Intraplantar-injected ceramide in rats induces hyperalgesia through an NF-κB- and p38 kinase-dependent cyclooxygenase 2/prostaglandin E2 pathway

Abstract

Inflammatory pain represents an important unmet clinical need with important socioeconomic implications. Ceramide, a potent proinflammatory sphingolipid, has been shown to elicit mechanical hyperalgesia, but the mechanisms remain largely unknown. We now demonstrate that, in addition to mechanical hyperalgesia, intraplantar injection of ceramide (10 μg) led to the development of thermal hyperalgesia that was dependent on induction of the inducible cyclooxygenase (COX-2) and subsequent increase of prostaglandin E(2) (PGE(2)). The development of mechanical and thermal hyperalgesia and increased production of PGE(2) was blocked by NS-398 (15-150 ng), a selective COX-2 inhibitor. The importance of the COX-2 to PGE(2) pathway in ceramide signaling was underscored by the findings that intraplantar injection of a monoclonal PGE(2) antibody (4 μg) blocked the development of hyperalgesia. Our results further revealed that COX-2 induction is regulated by NF-κB and p38 kinase activation, since intraplantar injection of SC-514 (0.1-1 μg) or SB 203580 (1-10 μg), well-characterized inhibitors of NF-κB and p38 kinase activation, respectively, blocked COX-2 induction and increased formation of PGE(2) and thermal hyperalgesia in a dose-dependent manner. Moreover, activation of NF-κB was dependent on upstream activation of p38 MAPK, since SB 203580 (10 μg) blocked p65 phosphorylation, whereas p38 kinase phosphorylation was unaffected by NF-κB inhibition by SC-514 (1 μg). Our findings not only provide mechanistic insight into the signaling pathways engaged by ceramide in the development of hyperalgesia, but also provide a potential pharmacological basis for developing inhibitors targeting the ceramide metabolic-to-COX-2 pathway as novel analgesics.

Figure 1.

Ceramide-induced hyperalgesia. When compared to vehicle (Veh, ▵), intraplantar injection of 10 μg C₂-ceramide (●), but not dihydro-C₂-ceramide (◊), led to a time-dependent development of mechanical (A) and thermal hyperalgesia (B). Results are expressed as means ± se for 4 rats. BL, baseline. *P < 0.001 vs. Veh; 2-way ANOVA with Bonferroni post hoc tests.

Figure 2.

Role of COX-2 and PGE₂ in ceramide-induced hyperalgesia. A, B) When compared to baseline values at t = 0 h, intraplantar injection of C₂-ceramide (10 μg) led to a significant increase in COX-2 (A), but not COX-1 (B), expression by 2 h, which was sustained through 5 h, as measured by Western blot analysis. Top panels: bands are representative replicates of separate animals (n=3 rats) taken from the same gel and exposure. Bottom panels: band densities were normalized to α-tubulin (%COX-1 or -2/tubulin). C) Increased COX-2 corresponded with a significant increase in production of PGE₂ at 2 h, as measured in paw fluids (n=4 rats). Results are expressed as means ± se. *P < 0.05, **P < 0.01, ***P < 0.001 vs. t = 0 h; ANOVA with Dunnett's post hoc test (A, B) or unpaired Student's t test (C).

Figure 3.

Inhibition of ceramide-induced hyperalgesia by NS-398. A, B) When compared to rats administered intraplantar NS-398 vehicle and ceramide vehicle (Veh; ▵), an intraplantar injection of ceramide (Cer; 10 μg, ◊) led to a time-dependent development of mechanical (A) and thermal hyperalgesia (B) that was attenuated in a dose-dependent manner by intraplantar NS-398 given at 15 ng (●), 45 ng (■), or 150 ng (♦). Intraplantar administration of NS-398, when tested at the highest dose and compared to rats that received vehicle for ceramide, had no effect on baseline (BL) withdrawal latency (○). C) When tested at the highest dose, NS-398 (150 ng) attenuated the increased production of PGE₂ in paw tissue fluids, as measured at t =2 h. D, E) When compared to rats administered ceramide vehicle (▵), an intraplantar injection of ceramide (10 μg; ◊) led to a time-dependent development of mechanical (D) and thermal hyperalgesia (E) that was attenuated by intraplantar injection of a monoclonal antibody to PGE₂ (▾). Anti-PGE₂ antibody alone had no effect on baseline withdrawal latency (▿). Results are expressed as means ± se (n=4 rats). *P < 0.001 vs. Veh; ^†P < 0.05, ^††P < 0.01, ^†††P < 0.001 vs. Cer; 2-way ANOVA with Bonferroni post hoc test (A, B, D, E) or 1-way ANOVA with Dunnett's post hoc test (C).

