JBC Papers in Press. Published on March 11, 2008 as Manuscript M707960200 The latest version is at http://www.jbc.org/cgi/doi/10.1074/jbc.M707960200 The CB cannabinoid receptor controls myeloid progenitor trafficking 2 INVOLVEMENT IN THE PATHOGENESIS OF AN ANIMAL MODEL OF MULTIPLE SCLEROSIS Javier Palazuelos1, Natalie Davoust2#, Boris Julien1, Eric Hatterer2, Tania Aguado1, Raphael Mechoulam3, Cristina Benito4, Julian Romero4, Augusto Silva5, Manuel Guzmán1, Serge Nataf2*, Ismael Galve-Roperh1* 1Department of Biochemistry and Molecular Biology I, School of Biology, and Centro de Investigación Biomédica en Red sobre Enfermedades Neurodegenerativas (CIBERNED), Complutense University, 28040 Madrid (Spain) 2INSERM U433, IFR des Neurosciences de Lyon, Faculté de Médecine Laënnec, 69372 Lyon (France) 3Department of Medicinal Chemistry and Natural Products, School of Pharmacy, The Hebrew University, 91120 Jerusalem (Israel) 4Laboratorio de Apoyo a la Investigación, Fundación Hospital Alcorcón, 28922 Madrid (Spain) 5Centro de Investigaciones Biológicas (CIB-CSIC), Ramiro de Maeztu 9, 28040 Madrid (Spain) Running title: Control of myeloid progenitor recruitment by CB receptors 2 Cannabinoids are potential agents for the trafficking and its contribution to microglial development of therapeutic strategies against activation; and support the potential use of multiple sclerosis. Here we analyzed the role non-psychoactive CB agonists in therapeutic D 2 o w of the peripheral CB2 cannabinoid receptor in strategies for multiple sclerosis and other nlo the control of myeloid progenitor cell neuroinflammatory disorders. ad e trafficking towards the inflamed spinal cord d fro and their contribution to microglial activation During the last years it has been shown that the m h in an animal model of multiple sclerosis endocannabinoid (eCB) system, the endogenous ttp (experimental autoimmune encephalomyelitis, system targeted by active ingredients of the ://w w EAE). CB receptor knock-out mice showed hemp plant Cannabis sativa L. (1), is altered in w an exacerb2ated clinical score of the disease experimental autoimmune encephalomyelitis .jbc .o when compared to their wild-type littermates, (EAE), an animal model of multiple sclerosis rg and this occurred in concert with extended (MS) (2-4). In addition, cannabinoid receptor by/ axonal loss, T-lymphocyte (CD4+) infiltration agonist administration or modulation of the eCB gu e s and microglial (CD11b+) activation. Immature tone with eCB reuptake/degradation inhibitors t o n bone marrow-derived CD34+ myeloid improves pathological signs of the disease, D e c progenitor cells, which play a role in notably spasticity and tremor (3-7). Moreover, em b neuroinflammatory pathologies, were shown cannabinoids have shown to be effective not e r 3 to express CB receptors and to be only in palliating EAE symptoms but also as 1 2 , 2 abundantly recruited towards the spinal cords neuroprotective agents that, by promoting 0 1 8 of CB knock-out EAE mice. Bone marrow- oligodendrocyte survival (8), reducing 2 derived cell transfer experiments further demyelinated lesions (9) and attenuating evidenced the increased contribution of these neuronal loss (10, 11) contribute to delayed cells to microglial replenishment in the spinal progression of the disease. This protective role of cords of CB -deficient animals. In line with the eCB system during neuroinflammation is 2 these observations, selective pharmacological exerted, at least in part, by decreasing immune CB activation markedly reduced EAE cell activation and infiltration (12, 13). 2 symptoms, axonal loss and microglial Cannabinoid actions are mediated by the activation. CB receptor manipulation altered activation of two different G-protein coupled 2 the expression pattern of different receptors, namely CB and CB receptors (1). 1 2 chemokines (CCL2, 3, 5) and their receptors Alleviation of EAE symptoms by cannabinoid CCR1, 2), thus providing a mechanistic intervention is mostly attributed to the explanation for its role in myeloid progenitor engagement of neuronal CB receptors (4, 14), 1 recruitment during neuroinflamation. These while CB receptor expression by infiltrating T- 2 findings demonstrate the protective role of cells and monocytes is involved in the control of CB receptors in EAE pathology; provide neuroinflammation (4). Likewise, CB receptors 2 2 evidence for a new site of CB receptor action, are functional in microglial cells, in which they 2 namely the targeting of myeloid progenitor inhibit the production of proinflammatory 1 Copyright 2008 by The American Society for Biochemistry and Molecular Biology, Inc. cytokines and oxygen and nitrogen reactive Animal procedures and EAE induction- Animal species (12, 15), and their expression is up- procedures were performed according to the regulated upon cell activation (16). The European Union guidelines (86/609/EU) for the involvement of microglial cell activation in the use of laboratory animals. Adult CB receptor- 2 evolution and progression of different deficient mice (8-week old) (24) and their neurological disorders (MS, stroke, Alzheimer’s respective wild-type littermates, housed in a disease, amyotrophic lateral sclerosis, HIV- temperature-controlled room with a fixed 12 h associated dementia) has been put forward (17, light/12 h dark cycle, were obtained from 18), and it is becoming