Neuroinflammatory response to lipopolysaccharide is exacerbated in mice genetically deficient in cyclooxygenase-2

Abstract

Background: Cyclooxygenases (COX) -1 and -2 are key mediators of the inflammatory response in the central nervous system. Since COX-2 is inducible by inflammatory stimuli, it has been traditionally considered as the most appropriate target for anti-inflammatory drugs. However, the specific roles of COX-1 and COX-2 in modulating a neuroinflammatory response are unclear. Recently, we demonstrated that COX-1 deficient mice show decreased neuroinflammatory response and neuronal damage in response to lipopolysaccharide (LPS).

Methods: In this study, we investigated the role of COX-2 in the neuroinflammatory response to intracerebroventricular-injected LPS (5 mug), a model of direct activation of innate immunity, using COX-2 deficient (COX-2-/-) and wild type (COX-2+/+) mice, as well as COX-2+/+ mice pretreated for 6 weeks with celecoxib, a COX-2 selective inhibitor.

Results: Twenty-four hours after LPS injection, COX-2-/- mice showed increased neuronal damage, glial cell activation, mRNA and protein expression of markers of inflammation and oxidative stress, such as cytokines, chemokines, iNOS and NADPH oxidase. Brain protein levels of IL-1beta, NADPH oxidase subunit p67phox, and phosphorylated-signal transducer and activator of transcription 3 (STAT3) were higher in COX-2-/- and in celecoxib-treated mice, compared to COX-2+/+ mice. The increased neuroinflammatory response in COX-2-/- mice was likely mediated by the upregulation of STAT3 and suppressor of cytokine signaling 3 (SOCS3).

Conclusion: These results show that inhibiting COX-2 activity can exacerbate the inflammatory response to LPS, possibly by increasing glial cells activation and upregulating the STAT3 and SOCS3 pathways in the brain.

Figure 1

Effects of COX-2 deficiency on LPS-induced neuronal degeneration. Representative photomicrographs of FluoroJade-B staining in the hippocampus of COX-2^+/+ and COX-2^-/- mice 24 h after icv injection of LPS or vehicle. High magnification (10×) images of FluoroJade-B staining in the hippocampus are shown. The inset shows details of the neurons stained by FluoroJade-B in the hippocampus of COX-2^-/- mice 24 h after icv injection of LPS (40× magnification). Bars represent 100 μm.

Figure 2

Effects of COX-2 deficiency on LPS-induced expression of glial markers. Quantitative real time-PCR analysis of astrocyte marker GFAP (A) and microglia marker SRA mRNA (B) in COX-2^+/+and COX-2^-/- mice 24 h after icv injection of LPS or vehicle. Data are presented as mean ± SEM (n = 4-6). ***P < 0.001 compared to the corresponding vehicle-injected mice; ^#P < 0.05 compared to the LPS-injected COX-2^+/+ mice. (C) Effects of COX-2 deficiency on LPS-induced activation of microglia. Representative photomicrographs of SRA immunohistochemistry in the striatum/caudate putamen and hippocampal area for COX-2^+/+ and COX-2^-/- mice 24 h after icv injection of LPS or vehicle. Bars represent 100 μm.

Figure 3

Effects of COX-2 deficiency on LPS-induced expression of cytokines and chemokines. Quantitative real time-PCR analysis of TNF-α (A), IL-6 (B), IL-1β (C), and CCL3/MIP-1α and CCL2/MCP-1 (E) for COX-2^+/+ and COX-2^-/- mice 24 h after icv injection of LPS or vehicle. (D) ELISA-based quantification of IL-1β protein levels in the brain of COX-2^+/+ and COX-2^-/-mice 24 h after icv injection of LPS or vehicle. Data are presented as mean ± SEM (n = 4-6). ***P < 0.001 compared to the corresponding vehicle-injected mice; ^#P < 0.05, ^##P < 0.01, ^###P < 0.001 compared to the LPS-injected COX-2^+/+ mice.

Figure 4

Effects of COX-2 deficiency on LPS-induced expression of enzymes involved in the arachidonic acid cascade. Quantitative real time-PCR analysis of mPGES-1 (A) and cPLA₂ (B) for COX-2^+/+ and COX-2^-/- mice 24 h after icv injection of LPS or vehicle. (C) Quantification of COX-1 protein levels, relative to GAPDH internal loading control in the whole brain. (D) Representative immunoblot of COX-1 expression in COX-2^+/+ and COX-2^-/- mice 24 h after icv injection of LPS or vehicle. