N -Arachidonoyl- l -Serine is Neuroprotective after Traumatic Brain Injury by Reducing Apoptosis

Abstract

N-arachidonoyl-L-serine (AraS) is a brain component structurally related to the endocannabinoid family. We investigated the neuroprotective effects of AraS following closed head injury induced by weight drop onto the exposed fronto-parietal skull and the mechanisms involved. A single injection of AraS following injury led to a significant improvement in functional outcome, and to reduced edema and lesion volume compared with vehicle. Specific antagonists to CB2 receptors, transient receptor potential vanilloid 1 (TRPV1) or large conductance calcium-activated potassium (BK) channels reversed these effects. Specific binding assays did not indicate binding of AraS to the GPR55 cannabinoid receptor. N-arachidonoyl-l-serine blocked the attenuation in phosphorylated extracellular-signal-regulated kinase 1/2 (ERK) levels and led to an increase in pAkt in both the ipsilateral and contralateral cortices. Increased levels of the prosurvival factor Bcl-xL were evident 24 hours after injury in AraS-treated mice, followed by a 30% reduction in caspase-3 activity, measured 3 days after injury. Treatment with a CB2 antagonist, but not with a CB1 antagonist, reversed this effect. Our results suggest that administration of AraS leads to neuroprotection via ERK and Akt phosphorylation and induction of their downstream antiapoptotic pathways. These protective effects are related mostly to indirect signaling via the CB2R and TRPV1 channels but not through CB1 or GPR55 receptors.

Keywords

apoptosis endocannabinoids neuroprotection traumatic brain injury

Introduction

The endocannabinoid (eCB) G_i/o-protein-coupled receptors CB1 and CB2 were previously identified in the brain (Devane et al, 1988; Munro et al, 1993), followed by identification of the eCB agonists anandamide (arachidonoylethanolamide, AEA) and 2-arachidonoylglycerol (2AG) (Devane et al, 1992; Mechoulam et al, 1995; Sugiura et al, 1995; Van Sickle et al, 2005).

Traumatic brain injury (TBI) leads to an increase in 2AG levels in the brain, which lasts for at least 24 hours after the injury. Treatment with exogenous 2AG following TBI reduces blood–brain barrier permeability, brain water content, lesion volume, and hippocampal cell death. In addition, behavioral function improves significantly in treated mice (Panikashvili et al, 2001). These beneficial effects were abolished when 2AG was coadministrated with the selective CB1-antagonist SR141716A (rimonabant) as well as in CB1^−/- knockout mice, suggesting that they are mediated via CB1 receptor signaling (Panikashvili et al, 2001, 2005). Additionally, administration of AEA reduced the volume of cytotoxic edema in a manner that was insensitive to SR141716A (van der Stelt et al, 2001), suggesting that at least part of the beneficial effects of AEA are produced via activation of mechanisms independent of CB1 signaling. Recently, it was proposed that also CB2 receptors mediate neuroprotection after middle cerebral artery occlusion (Zhang et al, 2007). Indeed, injection of CB2 agonist together with CB1 antagonist led to a better outcome in an ischemia–reperfusion model (Zhang et al, 2008).

N-arachidonoyl-l-serine (AraS) is an eCB-like compound with a structure similar to AEA that was isolated from bovine brain (Milman et al, 2006). N-arachidonoyl-l-serine causes endothelium-dependent arterial vasodilatation and stimulates phosphorylation of p44/42 mitogen-activated protein kinase and protein kinase B/Akt in cultured endothelial cells (Milman et al, 2006). These effects are similar to the stimulation observed following treatment with the classical eCBs (Chen et al, 2000; Golech et al, 2004), despite the very low binding affinity of AraS to CB1 and CB2 receptors (Milman et al, 2006). Moreover, AraS was recently demonstrated to be an activator of large conductance calcium-activated potassium channels in vitro (Godlewski et al, 2009). Therefore, the present study aimed to investigate whether administration of exogenous AraS would result in reduced disability and smaller lesion volumes following TBI and to elucidate the mechanisms involved.

Materials and methods

Reagents

N-arachidonoyl-l-serine was synthesized in our laboratory as previously described (Milman et al, 2006). Cremophor, 2,3,5-triphenyltetrazolium chloride (TTC), capsazepine, paxilline, bovine serum albumin, guanosine 5'-diphosphate, ethylenediaminetetraacetic acid, ethylene glycol-bis(2-aminoethylether)-N,N,N',N'-tetraacetic acid, HEPES (4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid), Tris–HCl, sodium orthovanadate (Na₃VO₄), protease inhibitor cocktail, phenylmethanesulfonyl fluoride, sodium pyrophosphate, sodium β glycerophosphate, and 2-β-mercaptoethanol were purchased from Sigma-Aldrich (St Louis, MO, USA). Sodium chloride, magnesium chloride, and sodium fluoride were purchased from Merck (Darmstadt, Germany). CB1-antagonist rimonabant (SR141716A) and CB2-antagonist (SR144528) were received from the Research Triangle Institute (Research Triangle Park, NC, USA). A Bio-Rad protein assay was purchased from Bio-Rad (Bio-Rad Laboratories, Munich, Germany), [³⁵S]-GTPγS ([³⁵S]-Guanosine 5'-[γ⁻³⁵S]-triphosphate) was purchased from GE Healthcare (Piscataway, NJ, USA) and NP-40 from USB (Cleveland, OH, USA).

Animals

The study was performed according to the Institutional Animal Use and Care Committee guidelines and was approved by the institution's Animal Care and Use Committee. Male Sabra mice (Harlan, Jerusalem, Israel) aged 6 to 7 weeks and weighing 35 to 45 g were used in this study. This is a strain bred at the Hebrew University and provided from the same supplier (Harlan). No variations in dominant/submissive characteristics were observed with this strain. For each experimental group, 6 to 9 mice were used unless otherwise mentioned.