evident that, besides CNS heterozygote crosses. For EAE induction, mice resident microglial cells, bone marrow-derived were immunized at days 0 and 7 by cells infiltrate the inflamed CNS and subcutaneous injection of 150 μg myelin differentiate into functional microglia (19-21). oligodendrocyte glycoprotein (MOG) 35–55 Among those cells, CD34+ myeloid progenitors peptide emulsified in complete Freund’s expand in the blood of EAE mice, target the adjuvant, as previously described (25). In inflamed CNS and display differentiation addition, at days 0 and 2 mice were administered potential towards microglia (22). As modulation i.p. with 500 ng pertussis toxin (Sigma, St. of cannabinoid signaling may result in unwanted Louis, MO, USA). Clinical scores were psychoactive actions mediated by neuronal CB monitored using the following scale: 0, lack of 1 receptors, the use of CB -selective agonists is an clinical signs; 1, tail weakness; 2, hind limb 2 D attractive therapeutic possibility (15, 23). The paraparesis, hemiparesis or ataxia; 3, hind limb o w present study was therefore aimed at elucidating paralysis or hemiparalysis; 4, complete paralysis; nlo a the role and action mechanism of the CB2 5, moribund; 6, death. In pharmacological ded cannabinoid receptor in EAE. In particular, we experiments HU-308 (15 mg/kg) was fro m focused on the trafficking of myeloid progenitors administered i.p. starting at the day of maximal h towards the inflamed spinal cord and their score and thereafter daily until sacrifice. Control ttp contribution to microglial activation during EAE animals received the corresponding vehicle ://w w pathogenesis. injections (100 μl PBS supplemented with 0.5 w.jb mg/ml defatted bovine serum albumin and 4 % c.o EXPERIMENTAL PROCEDURES dimethylsulfoxide). Each clinical score value brg/ was obtained from a representative experiment y g u Materials- The following materials were kindly of 4 or 3 independent experiments aimed at es donated: CB2 receptor knock-out mice by Nancy elucidating the effect of CB2 receptor genetic t on D Buckley (National Institute of Health, Bethesda, ablation or HU-308 administration, respectively. ec e MD, USA) and HU-308 by Pharmos (Rehovot, Cell transfer experiments were performed by mb e Israel). The following antibodies were used: intracardiac injection of 5 x 106 bone marrow r 3 1 polyclonal anti-200 kDa neurofilament heavy cells derived from healthy C57BL/6 (ACTβ- , 2 0 protein and anti-human CD11b (clone M1/70) EGFP) mice into EAE-induced WT and CB -/- 18 2 from Abcam (Cambridge, UK); monoclonal rat syngenic mice at day 9 before symptom anti-mouse CD45R/B220 (clone RA3–6B2) appearance. Recipient mice were sacrificed and FITC conjugated antibody, rat anti-mouse analyzed 10 days after engraftment. CD11b (clone M1/70) and anti-CD4 from Magnetic resonance imaging was Becton Dickinson PharMingen (San Diego, CA, performed in EAE mice (n=4 each group) the USA); monoclonal anti-mouse CD34-biotin day before sacrifice at the Nuclear Magnetic (clone RAM34) from E-bioscience (San Diego, Resonance Center of Complutense University CA, USA); polyclonal anti-CB2 receptor from (Madrid, Spain). Anesthesized mice were placed Affinity Bioreagents (Golden, CO, USA); in a Biospec 47/40 (Bruker, Ettlingen, Germany) monoclonal anti-human CD45RB (clone operating at 4.7 Teslas, equipped with a 12 cm PD7/26) from Dako (CA, USA), polyclonal anti- gradient set and using a 4 cm radio frequency GFP from Invitrogen (Carlsbad, CA, USA) and surface coil. 3D T2-weighted spin-echo images rabbit monoclonal anti-Ki-67 (SP6) from were acquired using a fast spin-echo sequence. LabVision (Fremont, CA). Macrophage-colony The acquisition parameters were: TR=2226 ms, stimulating factor and Flt-3 were from effective TE=117 ms, FOV=2.5x1.6x1.0 cm3 and PreproTech Inc (London, UK). 2 averages. The acquisition matrix size was 256x128x32, which was zero-filled to get a reconstructed matrix size of 256x256x32. The 2 total acquisition time was 19 min. Diffusion CD34 and CD31 double immunostaining was water images delineated the area of routinely performed in order to confirm that neuroinflammation evidenced as hyperintense CD34 immunoreactivity included, in addition to signals. CD31+ vascular endothelial cells, cells with microglial features. Only CD34+ cells displaying Immunofluorescence and confocal microscopy- a rounded morphology (similar to ameboid EAE mice were sacrificed and isolated spinal microglia) or extending short processes (similar cords were frozen on dry ice. to reactive microglia) were counted. Both Immunofluorescence analysis (26) was perivascular and parenchymal CD34+ cells were performed on ethanol-fixed 14 µm-thick cryostat observed, and a significant fraction of those cells sections. Spinal cord sections were rinsed and did not express the endothelial marker CD31 and blocked for 30 min in phosphate-buffered saline co-expressed the microglial marker CD11b. (PBS) supplemented with 10% goat serum and Results are given as mean cell number per mm2. 