Inhibition of sEH blocks inflammatory and neuropathic pain. (A) Intraspinal administration of the sEHI AEPU (n = 3–4) at low microgram amounts reduced carrageenan-elicited peripheral thermal hyperalgesia (black bars, expressed as percentage control latency) and mechanical allodynia (gray bars, expressed as percentage control threshold). *, P = 0.012; **, P = 0.003; ‡, P < 0.001 (ANOVA followed by Games–Howell post hoc). (B) The piperidine sEHI TPAU (n = 6–10) eliminated LPS (i.pl., n = 8, 10 μg)-elicited thermal hyperalgesia (BL, baseline before LPS) in a dose-dependent manner. The metabolically stable TPAU is equipotent to morphine (n = 6) but with significantly prolonged efficacy. None of the sEHIs have significant in vitro inhibitory activity on cox-1 or cox-2 (IC₅₀ > 100 μM, data not shown). (C) Spinal COX2 message is rapidly up-regulated after LPS but significantly suppressed by AEPU or TPAU administration (n = 6 per group). Quantitative RT-PCR measurements reflect fold induction compared with untreated animals in which expression level is set to a value of 1. (D) Brain tissue concentrations of AEPU (n = 4) and TPAU (n = 4) upon dermal and systemic administration, respectively. (E) TPAU and AUDA, 2 structurally different sEHIs, both reduced mechanical allodynia elicited by streptozocin-induced diabetic neuropathy (n = 6 per group). Allodynia was measured by Von Frey's test (BL, baseline withdrawal threshold before streptozocin). Thermal and mechanical withdrawal latencies were converted to percent baseline response and are shown on the y axis. Data are expressed as mean ± SEM for all figures.

EET- or sEHI-mediated antihyperalgesia occurs through 2 distinct mechanisms. Several cytochrome P450 family enzymes naturally produce EETs by oxidation of the unsaturated bonds of arachidonic acid to result in 4 regioisomers with pleiotropic biological activities. These are degraded by sEH, which introduces a water molecule opening the epoxide moieties to their corresponding diols or DHETs. The DHETs are widely assumed to be less active. EETs have little effect on the expression of the COX2 gene in normal animals but down-regulate induced COX2 possibly through an NF-κB-related pathway (11). Thus, increased EETs can mimic antiinflammatory and analgesic effects of nonsteroidal antiinflammatory drugs but as transcriptional regulators rather than enzyme inhibitors. EETs also up-regulate StARD1 gene expression in the presence of elevated cAMP levels. The StARD1 gene expression leads to an acute increase in steroid/neurosteroid synthesis, which then results in analgesia through an agonistic activity on GABA channels. This results in analgesia in both inflammatory and neuropathic pain states. Paradoxically, COX2 that is repressed by EETs is responsible for producing prostaglandins that through EP receptor activation lead to a rapid rise in intracellular cAMP levels, which appear important for EET-mediated analgesia. The dashed arrows indicate the novel, hypothesized steps in this cascade.

sEHIs cause a rapid up-regulation of spinal StARD1 expression in the presence of elevated intracellular cAMP. (A) In inflamed animals spinal StARD1 mRNA expression was briefly induced in response to peripheral inflammation elicited by LPS (n = 4, black bars), but this induction is sustained with AEPU (n = 4–5, gray bars) or TPAU (n = 4–6, white bars). *, P = 0.03; **, P < 0.0001; ♣, P = 0.04; ♦, P = 0.002; ♦♦, P = 0.013 (ANOVA followed by Games-Howell post hoc). (B) In inflamed animals brain StARD1 mRNA expression was induced only in response to LPS plus AEPU treatment [n = 6, in all groups, *, P = 0.018 (1-way ANOVA followed by Tukey's HSD post hoc)] but not by LPS or AEPU alone. (C) In noninflamed animals direct intraspinal administration of the cell-permeable cAMP analogue 8-Br cAMP (100 μg), methyl esters of EETs (5 μg), and AEPU (1 μg) in saline (with 1% DMSO) led to changes in spinal StARD1 expression after 30 min. Saline, AEPU alone, or 8-Br cAMP alone did not influence baseline StARD1 levels, but EETs alone led to a significant decrease. However, the combination of cAMP with either EETs of AEPU led to significant increases in spinal StARD1 expression [n = 4 for all groups, *, P < 0.01 (1-way ANOVA followed by Tukey's HSD post hoc)]. (D) Brain expression levels of StARD1 of animals shown in C were also monitored. In brain slices, only the spinal administration of 8-Br cAMP (100 μg) and AEPU (1 μg) led to a significant increase in brain StARD1 expression. However, it is plausible that intraspinal EETs did not reach the brain. Spinal cords and brain slices from saline-treated animals were used as calibrators for C and D.