Trauma Model

Mice were subjected to closed head injury (CHI) under isoflurane anesthesia, using a weight-drop device that falls over the left hemisphere, as described elsewhere (Tsenter et al, 2008). In brief, after a longitudinal scalp incision, mice were immobilized under a cylindrical calibrated weight-drop device. A tipped teflon cone was placed (upside down) 2 mm lateral to the midline and 1 mm caudal to the left coronal suture, and a metal rod (94 g) was dropped down on the cone from a height of 11 to 14 cm (adjusted to body weight, to ensure the severity of injury (Tsenter et al, 2008) required to produce CHI). Sham-treated mice were anesthetized with isoflurane, their scalps were incised, but trauma was not induced. It should be noted that < 10% of the injured mice were excluded from the study, mostly because of death by apnea within minutes of injury.

Drug Application

Following preliminary experiments in which dose and time window were determined, AraS 3 mg/kg dissolved in ethanol:cremophor:saline 1:1:18 or vehicle alone was injected intraperitoneally into the mice 1 hour after CHI, as described elsewhere (Panikashvili et al, 2001). To evaluate the involvement of CB1R, CB2R and transient receptor potential vanilloid 1 (TRPV1) in the neuroprotective effects of AraS, CB1-antagonist rimonabant (SR141716A) 1 mg/kg (Barna et al, 2009; Hayakawa et al, 2007), or the CB2-antagonist SR148522, 1 mg/kg (Fride et al, 2005; Hanus et al, 1999), or Capsazepine, a TRPV1 antagonist, 1 mg/kg (E Berry, Laboratory of Human Nutrition Hebrew University Jerusalem Israel; personal communication), together with AraS were also injected into the mice. As AraS was found to be an activator of large conductance Ca⁺-activated K⁺ (BK) channels, we also tested the involvement of these channels in the neuroprotective effect of AraS, using the BK channel blocker paxilline (6 mg/kg) (Imlach et al, 2008) alone or with AraS. Paxilline was dissolved in ethanol, sonicated, and diluted in ethanol:-cremophor:saline (1:1:18) to the final concentration.

Evaluation of Functional Outcome

At 1 hour after CHI, the functional status of the mice was evaluated according to a set of 10 neurobehavioral tasks (neurologic severity score, NSS), which examine reflexes, alertness, coordination, motor abilities, and balancing (Beni-Adani et al, 2001). Failure to perform a task scores one point and a success scores 0. Hence, normal animals score 0, reflecting healthy mice, whereas a score of 10 reflects maximal neurologic impairment. Only mice with NSS 6 to 8 at 1 hour after injury (NSS 1hour) were included in the study. Immediately after evaluation of NSS at 1 hour, the mice were randomly assigned to treatment groups (n = 6 to 9 mice/group), which received vehicle or drug and NSS was reevaluated on days 1, 2, 3, 7, and 14 after CHI. The analyses were performed by an investigator that was blinded to treatment.

Lesion Volume

Lesion volume was evaluated with the aid of TTC staining, as described elsewhere (Povlishock and Katz, 2005). The mice were subjected to CHI followed by the different treatments (n = 5 to 16 mice/group). After 24 hours they were killed and their brains were removed and sliced at 2 mm, intervals, using a brain mold. The slices were placed in 2% TTC solution in saline for 30 minutes and preserved in 4% paraformaldehide in PBS. The brains were photographed 24 hours later with the aid of a Stemi SV11 Stereoscope (Zeiss, Goettingen, Germany) and a digital camera (Nikon, Coolpix 4500, Natori, Miyagi, Japan). ImageJ 1.40g software (National Institutes of Health, Bethesda, MD, USA) was used to quantify lesion volume. To avoid inaccuracies due to changes in the ipsilateral hemispheric volume, the lesion volume was calculated as the area of unstained tissue divided by the area of contralateral hemisphere (Swanson et al, 1990).

Cerebral Edema

The edema was determined at 24 hours after CHI, as previously described (Panikashvili et al, 2005). Cortices of both ipsilateral and contralateral hemispheres were removed and weighted immediately and then dried for 24 hours in a 95°C oven (n = 10 to 12 mice/group). The water content of the tissue was calculated as:

% H_{2} O = \frac{w e t w e i g h t - d r y w e i g h t}{w e t w e i g h t} \times 100

Rectal Temperature Measurements

As cannabinoids have a hypothermic effect, which may contribute to neuroprotection, we assessed the effect of AraS on rectal temperature of mice after trauma (n = 5 to 10 mice/group), which correlates well with brain temperature. The temperature was measured 2, 4, 8, and 24 hours following the insult, using rectal thermistors (Digisense, Thermistor 400 series, Euthech Instruments, Singapore) inserted 3 cm beyond the anal sphincter.