4% bovine serum albumin and, after washing, In vitro immunofluorescence was performed incubated with the indicated primary antibodies. with the indicated primary antibodies and the Secondary antibody incubation (1 h at room corresponding secondary fluorescent antibodies temperature) was performed with the appropriate (as above). Immunofluorescence controls were mouse, rat and rabbit highly cross-adsorbed routinely performed with incubations in which AlexaFluor 488, AlexaFluor 594, AlexaFluor primary antibodies were not included. D 647 secondary antibodies, or with Immunofluorescence images shown in o w AlexaFluor488 or AlexaFluor594 streptavidin Figs. 2 and 3 were obtained with a fluorescence nlo a conjugates (Invitrogen). Washed sections were microscope (Zeiss Axioplan II) and acquired de d incubated with Hoechst 33342 (5 μg/ml) in PBS with a CDD camera (F-View II; Soft Imaging fro m prior to mounting. System). Images in Figs. 4-6 and 9 were h The specificity of CB receptor examined using Leica TCS-SP2 software ttp immunoreactivity was corroborated2 by using (Wetzlar, Germany) and SP2 microscope with 2 ://w w CB2-/- mouse sections, in which no passes by Kalman filter and a 1024X1024 w.jb immunoreactivity was observed, and allowed to collection box. c.o adjust optimal confocal microscope settings. CB2 brg/ receptor expression was analyzed with anti-CB2 Cell culture and flow cytometry analysis- Bone y gu receptor antibody together with anti-CD11b and marrow cells were cultured as described (22, 27) es anti-CD34-biotinilated antibodies (overnight with slight modifications. Briefly, 8 week-old t on D incubation at 4ºC) followed by secondary female mice were sacrificed and bone marrows e c e staining for rabbit and mouse IgGs with highly harvested by flushing out tibiae and femurs and m b e cross-adsorbed AlexaFluor 647, AlexaFluor 594 cultured in Iscove’s modified Dulbecco’s r 3 1 and streptavidin-AlexaFluor 488 secondary medium supplemented with 4% N2 (Invitrogen), , 2 0 antibodies, respectively. In addition, triple 15 ng/ml of Flt-3 and M-CSF and penicillin and 18 immunostaining was performed with a streptomycin. Six days after plating non-adherent combination of anti-CD34, CD45R/B220 and cells were harvested and, after washing, CB primary antibodies and their appropriate characterized by flow cytometry or replated for 2 secondary fluorescent antibodies. All the generation of microglial-like cells. Myeloid immunofluorescence data were obtained in a progenitors were differentiated for 5 days in blinded manner by two independent observers in bone marrow culture medium (described above) a minimum of 5-7 adjacent slices of two supplemented with one volume (50%) of glial different samples from the lumbo-thoracic spinal cell conditioned medium. Prior or after cord of the same animal. To determine the loss differentiation adherent cells were either fixed of axonal surface and the extent of microglial for immunofluorescence, frozen for RNA activation, neurofilament H and CD11b staining extraction or detached for FACS analysis. was quantified with Metamorph-Offline software Peripheral blood mononuclear cells (Universal Imaging, Downingtown, PA) and (PBMCs) were isolated by density gradient referred to the total white matter area of spinal centrifugation on Ficoll (Histopaque 1077; cord sections. T-cells, CD34+ cells and bone Sigma) from blood samples obtained from the marrow GFP infiltration were quantified by cave vein. Cells were washed with PBS and immunoreactive positive cell counting and resuspended in PBS supplemented with 2% goat results are given as mean cell number per mm2. calf serum for flow cytometry analysis or frozen 3 for RNA analysis. Flow cytometry was RESULTS performed with 0.5 x 106 cells per condition fixed in 1% paraformaldehyde at 4ºC. Antibodies CB cannabinoid receptor deficiency 2 and their corresponding controls were incubated exacerbates EAE pathogenesis- Wild-type mice for 30 min at 4ºC in 2% goat serum-PBS and, immunized by MOG injection developed EAE after washing, samples were subjected to with the appearance of symptoms starting at day secondary antibody incubation. Ten thousand 12.4 ± 4.2 and, after reaching a maximal stage at cells per recording were analyzed using a day 20.2 ± 2.7 (Fig. 1A), entered a chronic FACSCalibur flow cytometer. clinical phase (22, 25) The involvement of the CB receptor in the appearance of EAE 2 mRNA detection and quantification- RNA was symptoms was investigated by comparing wild- obtained with the RNeasy Protect kit (Qiagen, type and CB -/- littermates, which showed that the 2 Valencia, CA,USA) using the RNase-free DNase latter developed a notably higher symptomatic kit. cDNA was subsequently obtained using the EAE score (Fig. 1A, B). The day before sacrifice Superscript First-Strand cDNA synthesis kit magnetic resonance imaging analysis was (Roche, Welwyn Garden City, UK) and performed, confirming the existence of an amplified with the primers indicated in exacerbated neuroinflammatory process in the supplemental Table 1A. CB and CB PCR dorsal spinal cords of CB -/- animals as 1 2 2 amplifications were performed using the evidenced by the bright signal of low water D following conditions: 93°C for 1 min, 2 rounds diffusion areas (Fig. 1C). Spinal cord lesions o w (30s at 59°C, 1 min at 72°C and 30 s at 93ºC), 2 were characterized in further detail by nlo a rounds (30 s at 57°C, 1 min at 72°C and 30 s at histological analysis. Quantification of de d 93ºC) and 35 cycles (30 s at 55°C, 1 min at 72°C neurofilament H immunofluorescence showed an fro m and 30 s at 93ºC). Finally, after a final extension increased axonal loss (Fig. 2A) together with h step at 72°C for 5 min, PCR products were extended