Primary Rat Microglial Cultures.

Primary mixed glial cultures were prepared from neonatal rat cortex as described previously (Ramírez et al., 2005). In brief, mechanically dissociated cortices were seeded onto 75-cm² flasks in Dulbecco’s modified Eagle’s medium (Lonza France Sàrl, Paris, France), supplemented with 10% fetal calf serum (FCS; Invitrogen, Carlsbad, CA) and 40 μg/ml gentamicin. Cells were cultured in a humidified atmosphere of 5% CO₂/95% air at 37°C, and the medium was changed the day after seeding and once every week afterward. When confluence was reached, after being cultured for 2 to 3 weeks, flasks were shaken for 2 to 3 h at 230 rpm at 37°C, and floating cells were pelleted and seeded onto poly(lysine)-coated 96-well plates in medium with 0.1% FCS. The cultures were at least 99% pure, as judged by immunocytochemical criteria. Drugs were added in one tenth of the final volume to maintain the aggregation of peptides.

Because of the poor yield of the microglial cultures, we took advantage of microglial cell lines N13 and BV-2 to construct concentration-response curves with the cannabinoids under study. Thereafter, concentrations that approximated the EC₅₀ were assayed in primary microglial cultures. Furthermore, the involvement of either CB₁ or CB₂ receptors in the microglial cell functions, difficult to study in vivo, was investigated.

BV-2 Microglial Cells.

The transformed microglial cell line (v-raf/v-mic) was obtained from the Interlab Cell Line Collection (National Institute for Cancer Research and Advanced Biotechnology Center, Geneva, Italy).The cells were grown in RPMI 1640 (Sigma-Aldrich) supplemented with 10% fetal calf serum and 2 mM glutamine, detached from the flasks by manual shaking, and seeded onto poly(l-lysine) (10 μg/ml)-coated plates.

N13 Microglial Cells.

In other experiments, the microglial cell line N13 was used (Righi et al., 1989). Cells were cultured in RPMI 1640 containing 10% FCS in F75 flasks, and the medium was changed every 2 or 3 days. At confluence, the cells were trypsinized (trypsin-EDTA; Sigma-Aldrich) and seeded onto poly(l-lysine)-coated plates in medium containing 0.1% FCS.

Measurement of [Ca²⁺]_i.

Microglial cells, either N13 cells or primary rat microglia, were plated in 96-well plates (90.000 cells/well in 100 μl of medium) in RPMI culture medium with 0.1% FCS for 1 day. After washing twice with Krebs’ solution (105 mM NaCl; 5 mM KCl; 10 mM HEPES-sodium; 5 mM NaHCO₃; 60 mM mannitol; 5 mM sucrose; 0.5 mM MgCl₂; and 1.3 mM CaCl₂, pH 7.4), the cells were loaded with 50 μl/well of 10 μM Fura-2 containing 0.2% Pluronic acid in Krebs’ solution for 30 min and 37°C. Antagonists were added to the dye solution at final concentration in the corresponding wells. Then, the wells were washed twice with Krebs’ solution, and 40 μl were added to each well, adding the antagonist to final concentration where appropriate. The intracellular calcium concentration was estimated by alternatively exciting with 340 and 380 nm and measuring the fluorescence emission at 510 nm in a fluorescence plate reader (Fluostar Optima; BMG Labtech GmbH, Offenburg, Germany) at 37°C. After 15-s reading, 10 μl of control (Krebs’ buffer) or a 5-fold concentrated agonist solution was added to each well by means of the injector system, and the fluorescence was measured for an additional 1 to 2 min. The ratio of emission levels at 340 and 380 nm was calculated at each time point. The conversion to intracellular calcium concentration was performed using calibrating solutions consisting in 5 μM ionomycin in Krebs’ buffer (maximal response) or 5 μM ionomycin plus 20 mM EGTA in Krebs’ buffer (minimal response). In addition, any background signal was subtracted as measured in nonloaded wells. The increase in [Ca²⁺]_i was expressed as a percentage of the peak level in comparison with the preinjection baseline according to the following formula: Δ[Ca²⁺]_i = [(peak − baseline)/baseline] × 100. In other experiments, data were normalized versus the ATP response. In preliminary experiments, it was shown that the reduction of extracellular calcium, without added calcium in the buffer and addition of EGTA 10 mM, decreased the intracellular calcium levels by 50%.