GTPγS Binding (Radio-Labeling) Assay

As AraS was shown to have a very low binding affinity for CB1 and CB2 receptors (Milman et al, 2006), we decided to examine its binding to other candidate receptors from the G-protein-coupled receptor (GPCR) family, using a radio-labeling assay as described elsewhere (Ryberg et al, 2007). A dilution series of AraS was prepared in DMSO in 96-well plates, at 2 μL/well. This gave a final DMSO concentration of 1%. The highest concentration of compound in the assay was 33 μmol/L. [³⁵S]-Guanosine 5'-[γ⁻³⁵S]-triphosphate binding assays were conducted at 30°C for 45 minutes in membrane buffer (100 mmol/L NaCl, 5, 1 mmol/L ethylenediaminetetraacetic acid, 50 mmol/L HEPES, pH 7.4) containing 0.025 μg/μL of membrane protein with 0.01% bovine serum albumin (fatty-acid free), 10 μmmol/L guanosine 5'-diphosphate, 100 μmmol/L dithiothreitol, and 0.53 nM [³⁵S]-GTPγS in a final volume of 200 μL. Nonspecific binding was determined in the presence of 20 μmmol/L unlabeled GTPγS. The reaction was terminated by the addition of ice-cold wash buffer (50 mmol/L Tris–HCl, 5 mmol/L MgCl₂, 50 mmol/L NaCl, pH 7.4), followed by rapid filtration under vacuum through GF/B glass-fiber filters using a cell harvester. The filters were left to dry for 30 minutes at 50°C. Then 15 μL scintillation liquid was added to each well before the bound radioligand was detected with the aid of a microbeta scintillation counter (Wallac 1450 Microbeta TRILUX, Turku, Finland). Antagonist potency was determined versus an EC₈₀ concentration of CP55940 that was determined empirically on the day of the experiment. Data were fitted according to the equation y = A + ((B - A)/1 + ((C/x)^D))) and the EC₅₀ was estimated where A = the nonspecific binding, B = the total binding, C = the EC₅₀, and D = the slope. Antagonist potency was determined versus an EC₈₀ concentration of GPR55 agonist O-1602 that was determined empirically on the day of the experiment. O-1602 (5-methyl-4-[(1R,6R)-3-methyl-6-(1-methylethenyl)-2-cyclohexen-1-yl]-1,3-benzenediol) is a synthetic atypical cannabinoid analog of cannabidiol, which binds as an agonist to GPR55. It is similar in structure to abnormal cannabidiol but with a methyl group instead of pentyl group substitution of the benzene. The data were fitted as above.

Western Immunoblotting

To determine the levels of phosphorylated extracellular-signal-regulated kinase 1/2 (pERK1/2), pAkt, and Bcl-xL, the mice were decapitated under isoflurane anesthesia 2, 4, and 24 hours after injury (n = 5 to 6 mice/group). The cortical tissue adjacent to the site of injury was removed and homogenized using a manual homogenizer, in ice-cold buffer (50 mmol/L Tris–HCl, pH 7.5, 150 mmol/L NaCl, 1 mmol/L ethylenediaminetetraacetic acid, 1 mmol/L ethylene glycol-bis(2-aminoethylether)-N,N,N',N'-tetraacetic acid, 1 mmol/L Na₃VO₄, 50 mmol/L NaF, 10 mmol/L sodium β-glycerophosphate, 5 mmol/L sodium pyrophosphate, 1 mmol/L phenylmethanesulfonyl fluoride, supplemented with 0.1% (v/v) Triton X-100, 0.1% (v/v) 2-β-mercaptoethanol, and protease inhibitor cocktail). For comparison, tissue from the contralateral hemisphere was removed and homogenized separately in the same buffer. After 15 minutes incubation on ice, the extracts were centrifuged at 12,000g for 15 minutes at 4°C and the protein content in the supernatant was determined according to Bradford (1976) and the manufacturer's instructions. A total of 50 μg of the protein extract from each sample was mixed with SDS-sample buffer and boiled at 95°C for 5 minutes before loading. The proteins were separated by 10% SDS–PAGE and transferred onto trans-blot nitrocellulose membranes. The membranes were blocked in 5% nonfat dry milk in Tris buffered saline (TBS), pH 7.4, with 0.1% Tween 20 (TBS–Tween) for 1 h at room temperature. Primary antibodies (phospho-ERK1/2 1:1,000, phospho-AKT Ser473 1:1,000, Bcl-xL 1:1,000, β-actin 1:1,000 (Cell Signaling Technology, Danvers, MA, USA) were diluted in 5% bovine serum albumin solution, and the membranes were incubated overnight at 4°C. The primary antibody was removed, and the blots were washed in TBS–Tween and incubated for 1 h at room temperature with horseradish peroxidase-conjugated secondary antibodies (1:10,000; Jackson Immunoresearch, Cambridgeshire, UK). Reactive proteins were visualized using Enhanced Chemiluminescence Reagent (Biological Industries, Beit Haemek, Israel) by direct chemiluminescence using FUJIFILM Luminescent Image Analyzer camera LAS-3000 (Tokyo, Japan). The optical density was determined using ImageJ 1.40g software (National Institutes of Health).

Caspase-3 Activity Assay

Caspase-3 activity was measured in the cortical tissue of the injured mice using the commercially available BioVision kit (Mountain View, CA, USA) according to the manufacturer's instructions (n = 5 to 9 mice/group). In brief, the mice were subjected to CHI followed by treatment with AraS (3 mg/kg), AraS + SR141716A, AraS + SR144528, AraS + capsazepine, or vehicle 1 hour after injury. The mice were killed 72 hours later, and the ipsilateral cortices were removed, lysed in 200 μL of the supplied lysis buffer using a manual homogenizer, and frozen immediately at −80°C until assayed. Total protein was calculated and 150 μg of the samples were loaded in each well of a 96-well plate. In all, 50 μL of reaction buffer containing 10 mmol/L dithiothreitol and 5 μL of DEVD-AFC (Asp-Glu-Val-Asp7-amino-4-trifluoromethylcoumarin) substrate were added and the plates were incubated at 37°C for 2 hours. The cleavage product, representing active caspase-3, was detected using a fluorometer (FluoroStar 403, BMG LabTech, Offenburg, Germany) equipped with 390 nm excitation and 520 nm emission filters.

Statistical Analysis

Statistical analysis was performed using SigmaStat 2.03 software (SPSS, Inc., Chicago, IL, USA). The data are presented as the mean ± s.e.m. The statistical significance of differences between means was evaluated by two-way analysis of variance for repeated measures, followed by Tukey post hoc for NSS tests, and with parametric one-way analysis of variance for lesion volume and for fluorometric measurements. The statistical significance of the optical density measurements was calculated using two-way analysis of variance followed by Tukey correction. Probability values (P) < 0.05 were considered statistically significant.