microglial-cell (CD11b+) activation ttp separated on 1.5% agarose gels. Real-time and T-lymphocyte (CD4+) infiltration in CB - ://w 2 w quantitative PCR was performed with Universal deficient mice (Fig. 2B, C), thus providing w.jb probe system (Roche, Basel, Switzerland) using further evidence for the exacerbated phenotype c.o the primers indicated in supplemental Table 1B. of these animals. As CB2 receptors are highly brg/ Amplifications were run in a 7900-HT Fast Real- expressed in bone marrow immune cells (13, y g u time PCR system and obtained values adjusted 15), and myeloid progenitor cells can be es using 18S RNA levels as reference. recruited into the neuroinflamed CNS (19-22), t on D we analyzed the presence of cells expressing e c e Multiple sclerosis human tissue samples- Tissue CD34, a marker for primitive myeloid m b e samples were supplied by the UK Multiple progenitors. Of interest, an increased number of r 3 Sclerosis Tissue Bank, funded by the Multiple CD34+ cells was evident in CB2-deficient spinal 1, 20 Sclerosis Society of Great Britain and Northern cords (Fig. 2D). Increased proliferation (Ki-67+ 18 Ireland, Registered Charity 207495. Cortical cells) of CD11b and CD34 cell populations was brain samples fixed in formalin, embedded and also observed in CB knock-out mice (Fig. 2E). 2 cut in 4 μm-thick sections were employed for immunohistochemical study as previously Selective CB cannabinoid receptor activation 2 described (28). Briefly, sections were palliates EAE symptoms and pathogenesis- As deparaffinized and, after washing, subjected to CB receptor genetic ablation exacerbates EAE 2 antigen retrieval procedure. Tissue samples were pathogenesis, we assessed the impact of CB 2 incubated with mouse anti-human CD45RB and receptor selective activation on disease rat anti-human CD11b monoclonal antibodies progression. Daily injections of the CB - 2 together with rabbit polyclonal anti-CB receptor selective agonist HU-308 (29) were performed 2 antibody. starting at the day of maximal disease score. In agreement with the observations from CB - 2 Statistical analysis- Results shown represent the deficient mice, HU-308 treatment improved EAE means ± SEM of the number of experiments score and the evolution of the disease when indicated in every case. Statistical analysis was compared to vehicle administration (Fig. 3A). performed by ANOVA. A post hoc analysis was The specificity of HU-308 action was confirmed made by the Student-Neuman-Keuls test. In vivo by the use of CB -/- mice, in which this agonist 2 data were analyzed by an unpaired Student t-test. was unable to decrease EAE score (data not 4 shown). The analysis of HU-308 administration Microglial-like cell differentiation reduced the on tissue histology revealed a strong reduction of expression of CD34, whereas increased spinal cord infiltrates (Fig. 3B) that correlated expression of CD11b and CD11c were observed. with reduced axonal loss (Fig. 3C, left panel), CB receptor transcripts were highly expressed 2 microglial activation (Fig. 3C, middle panel), in myeloid progenitors and were also evident in and myeloid CD34+ cell infiltration (Fig. 3C, differentiated microglial-like cells (Fig. 6B, D). right panel). Of importance, HU-308 In contrast, CB receptor mRNA was hardly 1 administration decreased microglial and detectable in myeloid progenitors and was below infiltrating myeloid cell proliferation (Fig. 3D). detection limits after differentiation. The eCB degrading enzymes showed opposite patterns of Expression of the eCB system in myeloid expression with decreased FAAH and up- progenitor cells- In order to confirm if myeloid regulated MAGL transcripts upon differentiation progenitors with the ability to generate (Fig. 6D, right panel). microglial cells may be directly targeted by CB 2 receptor activation we examined spinal cord The CB cannabinoid receptor controls bone 2 sections of EAE mice. Confocal microscopy marrow-derived myeloid cell trafficking- The confirmed the expression of CB receptors in mechanism of CB receptor regulation on 2 2 CD34+ myeloid progenitor cells (Fig. 4A). myeloid cell recruitment and microglial Importantly, these cells were also positive for replenishment during EAE pathogenesis was D CD11b, supporting that recruited cells are prone further investigated. Bone marrow cells from o w to microglial differentiation (19-22). In addition, EAE mice were obtained at the end of the nlo a CB2 receptors were expressed in CD11b+ cells experiment and analyzed by flow cytometry. The ded that co-express CD45R/B220 (Fig. 4A), a marker number of CD11b+ cells was not altered in CB2-/- fro m that has been shown to identify myeloid mice when compared to wild-type mice (Fig. 7A, h progenitors with the potential to differentiate to upper panel). In contrast, CD34+ cells were ttp microglial cells (22). Real-time PCR increased in mice lacking the CB receptor (Fig. ://w 2 w quantification analysis revealed that in EAE 7A, lower panel), while the opposite was w.jb mice transcript levels of CD34, CD45R/B220, observed upon HU-308 administration (CD34+ c.o CD11b and CD11c were increased in spinal cord cells 2.4 ± 0.2% and 1.4 ± 0.3% in vehicle- and brg/ extracts (Fig. 4B). The eCB system elements HU-308-treated mice, respectively; P<0.01). y g u were also up-regulated during EAE and the Flow cytometry (Fig. 7B) and real-time PCR es maximal induction was observed for CB quantification analysis (Fig. 7C) of PBMCs t on 2 D receptor expression, which was accompanied by obtained from the same EAE mice supported e c e increased expression of the CB receptor and the that, in parallel with bone marrow cell profiles, m 1 b e degrading enzymes fatty acid amide hydrolase CD34+ but not CD11b+ cells were expanded into r 3 1 (FAAH) and monoacylglycerol lipase (MAGL; circulating blood cells in mice lacking the CB2 , 20 Fig. 4C). Next, we analyzed the expression of receptor. These findings are in line with the 18 CB receptors in brain sections of MS patients. aforementioned higher number of CD34+ and 2 CB receptors were expressed in microglial cells CD11b+ cells found in spinal cords of CB -/- 2 2 of plaques located in the vicinity of blood vessels mice, suggesting that the CB receptor can 2 (28), and importantly a subpopulation of these control myeloid progenitor cell trafficking cells were shown to co-express the myeloid towards neuroinflamed tissue. marker CD45RB (Fig. 5). An analysis of the in vivo expression of Finally, we verified the expression of chemokine ligands and receptors was performed CB receptors in myeloid progenitors from bone in EAE spinal cord and bone marrow. CB 2 2 marrow-derived cultures. Immunofluorescence receptor ablation upregulated chemokines and analysis showed the co-expression of CB receptors known to be important in microglial 2 receptors with CD34 and CD45R/B220 (Fig. recruitment to inflammatory lesions (30). Thus, 6A). CB receptors were also present in CCL2, CCL3 and CCL5 transcript levels were 2 committed cells that express CD11b together increased, and similarly their principal receptors with CD45R/B220 or CD34 (Fig. 6A). Gene CCR2 and CCR1 were also induced in bone expression analysis by RT-PCR (Fig. 6B) and marrow (Fig. 8A, B). On the other hand, HU-308 quantitative PCR (Fig. 6C, D left panel) of administration resulted in an overall reduction of myeloid progenitor cell cultures confirmed the these chemoattractant ligands and receptors (Fig. expression of CD34, CD45R/B220 and CD11b. 8C, D). 5 Increased bone marrow-derived cell recruitment pathology is still a matter of debate. in spinal cords of CB receptor-deficient EAE Infiltrating T-cells and resident microglia 2 mice- In order to confirm the involvement of the make a major contribution to the ethiopathology CB receptor in microglial replenishment from of neuroinflammation and neurodegeneration in 2 myeloid progenitors, bone marrow GFP-labeled MS patients, as well as in animal models of the cells derived from healthy EGFP transgenic mice disease (17). In addition, recent research has were transferred into EAE-induced wild-type shown that brain resident microglia can be and CB -/- mice before symptom appearance replenished by grafted bone marrow-derived 2 (Fig. 9A). At the end of the experiment, flow myeloid progenitors (19-21, 34). In particular, cytometry analysis revealed the presence of a under neuroinflammatory conditions such as significant population of grafted cells in the EAE, bone marrow-derived CD34+ myeloid PBMC fraction obtained from the blood of progenitors mobilize through the blood and recipient animals (9.6 ± 1.1%), of which the target the inflamed brain, allowing the majority (96.1 ± 2.7%) co-expressed CD11b (Fig recruitment of new microglial cells (18, 22). By 9B, upper panel). CB -/- mice showed an using various experimental approaches, here we 2 increased number of total circulating CD34+ demonstrate that CB receptors play an important 2 cells, which was also evident within the GFP+ role in the control of EAE pathology and provide population (Fig. 9B, lower panel). Spinal cord evidence for the mechanism of CB action, 2 sections were also analyzed and bone marrow- namely the targeting of myeloid progenitor cells. D derived GFP+ cells were observed (Fig. 9C). Thus, the absence of CB receptors significantly o 2 w Quantification of GFP+ cells showed a higher exacerbates EAE symptoms and myeloid cell nlo a number of infiltrating myeloid cells in CB2-/- recruitment into the inflamed CNS, while the ded mice than in wild-type littermates (Fig. 9C). opposite is observed upon CB2-selective agonist fro m Further immunofluorescence analysis evidenced administration. Control of microglial recruitment h that transferred GFP+ cells constituted by CB receptors involves the regulation of ttp differentiated microglial infiltrates in the spinal importa2nt mediators of cell trafficking such as ://w w cord as they expressed the CD11b marker (91.2 chemokines and their receptors. Genetic and w.jb ± 3.6 %), while in some GFP+ cells (20.2 ± 3.6 pharmacological models of CB2 receptor c.o %) CD34 expression was still evident (Fig. 9D). manipulation elicit changes in the expression of brg/ Moreover, the fraction of GFP+CD11b+ cells chemokines that promote myeloid cell y g u within the total CD11b+ microglial cell pool was recruitment to neuroinflammatory lesions such es elevated in CB -/- mice (Fig. 9E). In summary, as CCL2, CCL3 and CCL5. In addition, t on 2 D these results support the involvement of CB expression of their receptors CCR2 and CCR1 e 2 ce receptors in bone marrow-derived myeloid cell was also affected. Our results are in line with the m b e recruitment towards neuroinflamed tissue. role of the CCL2-CCR2 axis in microglial r 3 1 replenishment derived from