of local injections of anandamide, ibuprofen, rofecoxib and their combinations in a model of neuropathic pain. Neuropharmacology 50, 814–823. Hailer, N.P., Jarhult, J.D., Nitsch, R., 1996. Resting microglial cells in vitro: analysis of morphology and adhesion molecule expression in organotypic hippocampal slice cultures. Glia 18, 319–331. Hailer, N.P., Wirjatijasa, F., Roser, N., Hischebeth, G.T., Korf, H.W., Dehghani, F., 2001. Astrocytic factors protect neuronal integrity and reduce microglial activation in an in vitro model of N-methyl-D-aspartate-induced excitotoxic injury in organotypic hippocampal slice cultures. Eur. J. Neurosci. 14, 315–326. Hailer, N.P., Vogt, C., Korf, H.W., Dehghani, F., 2005. Interleukin-1beta exacerbates and interleukin-1 receptor antagonist attenuates neuronal injury and microglial activation after excitotoxic damage in organotypic hippocampal slice cultures. Eur. J. Neurosci. 21, 2347–2360. Hampson, A.J., Grimaldi, M., Axelrod, J., Wink, D., 1998. Cannabidiol and (−) Delta9-tetrahydrocannabinol are neuroprotective antioxidants. Proc. Natl. Acad. Sci. U S. A. 95, 8268–8273. Heppner, F.L., Skutella, T., Hailer, N.P., Haas, D., Nitsch, R., 1998. Activated microglial cells migrate towards sites of excitotoxic neuronal injury inside organotypic hippocampal slice cultures. Eur. J. Neurosci. 10, 3284–3290. Jarai, Z., Wagner, J.A., Varga, K., Lake, K.D., Compton, D.R., Martin, B.R., Zimmer, A.M., Bonner, T.I., Buckley, N.E., Mezey, E., Razdan, R.K., Zimmer, A., Kunos, G., 1999. Cannabinoid-induced mesenteric vasodilation through an endothelial site distinct from CB1 or CB2 receptors. Proc. Natl. Acad. Sci. U S. A. 96, 14136–14141. Kaminski, N.E., 1998. Regulation of the cAMP cascade, gene expression and immune function by cannabinoid receptors. J. Neuroimmunol. 83, 124–132. Khaspekov, L.G., Brenz Verca, M.S., Frumkina, L.E., Hermann, H., Marsicano, G., Lutz, B., 2004. Involvement of brain-derived neurotrophic factor in cannabinoid receptor-dependent protection against excitotoxicity. Eur. J. Neurosci. 19, 1691–1698. Kim, W.K., Ko, K.H., 1998. Potentiation of N-methyl-D-aspartate-mediated

neurotoxicity by immunostimulated murine microglia. J. Neurosci. Res. 54, 17–26. Kishimoto, S., Gokoh, M., Oka, S., Muramatsu, M., Kajiwara, T., Waku, K., Sugiura, T., 2003. 2-Arachidonoylglycerol induces the migration of HL-60 cells differentiated into macrophage-like cells and human peripheral blood monocytes through the cannabinoid CB2 receptor-dependent mechanism. J. Biol. Chem. 278, 24469–24475. Kishimoto, S., Muramatsu, M., Gokoh, M., Oka, S., Waku, K., Sugiura, T., 2005. Endogenous cannabinoid receptor ligand induces the migration of human natural killer cells. J. Biochem. (Tokyo) 137, 217–223. Klegeris, A., Bissonnette, C.J., McGeer, P.L., 2003. Reduction of human monocytic cell neurotoxicity and cytokine secretion by ligands of the cannabinoid-type CB2 receptor. Br. J. Pharmacol. 139, 775–786. Kohl, A., Dehghani, F., Korf, H.W., Hailer, N.P., 2003. The bisphosphonate clodronate depletes microglial cells in excitotoxically injured organotypic hippocampal slice cultures. Exp. Neurol. 