Results

N-Arachidonoyl-l-Serine Improves Motor Function and Reduces Lesion Volume and Water Content After Closed Head Injury

Mice treated with AraS or vehicle had a similar NSS at 1 hour, indicating a similar severity of injury (6.36 ± 0.11 and 6.32 ± 0.09, respectively). The motor function of mice was evaluated for up to 14 days after CHI. Significantly lower NSS values were recorded in AraS-treated mice compared with those in vehicetreated mice, from day 3 up to 14 days of follow-up (Figure 1; P < 0.01 on day 3 and P < 0.001 on days 7 and 14), indicating better recovery rates in the AraS-treated mice. To examine the effect of AraS on lesion volume, brains were stained with the vital dye TTC, which stains live tissue purple and leaves dead tissue unstained. Lesion volumes constituted 8.00% ± 0.87% of the hemispheric volume in the AraS-treated mice, compared with 14.39% ± 2.35% in the vehicle-treated mice (Figure 2). The results indicate a significant reduction of 45% in lesion volume in the AraS-treated mice (P < 0.01). The water content in the brain was determined 24 hours after CHI in both ipsilateral and contralateral hemispheres. A significant reduction in water content was observed in the ipsilateral cortex of AraS-treated mice compared with vehicle (81.82% ± 0.16% and 82.77% ± 0.25%, respectively; P < 0.01, data not shown). No difference in water content was detected in the contralateral hemispheres of both groups.

Figure 1

N-arachidonoyl-l-serine (AraS) reduces neurologic disability after closed head injury (CHI). Neurologic severity score (NSS) of mice treated with vehicle, AraS, AraS + rimonabant (SR141716A), AraS + SR144528, AraS + capsazepine, or AraS + paxilline (**P < 0.01 versus vehicle, ***P < 0.001 versus vehicle, ^#P < 0.05 versus AraS, ^##P < 0.01 versus AraS, ^###P < 0.001 versus AraS). AraS induced better neurobehavioral function after CHI was abolished by CB2 and transient receptor potential vanilloid 1 (TRPV1) antagonists SR144528 and capsazepine, respectively. The CB1 receptor antagonist rimonabant (SR141716A) did not alter the neuroprotective effect of AraS.

Figure 2

N-arachidonoyl-l-serine (AraS) reduces lesion volume after closed head injury (CHI). Lesion volumes measured 24 hours after CHI from 2,3,5-triphenyltetrazolium chloride (TTC)-stained brains after treatment with vehicle, AraS, AraS + rimonabant (SR141716A), AraS + SR144528, AraS + capsazepine, or AraS + paxilline (**P < 0.01 versus vehicle, ^#P < 0.05 versus AraS, ^##P < 0.01 versus AraS, ^###P < 0.001 versus AraS). AraS reduced hemispheric lesion volume after CHI, which was completely abolished by CB2 and transient receptor potential vanilloid 1 (TRPV1) antagonists SR144528 and capsazepine, respectively, and to some extent by BK channels blocker (paxilline). The reduction was not affected by CB1 antagonist (SR141716A).

N-Arachidonoyl-l-Serine Neuroprotective Effects Involve CB2 and Transient Receptor Potential Vanilloid 1 Receptors

We next sought to identify the receptors involved in mediating the neuroprotective effects of AraS. The CB1 receptors were considered likely candidates as the effects of AraS on motor disability were similar in magnitude to those previously observed for 2AG (Panikashvili et al, 2001). However, coadministration of AraS with the CB1-antagonist rimonabant (SR141716A) did not alter the recovery of the AraS-treated animals (Figure 1) nor reduce the lesion volume (Figure 2; 9.86% ± 1.03%), arguing against a CB1-mediated effect of AraS.

As 2AG also binds to CB2 and TRPV1 receptors, this raised the possibility that the protective effect of AraS may involve these binding sites. Coadministration of AraS with the CB2R-antagonist SR144528 significantly ameliorated the decrease in NSS values compared with those obtained for AraS-treated mice (Figure 1). The values were similar to those observed in vehicle-treated mice, 3, 7, and 14 days after injury (P < 0.05, P < 0.001, and P < 0.01, respectively, versus AraS). Moreover, a threefold increase in lesion volume was observed in mice treated with a combination of AraS and SR144528 (Figure 2; 24.33% ± 2.85%; P < 0.001) compared with those treated with AraS alone (8.00% ± 0.87%). To examine whether AraS acts via ‘nonclassical’ CB receptors, mice were treated with capsazepine, a TRPV1 receptor antagonist. They exhibited significantly higher NSS values than those obtained for the AraS-treated group 7 and 14 days after CHI (Figure 1; P < 0.05 and P < 0.01, respectively) and also showed a twofold increase in lesion volume compared with the volume obtained in AraS-treated mice (Figure 2; 15.61% ± 4.51% versus 8.00% ± 0.87%, respectively; P < 0.01), suggesting that the TRPV1 channels have a role in the AraS neuroprotective effects.

N-arachidonoyl-l-serine was recently found to activate large conductance calcium-activated potassium (BK) channels in vitro (Godlewski et al, 2009). Therefore, the involvement of these channels in the neuroprotective effect of AraS was also examined. A single dose of 6 mg/kg of paxilline, a BK channels blocker, was injected to mice 1hour after CHI with or without AraS. The functional recovery of the AraS + paxilline-treated group tended to be less pronounced than that of the AraS-treated mice, yet, this tendency became significant only 2 weeks after injury (Figure 1). In parallel, a 36% increase in lesion volume was observed 24hours after injury in mice treated with AraS in combination with paxilline compared with the lesion volume in mice treated with AraS alone (Figure 2; 10.94% ± 1.51% and 8.00% ± 0.87%, respectively; P < 0.05).

Recently, a new G-protein-coupled cannabinoid receptor, GPR55, was identified but it has not been fully characterized (Ryberg et al, 2007). GTPγS binding assays showed no significant agonistic binding of AraS to GPR55 (Table 1A) and no significant antagonistic activity was observed against the GPR55 receptor (Table 1B).

Table 1

AraS does not bind to GPR55 receptors as an agonist (A) or as an antagonist (B)

Compound	GPR55 EC₅₀ (nmol/L)
(A)
Anandamide	28
WIN55,212-2	NE
Arachidonoyl-serine	NE
(B)
Cannabidiol	717
Arachidonoyl-serine	NE

AraS, N-arachidonoyl-l-serine; NE, nonexistent.

Cumulatively, these results suggest that AraS mediates its protective effects in vivo by indirect signaling via CB2R, TRPV1 and partially via BK channels, but not through the CB1R or GPR55 receptors.

N-Arachidonoyl-l-Serine Prevents Hypothermia After Closed Head Injury

Hypothermia, a known neuroprotective mechanism, was found to mediate CB1-induced neuroprotection of the potent synthetic cannabinoid HU-210 (Leker et al, 2003). To evaluate whether AraS induces hypothermia and whether such an effect has a role in AraS-induced neuroprotection, the rectal temperature of mice treated with AraS or vehicle was measured at different time points after injury. Although not a direct measure of brain temperature, rectal and brain temperature are highly correlated with each other. No significant decrease in body temperature was observed in AraS-treated mice 2 hours after injury compared with preinjury temperature (36.77°C ± 0.21°C and 37.64°C ± 0.46°C, respectively), whereas a reduction in body temperature was detected in vehicle-treated mice at the same time point (35.19°C ± 0.37°C and 37.31°C ± 0.09°C, respectively). At all other time points, no significant changes were observed between the different treatment groups (data not shown), arguing against the involvement of a hypothermic mechanism of neuroprotection for AraS.

N-Arachidonoyl-l-Serine Prevents Attenuation in Phosphorylation of Extracellular-Signal-Regulated Kinase 1/2 Levels and Increases pAkt Levels of After Closed Head Injury

To further delineate the downstream signaling involved in the neuroprotective effects of AraS, we studied its effect on ERK, a key antiapoptotic prosurvival mediator. Mice were subjected to trauma, treated with AraS or vehicle 1 hour later and decapitated 2, 4, or 24 hours after injury. The levels of pERK were evaluated in the ipsilateral cortices. Although a significant decline in the levels of pERK was observed at 2 and 4hours after injury in the vehicle-treated mice (P < 0.01 versus sham), no decline was observed in the AraS-treated mice at 2 hours after injury (P < 0.01 versus vehicle). Interestingly, a dramatic decrease in pERK levels was noted 4 hours after CHI (P < 0.001 versus sham) in the AraS-treated mice. In both groups, the pERK levels increased, reverting to normal 24 hours after injury (Figure 3).

Figure 3

N-arachidonoyl-l-serine (AraS) prevents reduction in phosphorylated extracellular-signal-regulated kinase 1/2 (pERK) levels 2 hours after closed head injury (CHI). (A) Western blots for phosphorylated extracellular-signal-regulated kinase (pERK) at different time points after CHI from brains treated with AraS or vehicle. (B) Quantitative measurements of optical density of the bands shown in (A) (*P < 0.05 versus sham, **P < 0.01 versus sham, ***P < 0.001 versus sham, ^##P < 0.01 versus AraS 2 hours, ^###P < 0.001 versus AraS 2 hours). GAPDH, Glycer-aldehyde 3-phosphate dehydrogenase.

The effect of AraS on Akt phosphorylation was next measured by Western blot analysis in the cortex of both hemispheres at 4 and 24 hours after CHI. A significant increase in pAkt levels was observed 4 hours after CHI in the ipsilateral (59.7% ± 19.6%) as well as in the contralateral (54.96% ± 10.72%) cortex of AraS-treated mice, compared with those in the vehicle-treated and sham-operated mice (Figure 4). A drop to basal levels was observed in these regions 24 hours after injury (Figure 4).

Figure 4

N-arachidonoyl-l-serine (AraS) increases pAkt levels 4 hours after closed head injury (CHI). (A) Western blots for phosphorylated protein kinase B/Akt (pAkt) at the ipsilateral (ipsi) and contralateral (contra) cortex to the lesion, 4 and 24 hours after CHI. (B) Quantitative measurements of the optical density of the bands shown in (A) (*P < 0.05 versus vehicle, ***P < 0.01 versus vehicle).

N-Arachidonoyl-l-Serine Increases Bcl-xL Levels and Reduces Caspase-3 Activity in a CB2 Involved Mechanism

As AraS reduced lesion volume and since pERK and pAkt are involved in the regulation of apoptosis, we examined the levels of Bcl-xL, an antiapoptotic member of the bcl-2 family, 24 hours after injury. A significant increase of 62.6% ± 12.8% in Bcl-xL level was found in the ipsilateral cortex of mice treated with AraS compared with that in the vehicle (Figure 5). The effect of AraS on the proapoptotic protease caspase-3 activity was then evaluated in the ipsilateral cortex 3 days after CHI. Caspase-3 activity, expressed as fluorescence units, was almost 300% higher in the ipsilateral cortex of vehicle-treated mice compared with that in sham mice. N-arachidonoyl-l-serine significantly attenuated this increase (Figure 6A; P < 0.01). Treatment with AraS together with SR144528 blocked the antiapoptotic effect of AraS, with caspase-3 activity similar to that of vehicle-treated mice (P < 0.05 versus AraS). These results suggest a CB2-dependent mechanism. Surprisingly, administration of AraS in combination with capsazepine did not block the attenuation in caspase-3 activity, suggesting that TRPV1-mediated neuroprotection does not necessarily involve an antiapoptotic mechanism that is mediated via caspase-3. As expected, injection of AraS together with rimonabant did not affect caspase-3 activity (Figure 6A).

Figure 5

N-arachidonoyl-l-serine (AraS) increases Bcl-xL levels after closed head injury (CHI). (A) Western blots for Bcl-xL levels, 4 and 24hours after CHI. (B) Quantitative measurements of the optical density of the bands shown in (A) (***P < 0.001 versus vehicle, ipsi—cortex ipsilateral to the lesion, contra—cortex contralateral to the lesion).

Figure 6

N-arachidonoyl-l-serine (AraS) reduces active caspase-3 levels after closed head injury (CHI). (A) The effect of AraS treatment on caspase-3 activity (expressed as fluorescence units of DEVD-AFC cleaved product) in the brains of sham-operated controls and CHI animals treated with vehicle, AraS, AraS + SR141716A, AraS + SR144528, or AraS + capsazepine 72 hours after injury (&P < 0.05 versus AraS, ^##P < 0.01 versus vehicle, ***P < 0.001 versus sham). (B) Scheme describing the suggested mechanism triggered by AraS (Adapted from Cudaback et al, 2010; Zhuang et al, 2004; Datta et al, 1997; Yang et al, 1995; Chen et al, 2005).

Cumulatively, these findings suggest an antiapoptotic mechanism of action for the neuroprotective effects of AraS involving CB2, but not CB1 receptors as is shown in the schematic Figure 6B.

Discussion

In the present study, we show for the first time that the recently described eCB-like compound AraS mediates neuroprotection in vivo after TBI via the induction of a prosurvival and antiapoptotic cascade. This mechanism involves Akt and ERK phosphorylation and its downstream signaling includes increased levels of the antiapoptotic protein Bcl-xL and the reduced activity of the proapoptotic enzyme caspase-3. N-arachidonoyl-l-serine-induced neuroprotection involves signaling through CB2 receptors, TRPV1 and BK channels. Although the protective effects exerted by AraS are similar in magnitude to those observed by 2AG (Panikashvili et al, 2001), they are not sensitive to the specific CB1-antagonist rimonabant and are thus independent of CB1 signaling. These findings are in agreement with those of Milman et al (2006) who did not find binding of AraS to CB1R. Furthermore, we did not detect binding of AraS to GPR55—another candidate cannabinoid receptor that was recently found to be involved in AraS signaling (Zhang et al, 2010).

Cannabinoids have been shown to provide neuroprotection via hypothermia, in addition to other mechanisms (Leker et al, 2003). Therefore, we assessed the hypothermic effect of AraS following brain injury. No decrease in body temperature was observed in AraS-treated mice, indicating that the neuroprotective effects of AraS are not mediated via the induction of hypothermia.

Many deleterious and protective processes are induced following TBI, including massive release of glutamate from presynaptic terminals, production of reactive oxygen species, accumulation of inflammatory cytokines in the surrounding tissue, axonal injury, and neuronal cell loss in specific brain regions (Leker and Shohami, 2002; Povlishock and Katz, 2005). In parallel, neuroprotective events are also set in motion and include secretion of antiinflammatory cytokines, growth factors, antioxidative compounds, and enzymes (Leker and Shohami, 2002). The eCB system was proposed to be part of the protective processes (Panikashvili et al, 2001, 2005). The fact that it is activated ‘on-demand’ (Piomelli, 2003) lends support to this supposition.

The neuroprotective effects of 2AG after brain injury were shown to be mediated through CB1 signaling (Panikashvili et al, 2001, 2005). However, since 2AG activates both the CB1 and CB2 receptors, some of the neuroprotective effects may be mediated by signaling via CB2 (Zhang et al, 2007, 2008). These effects may explain our findings on the reduction of cerebral edema in the AraS-treated mice. In addition, CB1 receptor agonists were shown to activate Akt (Gomez del Pulgar et al, 2000) and ERK (Moranta et al 2007), which participate in prosurvival processes, including those related to neuroprotection (Pignataro et al, 2008; Yune et al, 2008). Our present results, in agreement with those of Milman et al (2006), suggest that in the brain, as in endothelial cells, the novel eCB AraS also induces ERK and Akt signaling via mechanisms that probably depend on CB2 and/or TRPV1 and/or slightly on BK channels, but not on CB1. Importantly, Akt kinase participates in many prosurvival and antiapoptotic processes inside and outside the brain (Dasari et al, 2008), including processes that are cannabinoid related (Molina-Holgado et al, 2002; Ozaita et al, 2007), thus corroborating the proposed neuroprotective mechanism of AraS. Here, we show that administration of exogenous AraS also leads to an increase in the antiapoptotic factor Bcl-xL at 24hours and to the decreased activity of the proapoptotic protease caspase-3, 72 hours after injury. BAD is a proapoptotic member of the Bcl-2 family; Bcl-xL intervenes in the apoptotic cascade and attenuates it. When unphosphorylated BAD forms a complex with Bcl-xL, it prevents the antiapoptotic effect of the latter, resulting in apoptosis (Datta et al, 1997; Yang et al, 1995). The phosphorylation of BAD by Akt (Datta et al, 1997) and/or by ERK (Chen et al, 2005), prevents this interaction and allows Bcl-xL to exert its antiapoptotic effect and eventually to inhibit programmed cell death. The attenuation in caspase-3 activity demonstrated by AraS was reversed by CB2R blocker pointing to this receptor as the main conduit of the antiapoptotic effect of AraS.

Importantly, the prosurvival antiapoptotic cellular mechanisms induced in AraS-treated mice led to a 45% reduction in lesion volume, and to a decrease in the neurologic disability, as assessed by the NSS test. A correlation between caspase-3 activity, lesion volume, and neurobehavioral deficits was observed in mice after treatment with the different antagonists. Thus, higher caspase-3 activity found following treatment with AraS in combination with the CB2 blocker led to larger lesions and to a more severe functional outcome. In contrast, in mice treated with AraS in combination with CB1 antagonist, caspase-3 activity, lesion volume, and neurologic outcome were similar to those observed in AraS-treated mice. This indicates that the CB1 receptor is not involved in the neuroportective mechanism. Surprisingly, although lesion volume in mice treated with AraS along with capsazepine was higher and the functional outcome was worse compared with that of the AraS group, caspase-3 activity remained unchanged. These results may imply that the mechanism of AraS-induced neuroprotection mediated by TRPV1 channels does not necessarily involve an antiapoptotic effect or involves an antiapoptotic effect that is not mediated via modulation of caspase-3 activity. As the changes in neurobehavioral function of paxilline-treated mice compared with those of AraS-treated mice were mostly insignificant, lesion volume was hardly affected by the blocker and it appears that the neuroprotective mechanism of AraS does not involve BK channels.