Migration Studies in Chemotaxis Chambers.

Chemotaxis through porous membranes was assessed according to Boyden with some modifications. N13 cells or primary microglial cells (90,000 cells) were seeded onto the upper compartment of inserts (6.5 mm diameter) with 8-μm porous polycarbonate membranes in 24-well plates (Transwell Costar 3422; Corning Life Sciences, Lowell, MA).

In preliminary experiments, it was confirmed that to obtain a migratory response, it was necessary to activate the cells by exposure to LPS for 24 h, in agreement with previous studies (Cui et al., 2002). Under these conditions, LPS dose-dependently (1500–6000 ng/ml) or the chemotactic peptide fMLP (25–200 nM) induced migration of N13 cells after 3 h of its addition to the cultures (data not shown). Cells were treated by the addition of LPS (3 μg/ml final concentration; from Escherichia coli 0127:B8, Difco Laboratories, Detroit, MI) in culture medium to both compartments. After 24 h, the medium was changed, and the treatments were added to the lower compartment. Three hours later, cells were fixed with 4% paraformaldehyde for 30 min and stained with Coomassie Brilliant Blue (0.2% in 10% acetic acid/40% methanol). The insert membrane was cut and mounted onto a microscope slide, and cells on the lower face of the filter were counted by phase-contrast microscopy by an observer unaware of the treatments (four fields per condition in triplicate) in an Axiovert Zeiss microscope at 400× magnification.

Nitrite Assay.

These experiments were performed with BV-2 cells. Cells were plated (50,000 cells/well in 100 μl of medium) onto 96-well precoated plates in RPMI culture medium containing 0.1% FCS. After 24 h, the culture cells were treated with LPS (300 ng/ml) alone or with the cannabinoids, and they were cultured for an additional 24 h. In preliminary studies, we determined the appropriate LPS concentration and incubation time. Nitrite oxide production was assessed by the colorimetric Griess reaction (Sigma-Aldrich), which detects nitrite (NO₂⁻), a stable reaction product of NO and molecular oxygen, in cell cultures supernatants. Eighty microliters of each sample was incubated with 80 μl of Griess reagent for 15 min, and absorbance was measured at 540 nm in a microplate reader. The nitrite concentration was determined from a sodium nitrite standard curve.

Aβ-Injected Mice.

All experiments were performed according to ethical regulations on the use and welfare of experimental animals of the European Union and the Spanish Ministry of Agriculture, and the procedures were approved by the bioethical committee of the Consejo Superior de Investigaciones Cientificas.

These animals were used as a partial AD model, which develops glial activation and cognitive deficit in learning a spatial task (Ramírez et al., 2005). C57/Bl6 mice of 3 months of age were intraventricularly injected with 2.5 μg of fibrillar Aβ or saline (5 μl). The Hamilton syringe used for intraventricular injections was repeatedly washed with distilled water followed by flushing with 1 mg/ml bovine serum albumin solution, which reduces drastically binding to glass. This procedure was performed before every injection. The next day, the intraperitoneal treatment with the cannabinoids (20 mg/kg CBD; 0.5 mg/kg HU-308, JWH, and WIN) was initiated. During the first week, the mice were treated daily, then for 2 weeks, they were treated 3 days/week. Performance in the Morris water maze was conducted at the same time of the day (9:00 AM to 2:00 PM). To determine spatial learning, rats were trained to find a hidden platform in a water tank 100 cm in diameter. Four trials per day with different start positions, each 30 min apart, were conducted for 5 days (Ramírez et al., 2005), and latency to reach the platform was recorded. Cutoff time to find the platform was 60 s, and mice failing to find the platform were placed on it and left there for 15 s. Data acquisition was performed with a video camera (Noldus Information Technology, Wageningen, the Netherlands). The animals were sacrificed 18 days after the Aβ injection, and their brains were dissected, frozen, and stored at −80°C until assayed.

Expression of CB₁ and CB₂ in Microglial Cell Line N13.

