Effects of COX-2 inhibition on spinal nociception: the role of endocannabinoids

Abstract

Background and purpose: Recent studies suggest that the effects of cyclooxygenase-2 (COX-2) inhibition are mediated by cannabinoid receptor activation. However, some non-steroidal anti-inflammatory drugs inhibit the enzyme fatty acid amide hydrolase, which regulates levels of some endocannabinoids. Whether COX-2 directly regulates levels of endocannabinoids in vivo is unclear. Here, the effect of the COX-2 inhibitor nimesulide, which does not inhibit fatty acid amide hydrolase, on spinal nociceptive processing was determined. Effects of nimesulide on tissue levels of endocannabinoids and related compounds were measured and the role of cannabinoid 1 (CB(1)) receptors was determined.

Experimental approach: Effects of spinal and peripheral administration of nimesulide (1-100 microg per 50 microL) on mechanically evoked responses of rat dorsal horn neurones were measured, and the contribution of the CB(1) receptor was determined with the antagonist AM251 (N-(piperidin-1-yl)-5-(-4-iodophenyl)-1-(2,4-dichlorophenyl)-4-methyl-1H-pyrazole-3-carboxamide), in anaesthetized rats. Effects of nimesulide on spinal levels of endocannabinoids and related compounds were quantified using liquid chromatography-tandem mass spectrometry.

Key results: Spinal, but not peripheral, injection of nimesulide (1-100 microg per 50 microL) significantly reduced mechanically evoked responses of dorsal horn neurones. Inhibitory effects of spinal nimesulide were blocked by the CB(1) receptor antagonist AM251 (1 microg per 50 microL), but spinal levels of endocannabinoids were not elevated. Indeed, both anandamide and N-oleoylethanolamide (OEA) were significantly decreased by nimesulide.

Conclusions and implications: Although the inhibitory effects of COX-2 blockade on spinal neuronal responses by nimesulide were dependent on CB(1) receptors, we did not detect a concomitant elevation in anandamide or 2-AG. Further understanding of the complexities of endocannabinoid catabolism by multiple enzymes is essential to understand their contribution to COX-2-mediated analgesia.

Figure 2

Example trace of mechanically evoked responses of a single wide dynamic range dorsal horn neurone in a naïve anaesthetized rat before (pre-drug response) and after spinal administration of nimesulide (25 µg per 50 µL).

Figure 1

Mean maximal effects of spinal nimesulide on mechanically evoked (8–60 g) (A) non-noxious (8 g and 10 g) and (B) noxious (15 g, 26 g and 60 g) responses of wide dynamic range dorsal horn neurones in naive anaesthetized rats in vivo (n= 6 rats per dose, total n= 18 rats). Nimesulide reduced mechanically evoked responses in a dose-dependent manner. Statistical analyses were performed with one-way anova (Kruskal-Wallis) with post hoc Dunn's test; single symbol (#, $, &)P < 0.05; double symbol (**, ##, ++, $$) P < 0.01 versus vehicle (not shown, no significant difference to pre-drug controls). Data are expressed as a percentage of the pre-drug control ± SEM.

Figure 4

Spinal pretreatment with the CB₁ receptor antagonist AM251 (N-(piperidin-1-yl)-5-(-4-iodophenyl)-1-(2,4-dichlorophenyl)-4-methyl-1H-pyrazole-3-carboxamide) (1 µg per 50 µL) blocked the inhibitory effects of nimesulide (25 µg per 50 µL) on mechanically evoked (8–60 g) responses of wide dynamic range dorsal horn neurones in naïve anaesthetized rats in vivo (n= 6). Statistical analyses were performed with non-parametric Mann–Whitney test; *P < 0.05; **P < 0.01 versus vehicle; ##P < 0.01 versus 25 µg nimesulide. Data are expressed as a percentage of the pre-drug control ± SEM.

Figure 3

The CB₁ antagonist AM251 (N-(piperidin-1-yl)-5-(-4-iodophenyl)-1-(2,4-dichlorophenyl)-4-methyl-1H-pyrazole-3-carboxamide) (1 µg per 50 µL) alone did not alter mechanically evoked (8–60 g) responses of dorsal horn wide dynamic range neurones in naïve anaesthetized rats (n= 6). Data are expressed as a percentage of the pre-drug control ± SEM.

Figure 5

Effects of spinal nimesulide on levels of anandamide (AEA), N-oleoylethanolamine (OEA) and 2-arachidonoylglycerol (2-AG) in spinal cord of naïve anaesthetized rats. Two analytical runs were conducted; effects of nimesulide 25 µg per 50 µL versus 50 µL vehicle (n= 8 samples per group) were analysed in one LC/MS run, followed by nimesulide 100 µg per 50 µL versus 50 µL vehicle (n= 6 samples per group) in a second analytical run. Statistical analyses were performed with non-parametric Mann–Whitney test; **P < 0.01, ***P < 0.005 nimesulide versus vehicle. Data are expressed as individual values and median value is depicted by the line.