Although AraS does not directly bind to CB2 or to TRPV1, it may exert its neuroprotective effects in two optional mechanisms: (1) It may have an entourage effect. N-arachidonoyl-l-serine may occupy FAAH and thus allows other cannabinoids, such as AEA, to exert their neuroprotection for longer period. (2) N-arachidonoyl-l-serine binds to the receptors/channels although not at their classic binding site (e.g., in lipid rafts within the membrane). In this case, it may be a cofactor for other cannabinoids/compounds and allow them to bind to the binding site and exert their neuroprotective effects (Figure 6B).

Since AraS reduces water content and lesion volume 24 hours after CHI, we postulate that it affects straight on pERK and pAkt directly and rapidly that eventually leads to a reduction in the pathophysiological parameters. Therefore, we believe the effect of AraS on apoptosis is the mechanism by which it provides neuroprotection and that the reduced levels of caspase-3 are not merely the result of smaller lesions.

In conclusion, we show here for the first time that the novel eCB AraS exerts neuroprotective effects in vivo in a CHI model of TBI. These effects are mediated at least in part by the induction of antiapoptotic signaling involving pERK, pAkt, and Bcl-xL, leading to reduced caspase-3 activity. In contrast to the effects seen upon the administration of 2AG, these neuroprotective effects are probably related indirectly to CB2 receptors and TRPV1 and partially to BK channels, but not to signaling via CB1 or GPR55 receptors.

Footnotes

The authors declare no conflict of interest.

References

Barna

Till

Haller

(2009) Blood, adipose tissue and brain levels of the cannabinoid ligands WIN-55,212 and SR-141716A after their intraperitoneal injection in mice: compound-specific and area-specific distribution within the brain. Eur Neuropsychopharmacol 19:533–41

Beni-Adani

Gozes

Cohen

Assaf

Steingart

Brenneman

Eizenberg

Trembolver

Shohami

(2001) A peptide derived from activity-dependent neuroprotective protein (ADNP) ameliorates injury response in closed head injury in mice. J Pharmacol Exp Ther 296:57–63

Bradford

(1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72:248–54

Chen

Lau

Fung

(2005) Inactivation of Bad by site-specific phosphorylation: the checkpoint for ischemic astrocytes to initiate or resist apoptosis. J Neurosci Res 79:798–808

Chen

McCarron

Ohara

Bembry

Azzam

Lenz

Shohami

Mechoulam

Spatz

(2000) Human brain capillary endothelium: 2-arachidonogly-cerol (endocannabinoid) interacts with endothelin-1. Circ Res 87:323–7

Cudaback

Marrs

Moeller

Stella

(2010) The expression level of CB1 and CB2 receptors determines their efficacy at inducing apoptosis in astrocytomas. PLoS One 5:e8702

Dasari

Veeravalli

Saving

Gujrati

Fassett

Klopfenstein

Dinh

Rao

(2008) Neuroprotection by cord blood stem cells against glutamate-induced apoptosis is mediated by Akt pathway. Neurobiol Dis 32:486–98

Datta

Dudek

Tao

Masters

Gotoh

Greenberg

(1997) Akt phosphorylation of BAD couples survival signals to the cell-intrinsic death machinery. Cell 91:231–41

Devane

Dysarz

III Johnson

Melvin

Howlett

(1988) Determination and characterization of a cannabinoid receptor in rat brain. Mol Pharmacol 34:605–13

10.

Devane

Hanus

Breuer

Pertwee

Stevenson

Griffin

Gibson

Mandelbaum

Etinger

Mechoulam

(1992) Isolation and structure of a brain constituent that binds to the cannabinoid receptor. Science 258:1946–9

11.

Fride

Ponde

Breuer

Hanus

(2005) Peripheral, but not central effects of cannabidiol derivatives: mediation by CB(1) and unidentified receptors. Neuropharmacology 48:1117–29

12.

Godlewski

Offertaler

Osei-Hyiaman

Harvey-White

Liu

Davis

Zhang

Razdan

Milman

Pacher

Mukhopadhyay

Lovinger

Kunos

(2009) The endogenous brain constituent N-arachidonoyl-L-serine is an activator of large conductance Ca2+-activated K+ channels. J Pharmacol Exp Ther 328:351–61

13.

Golech

McCarron

Chen

Bembry

Lenz

Mechoulam

Shohami

Spatz

(2004) Human brain endothelium: coexpression and function of vanilloid and endocannabinoid receptors. Brain Res Mol Brain Res 132:87–92

14.

Gomez del Pulgar

Velasco

Guzman

(2000) The CB1 cannabinoid receptor is coupled to the activation of protein kinase B/Akt. Biochem J 347:369–73

15.

Hanus

Breuer

Tchilibon

Shiloah

Goldenberg

Horowitz

Pertwee

Ross

Mechoulam

Fride

(1999) HU-308: a specific agonist for CB(2), a peripheral cannabinoid receptor. Proc Natl Acad Sci USA 96:14228–33

16.

Hayakawa

Mishima

Nozako

Hazekawa

Irie

Fujioka

Orito

Abe

Hasebe

Egashira

Iwasaki

Fujiwara

(2007) Delayed treatment with cannabidiol has a cerebroprotective action via a cannabinoid receptor-independent myeloperoxidase-inhibiting mechanism. J Neurochem 102:1488–96

17.

Imlach

Finch

Dunlop

Meredith

Aldrich

Dalziel

(2008) The molecular mechanism of ‘ryegrass staggers,’ a neurological disorder of K+ channels. J Pharmacol Exp Ther 327:657–64

18.

Leker

Gai

Mechoulam

Ovadia

(2003) Drug-induced hypothermia reduces ischemic damage—Effects of the cannabinoid HU-210. Stroke 34:2000–6

19.