N13 cells have been immortalized from primary mice microglial cell cultures and shown to express microglial markers and release several cytokines upon LPS stimulation while being capable of FcR-mediated phagocytosis (Righi et al., 1989). Previous evidence has shown that primary microglia and BV-2 microglial cells express both CB₁ and CB₂ (Walter et al., 2003; Ramírez et al., 2005). Accordingly, the expression of cannabinoid receptor subtypes by N13 microglial cells was assessed by immunocytochemistry and compared with that in cultured primary microglia. As expected, the N13 cells expressed both receptors (Fig. 1, top). Immunoreaction was present in the cell membrane and excluded the nuclei. Similar results were obtained in rat microglial cells (Fig. 1, bottom). Therefore, microglial cells should be susceptible of activation by the different cannabinoid agonists selected for the present study.

Fig. 1.

N13 microglial cells and rat primary microglia express CB₁ and CB₂ receptors. Cells were immunostained with anti-CB₁ (1:900 dilution) and anti-CB₂ antibodies (1:900 dilution). Top, N13 cells; bottom, primary microglia. Initial magnification, 200×. …

Cannabinoid Agonists Inhibit ATP-Induced Intracellular Calcium Increase in Cultured Microglia.

Variations in intracellular calcium concentration ([Ca²⁺]_i) underlie several important cell signaling functions, and they are involved in microglia activation. Indeed, an increase in extracellular ATP, released by dying neurons and glia, can interact with purinergic receptors, increase [Ca²⁺]_i, and activate microglia (Färber and Kettenmann, 2006). On the other hand, cannabinoid agonists have been shown to inhibit calcium responses in a variety of cells, including glial cells (Mato et al., 2009).

Indeed, we found that ATP increased [Ca²⁺]_i in a concentration-dependent manner (10–400 μM; data not shown) in N13 cells, reaching concentrations as high as 700 nM (nearly a 3-fold increase over basal levels) with the highest concentration tested. Intracellular calcium increased almost immediately after adding ATP to the N13 cells (see Fig. 2 C), and after reaching its maximal, even in the presence of ATP, it returned to baseline calcium levels.

Fig. 2.

CBD, WIN, and JWH inhibited the ATP-induced increase in intracellular calcium in cultured N13 and primary microglial cells. A, concentration-dependent inhibition of ATP-induced increase in [Ca²⁺]_i by cannabinoids in N13 cells. The effect of CBD (B; 100 …

For the subsequent experiments, we selected the ATP concentration of 400 μM that may mimic the high concentrations released by dying neurons and glia (e.g., pathological concentrations). Cannabinoids per se did not affect basal [Ca²⁺]_ilevels when added to the cultures in a wide concentration range (10–1000 nM). However, CBD did reduce ATP-induced [Ca²⁺]_i in a concentration-dependent manner (Fig. 2A), and the maximal effect attained was a 25% reduction. The effect of CBD in N13 microglia was not changed in the presence of either the CB₁– or the CB₂-selective antagonists (100 nM; Fig. 2E), which per se showed no effect (data not shown). The [Ca²⁺]_i response of primary microglia to ATP was different from that observed in N13 cells, because after reaching its peak effect, approximately 20 s after its addition to the culture, it was maintained at least for 50 s (Fig. 2, B and C). It is noteworthy that the CB₂ antagonist fully reversed the CBD effect (Fig. 2, B and F). WIN reduced ATP-induced intracellular calcium (Fig. 2), and although both antagonists blocked its effect in N13 cells (Fig. 2E), the CB₂-selective agonist fully reversed its effect in microglia (Fig. 2F). Finally, JWH also decreased the calcium intracellular levels after ATP addition. This effect was counteracted by the CB₂-selective antagonist in primary microglia (Fig. 2F), although it was not the case in N13 cells (Fig. 2E). HU-308 did not change ATP-induced increase in [Ca²⁺]_i at any concentration tested (10–300 nM; data not shown).

Previous work has shown that the immunosuppressive effects of CBD involved the activation of adenosine receptors, given its blockade of the equilibrative nucleoside transporter (Carrier et al., 2006). To ascertain the possible involvement of a similar mechanism responsible for the CBD reduction of ATP-induced calcium responses in microglial cells, we examined whether an A_2Aagonist could mimic the CBD action (Fig. 3). In fact, 2-[p-(2-carboxyethyl)phenethylamino]-5′-N-ethylcarboxamidoadenosine [CGS21680 (CGS)], an A_2A agonist (Klotz et al., 1998), decreased at the same extent the ATP-induce increase in [Ca²⁺]_i in N13 and primary microglial cells (Fig. 3, A and B), which was blocked by 4-{2-[7-amino-2-(2-furyl)[1,2,4]triazolo-[2,3-a][1,3,5]triazin-5-ylamino]ethyl}phenol [ZM241385 (ZM)], an A_2A antagonist (Palmer et al., 1995), that per se had no effect. Of note is that ZM was able to inhibit the CBD effect both in N13 and primary microglial cells (Fig. 3, C and D). Therefore, we have shown that CBD and other cannabinoids counteract the effects of ATP on microglial cells by a mechanism involving both cannabinoid receptor-dependent and independent mechanisms.

Fig. 3.

Implication of A_2A adenosine receptor in the effect of CBD on ATP-induced increase in intracellular calcium. Activation of A_2A receptor by CGS (100 nM) inhibited the ATP-induced increase in intracellular calcium and was reverted by the A_2A antagonist …

CBD and Other Cannabinoids Promote Microglial Cell Migration.

The Aβ peptide (Tiffany et al., 2002) and different cannabinoids trigger migration (Walter et al., 2003) that may subserve a beneficial function as a requisite for phagocytosing aggregated Aβ. Therefore, migration of microglial cells through porous membranes toward the lower chamber containing the compounds was investigated.

The mixed CB₁/CB₂ agonist WIN and the CB₂-selective agonist JWH promoted N13 cell migration (Fig. 4, A and B). Note that in these experiments, the cells changed their oval morphology (Fig. 1) to a fully round morphology with no lamellipodia (Fig. 4 B). Migration was similarly increased (approximately 20% compared with controls) by the two agonists (200 nM). The CB₁-selective agonist SR1 did not affect migration, whereas the CB₂-selective antagonist SR2 induced a partial although statistically significant effect on migration (approximately 10% increase; Fig. 2A). The effect of WIN on N13 cells was not altered by either of the cannabinoid antagonists, but full prevention of its migratory effect was obtained when both antagonists were combined. In contrast, the CB₂ antagonist completely blocked the response induced by JWH, whereas the CB₁ antagonist had no effect (Fig. 4A). Fibrillar Aβ (2 μM) induced microglial migration (20%) in comparison with vehicle-treated control cultures (Fig. 4C) and to those treated with a scrambled peptide, which had no effect. Aβ combined with the cannabinoids exerted a similar effect to that obtained with the agonists alone (data not shown). Primary microglial cells showed a more robust response (2.5-fold compared with controls) upon the addition of Aβ, LPS, or the cannabinoid agonists (Fig. 4C). The migratory effect induced by CBD and WIN was fully reversed by either of the selective antagonists (Fig. 4C). In contrast, the migration promoted by JWH and HU were only inhibited by the CB₂ antagonist (Fig. 4C). Taken together, these results indicate that CBD and the other cannabinoids promote microglial cell migration in a cannabinoid receptor-dependent manner.

Fig. 4.

Cannabinoid agonists promote microglial cell migration. WIN and JWH (200 nM) promoted N13 cell migration (A and B), which was inhibited by the combination of CB₁ and CB₂antagonist (100 nM each), or the CB₂ antagonist (SR2), respectively. Cannabinoid …

NO Generation Is Inhibited by Cannabinoids.

Cannabinoids inhibit LPS NO synthase stimulation and NO generation as reflected in nitrate accumulation in the culture media (Waksman et al., 1999). Given that LPS challenge did not generate nitrites in N13 cells, BV-2 microglial cells were used in these experiments.

Nitrite generation, measured in the culture media after LPS stimulation, was concentration-dependently reduced by all the cannabinoids tested (Fig. 5). According to their IC₅₀, their relative potencies were CBD = HU-308>JWH-133WIN (Fig. 5A). At a concentration near their IC₅₀, the inhibition of nitrite generation was greater in primary microglia (Fig. 5C) in comparison with BV-2 cells (Fig. 8B). Nitrite inhibition by cannabinoids was resistant to the antagonists tested (data not shown), and these results suggest that this microglial response is independent of cannabinoid receptor activation.

Fig. 5.

Cannabinoid agonists decreased LPS-induced nitrite generation in microglial cells. A, cannabinoid agonists concentration-dependently inhibited LPS-induced nitrite generation in BV-2 microglial cells. Cannabinoid agonist inhibition of LPS-induced nitrite …

Cannabinoids Counteract Aβ-Induced Cognitive Impairment and Increased Cytokine Gene Expression.

Given that several parameters of microglial activation in vitro were affected by CBD and other cannabinoids, we sought to determine whether these compounds were able to prevent some features of a pharmacological AD model. First, we examined whether cognitive impairment of Aβ-injected mice was affected by subchronic treatment. Mice injected with Aβ showed increased latencies to find the hidden platform in comparison with control animals (SCR + vehicle). Both CBD at a dose of 20 mg/kg and WIN 0.5 mg/kg were able to prevent the cognitive impairment shown by the animals in the Morris water maze (Fig. 6A). In fact, from the third training day, the mice showed a significant reduction in the latency to reach the hidden platform in comparison with Aβ-injected mice, and they behaved similar to the controls (SCR + vehicle). In contrast, neither JWH-133 nor HU-308 ameliorated the cognitive impairment of the animals (Fig. 6B).

Fig. 6.

CBD and WIN 55,212-2 prevented Aβ-induced learning deficit and cytokine expression. Mice received a single Aβ intraventricular injection (2.5 μg/5 μl) and were treated with the cannabinoid agonists daily (see Materials …

Gene expression of two different proinflammatory cytokines, TNF-α and IL-6, were measured in cerebral cortex of the animals treated with CBD and WIN. CBD did not alter the increased TNF-α gene expression observed in the AD mice model (Fig. 6C), and WIN partially reduced it. However, the levels of IL-6, which were dramatically increased (6-fold) in the Aβ-injected mice, were markedly decreased by both cannabinoids (Fig. 6D).

AD

Alzheimer’s disease

Aβ

β-amyloid peptide

IL-6

interleukin 6

TNF-α

tumor necrosis factor-α

WIN

WIN 55,212-2, (R)-(+)-[2,3-dihydro-5-methyl-3-(4-morpholinylmethyl) pyrrolo-[1,2,3-d,e]-1,4-benzoxazin-6-yl]-1-naphthalenyl-methanone

JWH

JWH-133

JWH-133

1,1-dimethylbutyl-1-deoxy-Δ⁹-tetrahydrocannabinol

HU

HU-308, 4-[4-(1,1-dimethylheptyl)-2,6-dimethoxyphenyl]-6,6-dimethyl-bicyclo[3.1.1]hept-2-ene-2-methanol

LPS

lipopolysaccharide

iNOS

inducible nitric-oxide synthase

SCR

scrambled peptide

FCS

fetal calf serum

PCR

polymerase chain reaction

ANOVA

analysis of variance

SR1

SR141716, N-piperidino-5-(4-chlorophenyl)-l-(2, 4-dichlorophenyl)-4-methylpyrazole-3-carboxamide

SR2

SR144528, N-[(1S)-endo-1,3,3-trimethyl bicyclo [2.2.1] heptan-2-yl]-5-(4-chloro-3-methylphenyl)-1-(4-methylbenzyl)-pyrazole-3-carboxamide

CGS

CGS21680, 2-[p-(2-carboxyethyl)phenethylamino]-5′-N-ethylcarboxamidoadenosine

ZM

ZM241385, 4-{2-[7-amino-2-(2-furyl)[1,2,4]triazolo-[2,3-a][1,3,5]triazin-5-ylamino]ethyl}phenol

VDM11

N-(4-hydroxy-2-methylphenyl)-5Z,8Z,11Z,14Z-eicosatetraenamide.