181, 1–11. Lyons, S.A., Pastor, A., Ohlemeyer, C., Kann, O., Wiegand, F., Prass, K., Knapp, F., Kettenmann, H., Dirnagl, U., 2000. Distinct physiologic properties of microglia and blood-borne cells in rat brain slices after permanent middle cerebral artery occlusion. J Cereb. Blood Flow Metab. 20, 1537–1549. Maestroni, G.J., 2004. The endogenous cannabinoid 2-arachidonoyl glycerol as in vivo chemoattractant for dendritic cells and adjuvant for Th1 response to a soluble protein. FASEB J. 18, 1914–1916. Marsicano, G., Goodenough, S., Monory, K., Hermann, H., Eder, M., Cannich, A., Azad, S.C., Cascio, M.G., Gutierrez, S.O., van der Stelt, M., LopezRodriguez, M.L., Casanova, E., Schutz, G., Zieglgansberger, W., Di Marzo, V., Behl, C., Lutz, B., 2003. CB1 cannabinoid receptors and on-demand defense against excitotoxicity. Science 302, 84–88. Martinez-Contreras, A., Huerta, M., Lopez-Perez, S., Garcia-Estrada, J., Luquin, S., Beas, Z.C., 2002. Astrocytic and microglia cells reactivity induced by neonatal administration of glutamate in cerebral cortex of the adult rats. J. Neurosci. Res. 67, 200–210. Marty, S., Dusart, I., Peschanski, M., 1991. Glial changes following an excitotoxic lesion in the CNS-I. Microglia/macrophages. Neuroscience 45, 529–539. Matsuura, K., Makino, H., Ogawa, N., 1997. Cyclosporin A attenuates the decrease in tyrosine hydroxylase immunoreactivity in nigrostriatal dopaminergic neurons and in striatal dopamine content in rats with intrastriatal injection of 6-hydroxydopamine. Exp. Neurol. 146, 526–535. Mechoulam, R., 2004. The promise of advances in the field of endocannabinoids. Neuro. Endocrinol. Lett. 25, 11–12. Mechoulam, R., Lichtman, A.H., 2003. Neuroscience. Stout guards of the central nervous system. Science 302, 65–67. Mechoulam, R., Ben Shabat, S., Hanus, L., Ligumsky, M., Kaminski, N.E., Schatz, A.R., Gopher, A., Almog, S., Martin, B.R., Compton, D.R., 1995. Identification of an endogenous 2-monoglyceride, present in canine gut, that binds to cannabinoid receptors. Biochem. Pharmacol. 50, 83–90. Mechoulam, R., Spatz, M., Shohami, E., 2002. Endocannabinoids and neuroprotection. Sci. STKE. 2002, RE5. Mukherjee, S., Adams, M., Whiteaker, K., Daza, A., Kage, K., Cassar, S., Meyer, M., Yao, B.B., 2004. Species comparison and pharmacological characterization of rat and human CB2 cannabinoid receptors. Eur. J. Pharmacol. 505, 1–9. Nagata, K., Takei, N., Nakajima, K., Saito, H., Kohsaka, S., 1993. Microglial conditioned medium promotes survival and development of cultured mesencephalic neurons from embryonic rat brain. J. Neurosci. Res. 34, 357–363. Panikashvili, D., Simeonidou, C., Ben Shabat, S., Hanus, L., Breuer, A., Mechoulam, R., Shohami, E., 2001. An endogenous cannabinoid (2-AG) is neuroprotective after brain injury. Nature 413, 527–531. Panikashvili, D., Shein, N.A., Mechoulam, R., Trembovler, V., Kohen, R., Alexandrovich, A., Shohami, E., 2006. The endocannabinoid 2-AG protects the blood–brain barrier after closed head injury and inhibits mRNA expression of proinflammatory cytokines. Neurobiol. Dis. 22, 257–264. Piomelli, D., 2003. The molecular logic of endocannabinoid signalling. Nat. Rev., Neurosci. 4, 873–884.

