Palmitoylethanolamide increases after focal cerebral ischemia and potentiates microglial cell motility

Abstract

Focal cerebral ischemia (FCI) induces rapid neuronal death in the ischemic core, which gradually expands toward the penumbra, partly as the result of a neuroinflammatory response. It is known that propagation of neuroinflammation involves microglial cells, the resident macrophages of the brain, which are highly motile when activated by specific signals. However, the signals that increase microglial cell motility in response to FCI remain mostly elusive. Here, we tested the hypothesis that endocannabinoids mediate neuroinflammation propagation by increasing microglial cell motility. We found that, in mouse cerebral cortex, FCI greatly increases palmitoylethanolamide (PEA), only moderately increases anandamide [arachidonylethanolamide (AEA)], and does not affect 2-arachidonoylglycerol levels. We also found that PEA potentiates AEA-induced microglial cell migration, without affecting other steps of microglial activation, such as proliferation, particle engulfment, and nitric oxide production. This potentiation of microglial cell migration by PEA involves reduction in cAMP levels. In line with this, we provide evidence that PEA acts through Gi/o-coupled receptors. Interestingly, these receptors engaged by PEA are pharmacologically distinct from CB1 and CB2 cannabinoid receptors, as well as from the WIN and abn-CBD (abnormal-cannabidiol) receptors, two recently identified cannabinoid receptors. Our results show that PEA and AEA increase after FCI and synergistically enhance microglial cell motility. Because such a response could participate in the propagation of the FCI-induced neuroinflammation within the CNS, and because PEA is likely to act through its own receptor, a better understanding of the receptor engaged by PEA may help guide the search for improved therapies against neuroinflammation.

Figure 4.

LPS/IFNγ increases the ability of BV-2 cells to engulf particles. Cells were incubated with mouse IgG-opsonized beads (1 μm) for 2 hr and analyzed by FACS. Unbound beads were too small to be detected. Shown are representative dot plots of forward scatter versus fluorescence intensity for microglial cells incubated with beads at either 4°C (a) or 37°C (b). After analyzing the signal obtained at 4°C, we discounted >95% of the nonspecific signal by gating in R1. BV-2 cells were recovered from R1-FACS and plated onto coverslips, fixed, and labeled with phalloidin (red) to stain actin (c). Engulfed fluorescent latex beads appear yellow-green (scale bar, 50 μm). Inset shows higher magnification of one BV-2 cell that had engulfed eight beads, which was counterstained with 4′,6′-diamidino-2-phenylindole (blue) to label the nucleus (scale bar, 20 μm). Fluorescence intensity histograms of BV-2 cells at 37°C pretreated for 18 hr with either vehicle (d) or IFN-γ (100 U) plus LPS (10 μg/ml) (e) and incubated with beads for 2 hr. Also shown is the cutoff for engulfing versus nonengulfing cells (dotted line).

Figure 1.

Endocannabinoid levels in cerebral cortex after focal cerebral ischemia. Mice underwent either sham surgery (Sham) or 20 min of FCI (Ischemia) or were directly decapitated (Control), as described in Materials and Methods. AEA, 2-AG, PEA, HEA, and DEA in cerebral cortex was quantified by CI-GC/MS. Values shown are the means ± SEMs of quantities measured in n = 5 animals per group. ^*p < 0.05 and ^**p < 0.01 compared with sham (ANOVA followed by Dunnett's post-test).

Figure 2.

PEA and synthetic cannabinoids inhibit the forskolin-stimulated accumulation of cAMP in BV-2 cells. BV-2 cells were preincubated for 10 min with IBMX (1 mm) and either vehicle (0.1% DMSO), CP-55940 (1 μm), or PEA (0.3 μm). Cells were then incubated for an additional 10 min with the same agents plus forskolin (FSK) (100 μm). During both preincubation and incubation, SR141716A (SR1) (0.3 μm), SR144528 (SR2) (0.3 μm), capsazepine (CPZ) (3 μm), or O-1918 (1 μm) was present. To test for G_i/o-protein involvement, cells were pretreated for 18 hr with 1 μg/ml pertussis toxin (PTX). Results are means ± SEMs of 9-45 independent quantifications from 3 to 15 separate experiments performed in triplicate. ^*p < 0.05 and ^**p < 0.01 compared with FSK alone (ANOVA followed by Dunnett's post-test). Basal levels of cAMP were 24.7 ± 3.5 fmol/well, which increased to 87.3 ± 9.8 fmol/well in the presence of FSK (n = 54). Horizontal dotted lines correspond to 100% forskolin response.

Figure 3.

PEA selectively potentiates the anandamide-induced BV-2 cell migration. Control (0.1% DMSO), PEA (300 nm), db-cAMP (1 mm), or PEA plus db-cAMP were added to the lower compartment of the Boyden chamber in the absence (Basal) or presence of AEA (100 nm) or 2-AG (100 nm). BV-2 cell migration toward these ligands was quantified, and results are expressed as a percentage of basal/control BV-2 cell migration. Results are means ± SEMs of 9-45 independent quantifications from 3 to 15 separate experiments performed in triplicate. ^*p < 0.05 and ^**p < 0.01 compared with basal/control migration (ANOVA followed by Dunnett's post-test). The dotted horizontal line corresponds to control basal migration.