inflammatory , 2 0 DISCUSSION monocytes (35, 36). Likewise, CCR2 is required 18 for bone marrow-derived microglial recruitment Alterations of the eCB system have been and the development of neuropathic pain (37). In implicated in the pathogenesis of several this context, stimulation of human monocytes neurodegenerative disorders, and activation of with CB -selective agonists has been shown to 2 cannabinoid receptors exerts neuroprotection in inhibit CCL2- and CCL3-mediated chemotaxis various models of brain damage including by cross-desensitization of CCR2 and CCR1 excitotoxicity, traumatic brain injury and stroke (38). (1, 31). Moreover, cannabinoid administration CB receptors participate in the control 2 elicits an improvement of symptoms of different of cell proliferation, survival and differentiation neuroinflammatory situations including EAE (2, fate decisions (15). Thus, CB receptor activation 2 12). Genetic and pharmacological studies controls hematopoietic and neural progenitor cell support that neuronal CB receptor expression is proliferation and differentiation. An inverse 1 required for cannabinoid-mediated suppression relation between CB receptor expression and 2 of EAE symptoms (4, 14, 32). In contrast, other stage of cell differentiation has been shown in reports have pointed to the involvement of both neural cells (from neural progenitors to mature CB and CB receptors in the beneficial effects neurons and neuroglial cells) (39) and B-cells 1 2 of cannabinoids in MS models (4, 7, 9, 11, 33). (from virgin B-cells to centroblasts) (40), and Thus, the precise role of CB receptors in EAE CB receptor activation and overexpression has 2 2 6 been reported to block neutrophil cell bone-marrow-derived myeloid cell recruitment differentiation (41). It is therefore conceivable (present report) and microglial activation (11). that changes in cell proliferation, as observed in The preclinical studies evidencing the our study, and differentiation may also ability of cannabinoids to manage EAE contribute to CB receptor-regulated myeloid symptoms have fostered the investigation for 2 progenitor trafficking. their potential translation to the clinic (2, 45-46). Our data agree with the current notion Beneficial cannabinoid actions in MS patients, that microglial cells express CB receptors (42), supported by large-scale phase III clinical trials, 2 which are strongly up-regulated in animal include alleviation of spasticity and tremor, models of neuroinflammation (16) and in neuropathic pain and nocturia. Nonetheless, plaques of MS patients (28, 43). Microglial cells therapies for MS management should be able to synthesize and degrade eCB ligands such as 2- prevent not only those symptoms but also arachidonoylglycerol (12), the levels of which neuroinflammation, demyelination, axonal loss are negatively controlled by the IFN-γ released and neurodegeneration in order to exert a by primed T-cells invading the CNS during clinically relevant impact in the secondary phase EAE, thus indicating that impaired 2- of the disease (2, 17). In this context, arachidonoylglycerol production may be cannabinoids constitute a very attractive associated with neurodegeneration in EAE (44). possibility for therapeutic intervention as they In addition, cannabinoids down-regulate the are neuroprotective (31), prevent demyelination D production of proinflammatory cytokines (9) and exert a wide array of anti-inflammatory o w (mostly IL-1β, IL-6 and TNFα by microglial actions (12, 13). The use of selective CB2 nlo a cells, and IL-2, IFN-γ and GM-CSF by receptor agonists for the treatment of MS, and de d autoreactive T cells), as well as of nitrogen and perhaps of other neuroinflammatory conditions, fro m oxygen reactive species (6, 12, 13). Overall constitutes therefore an attractive possibility h eCBs via CB receptors appear to play a key owing to the selective role of this cannabinoid ttp neuroimmunom2 odulatory role in EAE not only receptor type in immune regulation and to the ://w w by preventing T-cell-mediated absence of marijuana-like psychoactive effects w.jb neurodegeneration (2, 4), but also by inhibiting associated to its activation (15, 23). c.o brg/ y g u e s t o n D e c e m b e r 3 1 , 2 0 1 8 7 REFERENCES 1. Piomelli, D. (2003) Nat. Rev. Neurosci. 4, 873-884 2. Pryce, G., and Baker, D. (2005) Trends Neurosci. 28, 272-276 3. Cabranes. A., Venderova, K., de Lago, E., Fezza, F., Sanchez, A., Mestre, L., Valenti, M., Garcia- Merino, A., Ramos, J.A., Di Marzo, V., and Fernandez-Ruiz, J. (2005) Neurobiol. Dis. 20, 207-217 4. Maresz, K., Pryce, G., Ponomarev, E.D., Marsicano, G., Croxford, J.L., Shriver, L.P., Ledent, C., Cheng, X., Carrier, E.J., Mann, M.K., Giovannoni, G., Pertwee, R.G., Yamamura, T., Buckley, N.E., Hillard, C.J., Lutz, B., Baker, D., and Dittel, B.N. (2007) Nat. Med. 13, 492-497 5. Baker, D., Pryce, G., Croxford, J.L., Brown, P., Pertwee, R.G., Huffman, J.W., and Layward, L. (2000) Nature 404, 84-87 6. Croxford, J.L., and Miller, S.D. (2003) J. Clin. Invest. 111, 1231-1240 7. Ortega-Gutierrez, S., Molina-Holgado, E., Arevalo-Martin, A., Correa, F., Viso, A., Lopez- Rodriguez, M.L., Di Marzo, V., and Guaza, C. (2005) FASEB J. 19, 1338-1440 8. Molina-Holgado, E., Vela, J.M., Arevalo-Martin, A., Almazan, G., Molina-Holgado, F., Borrell, J., and Guaza, C. (2002) J. Neurosci. 