Figure 4.

Role of NF-κB in ceramide-induced hyperalgesia. A, B) When compared to rats administered intraplantar SC-514 vehicle and ceramide vehicle (Veh; ▵), an intraplantar injection of ceramide (Cer; 10 μg, ◊) led to a time-dependent development of mechanical (A) and thermal (B) hyperalgesia that was attenuated in a dose-dependent manner by intraplantar SC-514 given at 0.1 μg (●), 0.3 μg (■), or 1 μg (♦). Intraplantar administration of SC-514, when tested at the highest dose and compared to rats that received vehicle for ceramide, had no effect on baseline (BL) withdrawal latency (○). C, D) When tested at the highest dose, SC-514 (1 μg) blocked the induction of COX-2 (C), and the increased production of PGE₂ (D), as measured at t = 2 h. Results are expressed as means ± se (n=4 rats; A–C). Band densities (C, bottom panel) were normalized to α-tubulin (%COX-2/tubulin). *P < 0.001 vs. Veh; ^†P < 0.05, ^††P < 0.01, ^†††P < 0.001 vs. Cer; 2-way ANOVA with Bonferroni post hoc test (A, B) or 1-way ANOVA with Dunnett's post hoc test (C, D).

Figure 5.

Role of p38 in ceramide-induced hyperalgesia. A, B) When compared to rats administered intraplantar SB 203580 vehicle and ceramide vehicle (Veh; ▵), an intraplantar injection of ceramide (Cer; 10 μg, ◊) led to a time-dependent development of mechanical (A) and thermal hyperalgesia (B) that was attenuated in a dose-dependent manner by intraplantar SB 203580 given at 1 μg (●), 3 μg (■), or 10 μg (♦). Intraplantar administration of SB 203580, when tested at the highest dose and compared to rats that received vehicle for ceramide, had no effect on baseline withdrawal latency (○). C, D) When tested at the highest dose, SB 203580 (10 μg) blocked the induction of COX-2 (C) and the increased production of PGE₂ (D) as measured at t = 2 h. Results are expressed as means ± se (n=4 rats). Gels (C, top panel) are replicates from separate animals representative of n = 4 rats. Band densities (C, bottom panel) were normalized to α-tubulin (%COX-2/tubulin). *P < 0.01, **P < 0.001 vs. Veh; ^†P < 0.05, ^††P < 0.01, ^†††P < 0.001 vs. Cer; 2-way ANOVA with Bonferroni post hoc test (A, B) or 1-way ANOVA with Dunnett's post hoc test (C, D).

Figure 6.

Ceramide-induced p65 phosphorylation is dependent on p38 MAPK activation. When compared with rats administered intraplantar vehicles for the inhibitors SB 203580 or SC-514 and for ceramide (Veh), an intraplantar injection of ceramide (Cer, 10 μg) increased the p-p38 (T180/Y182; A) and p-p65 (S536; B) at 2 h. When tested at the highest dose, SB 203580 (Cer+SB 203580; 10 μg) blocked ceramide-induced p65 phosphorylation (B). In contrast, SC-514 (Cer + SC-514; 1 μg) did not block ceramide-induced p38 phosphorylation (A). There were no significant differences in total p38 or p65 between groups at 2 h. Top panels: gels are replicates from n = 5 rats for p38 and n = 6 rats for p65. Bottom panels: band densities were normalized to α-tubulin (%p-p38/tubulin or %p-p65/tubulin) and expressed as means ± se. *P < 0.05, **P < 0.001 vs. Veh; ^†P < 0.05 vs. Cer; 1-way ANOVA with Dunnett's post hoc test.

Figure 7.

Cartoon summarizing the key findings of this study. Intraplantar injection of ceramide is a potent hyperalgesic sphingolipid acting, at least in part, via COX-2 induction through activation of a p38 MAPK-dependent NF-κB pathway. Whether ceramide acts alone or in concert with one of its active metabolites (such as sphingosine-1-phosphate) is certainly a possibility, but not a requirement, since it is well established that ceramide does not need to be metabolized in order to exert its biological responses (1). Inhibiting this pathway with p38 MAPK inhibitor, SB 203580, or the IKK-2 inhibitor, SC-514, prevents the increased expression of COX-2 and the subsequent formation of PGE₂, which are essential to the development of ceramide-induced hyperalgesia. Dual inhibition of ceramide formation and COX inhibitors may provide a novel therapeutic approach to the management of pain: effective analgesia with reduced side effects typically associated with the use of COX inhibitors.