Leker

Shohami

(2002) Cerebral ischemia and trauma-different etiologies yet similar mechanisms: neuroprotective opportunities. Brain Res Brain Res Rev 39:55–73

20.

Mechoulam

Ben-Shabat

Hanus

Ligumsky

Kaminski

Schatz

Gopher

Almog

Martin

Compton

(1995) Identification of an endogenous 2-monoglyceride, present in canine gut, that binds to cannabinoid receptors. Biochem Pharmacol 50:83–90

21.

Milman

Maor

Abu-Lafi

Horowitz

Gallily

Batkai

Offertaler

Pacher

Kunos

Mechoulam

(2006) N-arachidonoyl L-serine, an endocannabinoid-like brain constituent with vasodilatory properties. Proc Natl Acad Sci USA 103:2428–33

22.

Molina-Holgado

Vela

Arevalo-Martin

Almazan

Molina-Holgado

Borrell

Guaza

(2002) Cannabinoids promote oligodendrocyte progenitor survival: involvement of cannabinoid receptors and phosphati-dylinositol-3 kinase/Akt signaling. J Neurosci 22: 9742–53

23.

Moranta

Esteban

Garcia-Sevilla

(2007) Acute, chronic and withdrawal effects of cannabinoid receptor agonist WIN55212-2 on the sequential activation of MAPK/Raf-MEK-ERK signaling in the rat cerebral frontal cortex: short-term regulation by intrinsic and extrinsic pathways. J Neurosci Res 85:656–67

24.

Munro

Thomas

Abu-Shaar

(1993) Molecular characterization of a peripheral receptor for cannabinoids. Nature 365:61–5

25.

Ozaita

Puighermanal

Maldonado

(2007) Regulation of PI3K/Akt/GSK-3 pathway by cannabinoids in the brain. J Neurochem 102:1105–14

26.

Panikashvili

Mechoulam

Beni

Alexandrovich

Shohami

(2005) CB1 cannabinoid receptors are involved in neuroprotection via NF-kappa B inhibition. J Cereb Blood Flow Metab 25:477–84

27.

Panikashvili

Simeonidou

Ben-Shabat

Hanus

Breuer

Mechoulam

Shohami

(2001) An endogenous cannabinoid (2-AG) is neuroprotective after brain injury. Nature 13:527–31

28.

Pignataro

Meller

Inoue

Ordonez

Ashley

Xiong

Gala

Simon

(2008) In vivo and in vitro characterization of a novel neuroprotective strategy for stroke: ischemic postconditioning. J Cereb Blood Flow Metab 28:232–41

29.

Piomelli

(2003) The molecular logic of endocannabinoid signalling. Nat Rev Neurosci 4:873–84

30.

Povlishock

Katz

(2005) Update of neuropathology and neurological recovery after traumatic brain injury. J Head Trauma Rehabil 20:76–94

31.

Ryberg

Larsson

Sjogren

Hjorth

Hermansson

Leonova

Elebring

Nilsson

Drmota

Greasley

(2007) The orphan receptor GPR55 is a novel cannabinoid receptor. Br J Pharmacol 152:1092–101

32.

Sugiura

Kondo

Sukagawa

Nakane

Shinoda

Itoh

Yamashita

Waku

(1995) 2-Arachidonoyl-glycerol: a possible endogenous cannabinoid receptor ligand in brain. Biochem Biophys Res Commun 215: 89–97

33.

Swanson

Morton

Tsao-Wu

Savalos

Davidson

Sharp

(1990) A semiautomated method for measuring brain infarct volume. J Cereb Blood Flow Metab 10:290–3

34.

Tsenter

Beni-Adani

Assaf

Alexandrovich

Trembovler

Shohami

(2008) Dynamic changes in the recovery after traumatic brain injury in mice: effect of injury severity on T2-weighted MRI abnormalities, and motor and cognitive functions. J Neurotrauma 25:324–33

35.

Van der Stelt

Veldhuis

van Haaften

Fezza

Bisogno

Bar

Veldink

Vliegenthart

Di Marzo

Nicolay

(2001) Exogenous anandamide protects rat brain against acute neuronal injury in vivo. J Neurosci 21:8765–71

36.

Van Sickle

Duncan

Kingsley

Mouihate

Urbani

Mackie

Stella

Makriyannis

Piomelli

Davison

Marnett

Di Marzo

Pittman

Patel

Sharkey

(2005) Identification and functional characterization of brainstem cannabinoid CB2 receptors. Science 310:329–32

37.

Yang

Zha

Jockel

Boise

Thompson

Korsmeyer

(1995) Bad, a heterodimeric partner for Bcl-XL and Bcl-2, displaces Bax and promotes cell death. Cell 80:285–91

38.

Yune

Park

Lee

(2008) Estrogen-induced Bcl-2 expression after spinal cord injury is mediated through phosphoinositide-3-kinase/Akt-dependent CREB activation. J Neurotrauma 25:1121–31

39.

Zhang

Martin

Adler

Razdan

Ganea

Tuma

(2008) Modulation of the balance between CB(1) and CB(2) receptor activation during cerebral ischemic/reperfusion injury. Neuroscience 152:753–60

40.

Zhang

Martin

Adler

Razdan

Jallo

Tuma

(2007) Cannabinoid CB(2) receptor activation decreases cerebral infarction in a mouse focal ischemia/reperfusion model. J Cereb Blood Flow Metab 27: 1387–96

41.

Zhang

Maor

Wang

Konus

Groopman

(2010) Endocannabinoid-like N-arachidonoyl serine is a novel pro-angiogenic mediator. Br J Pharmacol 160:1583–94

42.

Zhuang

Clapham

(2004) Phosphatidylinositol 3-kinase activates ERK in primary sensory neurons and mediates inflammatory heat hyperalgesia through TRPV1 sensitization. J Neurosci 24:8300–9