22, 9742-9753 9. Arévalo-Martin, A., Vela, J.M., Molina-Holgado, E., Borrell, J., and Guaza, C. (2003) J. Neurosci. 23, 2511-2516 10. Jackson, S.J., Pryce, G., Diemel, L.T., Cuzner, M.L., and Baker, D. (2005) Neuroscience 134, 261- D 268 o w 11. Eljaschewitsch, E., Witting, A., Mawrin, C., Lee, T., Schmidt, P.M., Wolf, S., Hoertnagl, H., nlo a Raine, C.S., Schneider-Stock, R., Nitsch, R., and Ullrich, O. (2006) Neuron 49, 67-79 de d 12. Walter, L., and Stella, N. (2004) Br. J. Pharmacol. 141, 775-785 fro m 13. Klein, T.W. (2005) Nat. Rev. Immunol. 5, 400-411 h 14. Pryce, G., and Baker, D. (2007) Br. J. Pharmacol. 150, 519-525 ttp 15. Fernández-Ruiz, J., Romero, J., Velasco, G., Tolón, R.M., Ramos, J.A., and Guzmán, M. (2007) ://w w Trends Pharmacol. Sci. 28, 39-45 w.jb 16. Maresz, K., Carrier, E.J., Ponomarev, E.D., Hillard, C.J., and Dittel, B.M.(2005) J. Neurochem. c.o 95, 437-445 brg/ 17. Hauser, S.L., and Oksenberg, J.R. (2006) Neuron 52, 61-76 y g u 18. Davoust, N., Vuaillat, C., Androdias, G., and Nataf, S. (2008) Trends Immunol. (in press) es 19. Vallieres, L., and Sawchenko, P.E. (2003) J. Neurosci. 23, 5197-5207 t on D 20. Simard, A.R., and Rivest, S. (2004) FASEB J. 18, 998-1000 e c e 21. Djukic, M., Mildner, A., Schmidt, H., Czesnik, D., Brück, W., Priller, J., Nau, R., and Prinz, M. m b e (2006) Brain 29, 2394-2403 r 3 1 22. Davoust, N., Vuaillat, C., Cavillon, G., Domenget, C., Hatterer, E., Bernard, A., Dumontel, C., , 2 0 Jurdic, P., Malcus, M., Confavreux, C., Belin, M.F., and Nataf, S. (2006) FASEB J. 20. 2081-2092 18 23. Mackie, K. (2006) Annu. Rev. Pharmacol. Toxicol. 46, 101-122 24. Buckley, N.E., McCoy, K.L., Mezey, E., Bonner, T., Felder, C.C., Glass, M., and Zimmer, A. (2000) Eur. J. Pharmacol. 396, 141-149 25. Nataf, S., Carroll, S.L., Wetsel, R.A., Szalai, A.J., and Barnum, S.R. (2000) J. Immunol. 165, 5867-5873 26. Aguado, T., Palazuelos, J., Monory, K., Stella, N., Cravatt, B., Lutz, B., Marsicano, G., Kokaia, Z., Guzmán, M., Galve-Roperh, I. (2006) J. Neurosci. 26, 1551-1561 27. Servet-Delprat, C., Arnaud, S., Jurdic, P., Nataf, S., Grassetn, M.F., Soulas, C., Domenget, C., Destaing, O., Rivollier, A., Perret, M., Dumontel, C., Hanau, D., Gilmore, G.L., Belin, M.F., Rabourdin-Combe, C., and Mouchiroud, G. (2002) BMC Immunol. 3, 15 28. Benito, C., Romero, J.P., Tolon, R.M., Clemente, D., Docagne, F., Hillard, C.J., Guaza, C., and Romero, J. (2007) J. Neurosci. 27, 2396-2402 29. Hanus, L., Breuer, A., Tchilibon, S., Shiloah, S., Goldenberg, D., Horowitz, M., Pertwee, R.G., Ross, R.A., Mechoulam, R., and Fride, E. (1999) Proc. Natl. Acad. Sci. USA 96, 14228-14233 30. Hanisch, U., and Kettenman, H. (2007) Nat. Neurosci. 10, 1387-1394 31. Mechoulam, R., Spatz, M., and Shohami, E. (2002) Sci. STKE 129:RE5 32. Croxford, J.L, Pryce, G., Jackson, S.J., Ledent, C., Giovannoni, G., Pertwee, R.G., Yamamura, T., and Baker, D. (2007) J. Neuroimmunol. 93, 120-129 8 33. Docagne, F., Muneton, V., Clemente, D., Ali, C., Loria, F., Correa, F., Hernangomez, M., Mestre, L., Vivien, D., and Guaza, C. (2007) Mol. Cell. Neurosci. 34, 551-561 34. Asheuer, M., Pflumio, F., Benhamida, S., Dubart-Kupperschmitt, A., Fouquet, F., Imai, Y., Aubourg, P., and Cartier, N. (2004) Proc. Natl. Acad. Sci. USA 101, 3557–3562 35. Mildner, A., Schmidt, H., Nitsche, M., Merkler, D., Hanisch, U.K., Mack, M., Heikenwalder, M., Brück, W., Priller, J., and Prinz, M. (2007) Nat. Neurosci. 10, 1544-1553 36. Babcock, A.A., Kuziel, W.A., Rivest, S., and Owens, T. (2003) J. Neurosci. 23, 7922-7930 37. Zhang, J, Shi, X.Q., Echeverry, S., Mogil, J.S., De Koninck, Y., and Rivest, S. (2007) J. Neurosci. 27, 12396-12406 38. Montecucco, F., Burger, F., Mach, F., and Steffens, S. (2008) Am. J. Physiol. Circ Physiol. [Epub ahead of print] 39. Palazuelos, J., Aguado, T., Egia, A., Mechoulam, R., Guzmán, M., and Galve-Roperh, I. (2006) FASEB J. 20, 2405-2407 40. Carayon, P., Marchand, J., Dussossoy, D., Derocq, J.M., Jbilo, O., Bord, A., Bouaboula, M., Galiegue, S., Mondiere, P., Penarier, G., Fur, G.L., Defrance, T., and Casellas, P. (1998) Blood 92, 3605-3615 41. Alberich, Jordá., M., Rayman, N., Tas, M., Verbakel, S.E., Battista, N., van Lom, K., Löwenberg, B., Maccarrone, M., and Delwel, R. (2004) Blood 104, 526-534 42. Carrier, E.J., Querrán, C.S., Barkmeier, A.J., Breese, N.M., Yang, W., Nithipatikom, K., Pfister, D S.L., and Campbell, W.B. (2004) Mol. Pharmacol. 65, 999-1007 o w 43. Yiangou, Y., Facer, P., Durrenberger, P., Chessell, I.P., Naylor, A., Bountra, C., Banati, R.R., and nlo a Anand, P. (2006) BMC Neurol. 6, 12 de d 44. Witting, A., Chen, L., Cudaback, E., Straiker, A., Walter, L., Rickman, B., Moller, T., Brosnan, C., fro m and Stella, N. (2006) Proc. Natl. Acad. Sci. USA 103, 6362-6367 h 45. Zajicek, J.P., Sanders, H.P., Wright, D.E., Vickery, P.J., Ingram, W.M., Reilly, S.M., Nunn, A.J., ttp Teare, L.J., Fox, P.J., and Thompson, A.J. (2005) J. Neurol. Neurosurg. Psychiatry 76, 1664-1669 ://w w 46. Wade, D.T., Makela, P.M., House, H., Bateman, C., and Robson, P. (2006) Mult. Scler. 12, 639- w.jb 645 c.o 47. Collin, C., Davies, P., Mutiboko, I.K., and Ratcliffe, S. (2007) Eur. J. Neurol. 