S. Kreutz et al. / Experimental Neurology 203 (2007) 246–257 Pozzo Miller, L.D., Mahanty, N.K., Connor, J.A., Landis, D.M., 1994. Spontaneous pyramidal cell death in organotypic slice cultures from rat hippocampus is prevented by glutamate receptor antagonists. Neuroscience 63, 471–487. Schatz, A.R., Lee, M., Condie, R.B., Pulaski, J.T., Kaminski, N.E., 1997. Cannabinoid receptors CB1 and CB2: a characterization of expression and adenylate cyclase modulation within the immune system. Toxicol. Appl. Pharmacol. 142, 278–287. Schwab, M.E., Bartholdi, D., 1996. Degeneration and regeneration of axons in the lesioned spinal cord. Physiol. Rev. 76, 319–370. Shoemaker, J.L., Joseph, B.K., Ruckle, M.B., Mayeux, P.R., Prather, P.L., 2005. The endocannabinoid noladin ether acts as a full agonist at human CB2 cannabinoid receptors. J. Pharmacol. Exp. Ther. 314, 868–875. Sinor, A.D., Irvin, S.M., Greenberg, D.A., 2000. Endocannabinoids protect cerebral cortical neurons from in vitro ischemia in rats. Neurosci. Lett. 278, 157–160. Stella, N., 2004. Cannabinoid signaling in glial cells. Glia 48, 267–277. Stella, N., Piomelli, D., 2001. Receptor-dependent formation of endogenous cannabinoids in cortical neurons. Eur. J. Pharmacol. 425, 189–196. Stella, N., Schweitzer, P., Piomelli, D., 1997. A second endogenous cannabinoid that modulates long-term potentiation. Nature 388, 773–778. Stoppini, L., Buchs, P.A., Muller, D., 1991. A simple method for organotypic cultures of nervous tissue. J. Neurosci. Methods 37, 173–182. Streit, W.J., Kreutzberg, G.W., 1987. Lectin binding by resting and reactive microglia. J. Neurocytol. 16, 249–260. Sugiura, T., Kondo, S., Sukagawa, A., Nakane, S., Shinoda, A., Itoh, K., Yamashita, A., Waku, K., 1995. 2-Arachidonoylglycerol: a possible endogenous cannabinoid receptor ligand in brain. Biochem. Biophys. Res. Commun. 215, 89–97. Sugiura, T., Oka, S., Gokoh, M., Kishimoto, S., Waku, K., 2004. New

257

perspectives in the studies on endocannabinoid and cannabis: 2-arachidonoylglycerol as a possible novel mediator of inflammation. J. Pharmacol. Sci. 96, 367–375. Ullrich, O., Diestel, A., Bechmann, I., Homberg, M., Grune, T., Hass, R., Nitsch, R., 2001a. Turnover of oxidatively damaged nuclear proteins in BV-2 microglial cells is linked to their activation state by poly-ADP-ribose polymerase. FASEB J. 15, 1460–1462. Ullrich, O., Diestel, A., Eyupoglu, I.Y., Nitsch, R., 2001b. Regulation of microglial expression of integrins by poly(ADP-ribose) polymerase-1. Nat. Cell Biol. 3, 1035–1042. van der Stelt, M., Di Marzo, V., 2005. Cannabinoid receptors and their role in neuroprotection. Neuromol. Med. 7, 37–50. van der Stelt, M., Veldhuis, W.B., Bar, P.R., Veldink, G.A., Vliegenthart, J.F., Nicolay, K., 2001. Neuroprotection by Delta9-tetrahydrocannabinol, the main active compound in marijuana, against ouabain-induced in vivo excitotoxicity. J. Neurosci. 21, 6475–6479. Veldhuis, W.B., van der Stelt, M., Wadman, M.W., van Zadelhoff, G., Maccarrone, M., Fezza, F., Veldink, G.A., Vliegenthart, J.F., Bar, P.R., Nicolay, K., Di Marzo, V., 2003. Neuroprotection by the endogenous cannabinoid anandamide and arvanil against in vivo excitotoxicity in the rat: role of vanilloid receptors and lipoxygenases. J. Neurosci. 23, 4127–4133. Walter, L., Franklin, A., Witting, A., Wade, C., Xie, Y., Kunos, G., Mackie, K., Stella, N., 2003. Nonpsychotropic cannabinoid receptors regulate microglial cell migration. J. Neurosci. 23, 1398–1405. Walter, L., Stella, N., 2003. Endothelin-1 increases 2-arachidonoyl glycerol (2AG) production in astrocytes. Glia 44, 85–90. Wolf, S.A., Fisher, J., Bechmann, I., Steiner, B., Kwidzinski, E., Nitsch, R., 2002. Neuroprotection by T-cells depends on their subtype and activation state. J. Neuroimmunol. 133, 72–80.