14, 290-296 brg/ y g u es t on D FOOTNOTES e c e m b e *Equally contributing authors r 3 1, 2 0 #Present address: INSERM U851, IFR 128 BioSciences Lyon Gerland, Lyon, F-69007, France 18 *Address correspondence to: Ismael Galve-Roperh, Department of Biochemistry and Molecular Biology I, School of Biology, Complutense University, 28040 Madrid, Spain Telephone: +34 913944668; Fax: +34 913944672; E-mail: [email protected] Acknowledgements- We are indebted to our lab colleagues for their support and encouragement, to I. del Valle and A. Sánchez for fruitful scientific discussions and collaboration, to A. Egia and E. Resel for technical support, and to M. Fernández and D. Castejón for excellent assistance in MRI imaging. J.P., B.J. and T.A. were supported by Ministerio de Educación y Ciencia (FPI program; Spain), Fondation pour Recherche Medicale (France) and Comunidad Autónoma de Madrid (Spain), respectively. Research in our laboratories was financially supported by Picasso Program (HF2005- 0017), Comunidad Autónoma de Madrid (S-SAL/0261/2006 and 950344), Santander Complutense (PR27/05-13988), Fundación de Investigación Médica Mutua Madrileña Automovilística, Ministerio de Educacion y Ciencia (SAF2004/00237), French Embassy in Spain and ARSEP. 9 FIGURE LEGENDS FIGURE 1. EAE is exacerbated in CB -deficient mice. A, EAE score was determined daily after 2 induction of the disease in wild-type (WT; open circles) and CB -/- littermates (closed circles; n=8 each 2 group). Scores were compared between the two groups day by day. * P< 0.05 versus WT from day 17 on. B, The mean values of the day of clinical onset, maximum score, and mean score from symptom appearance are shown. * P< 0.05 versus WT. C, Representative MRI images at day 24 of spinal cords from WT and CB -/- mice after EAE induction are shown. The bright signal of low water diffusion 2 areas corresponds to inflamed tissue. Scale bars: general axial projection (left panel) 2.0 mm; magnified axial projections (top panels) 1.3 mm; and magnified sagital projections (bottom panels) 1.7 mm. FIGURE 2. Characterization of spinal cord sections in CB -deficient mice during EAE. A, Axonal 2 loss was quantified (right panel) by neurofilament (NF) immunoreactivity in the white matter of spinal cord sections of wild-type (WT) and CB -/- mice at the end of the experiment. Representative 2 immunofluorescence images are shown (left panels). Phenotypic analysis of infiltrating cells in spinal cord sections was performed. Quantification of microglial activation was determined as the CD11b+ area referred to the total area examined (B). The number of infiltrating CD4+ (C) and CD34+ (D) cells per section were also determined and expressed as the mean cell number per mm2. E, Proliferating D cells were determined by double immunofluorescence with antibodies against Ki-67 and CD11b or ow n CD34. Scale bar 100 μm. ** P < 0.01 versus control. lo a d e d FIGURE 3. Administration of the CB2-selective agonist HU-308 improves EAE symptoms and from reduces spinal cord lesions and microglial activation. A, EAE score in wild-type (WT) mice treated h with vehicle (Veh; open circles) or HU-308 (closed triangles; 15 mg/kg daily; n=5 each group) at the ttp://w indicated time point (arrow). * P< 0.05 versus vehicle from day 24 on. The ratio EAE score at day 28 w w to EAE score at day 18 is shown in the right panel. ** P< 0.01 versus vehicle. B, Representative .jb c immunofluorescence images of neurofilament (NF) (red) and Hoechst 33342 nuclear infiltrates (blue) .o rg in vehicle- and HU-308-treated mice. Scale bar 100 μm. C, Quantification of axonal loss, CD11b+ and b/ y CD34+ cells in the spinal cords of the indicated groups at the end of the experiment. D, Proliferating gu e cells were determined by double immunofluorescence with antibodies against Ki-67 and CD11b or st o n CD34. ** P < 0.01 versus vehicle. D e ce m mFIiGcrUoRgElia l4 c. elElsx.p Are, sCsioonnf oocfa l tmheic rCosBco2 pcya nimnaagbeisn ooifd ErAeEce psptoinra l inc orsdp isneaclt iocnosr dw emrey oelbotiadin-deder iwvietdh ber 31 CD34 (green), CD11b (red) and CB2 receptor (blue) antibodies. The CB2 receptor was also present in , 201 CD45R/B220+ (green) and CD11b+ (red) cells. Total cell nuclei were counterstained with Hoechst 8 33342 (grey). The right column shows the merged images. Scale bars 50 μm and 10 μm in upper and middle panels respectively. B, Transcript levels of the indicated myeloid and microglial markers were analyzed in spinal cord extracts from wild-type mice, either healthy or EAE-induced (clinical score ranging from 2.5 to 3.5). C, Transcript levels of the indicated elements of the eCB system. Transcript levels were normalized to 18S RNA expression, referred to healthy mice levels and provided in arbitrary units (a.u.). ** P < 0.01 versus control. FIGURE 5. Expression of the CB cannabinoid receptor in CD45BR+ microglial cells in cortical 2 plaques of multiple sclerosis patients. A, Confocal microscopy analyses were performed by using CB receptor (green), CD45RB (red) and CD11b (blue) antibodies. Images show CB receptor 2 2 expression in CD45+ CD11b+ microglial cells (denoted by arrows) (1), CD11b+ CD45- microglial cells (2) and CD11b-CD45- cells (3) located in the vicinity of blood vessels. Cell nuclei were counterstained with Hoechst 33342 (grey). Scale bars 35 and 10 μm in upper and lower panels, respectively. FIGURE 6. The CB cannabinoid receptor is expressed in bone-marrow myeloid progenitors. A, 2 Immunofluorescence of cultured myeloid progenitors was performed with the indicated combinations of antibodies against CD34, CD45R/B220, CD11b and CB receptor. Total cell nuclei were 2 10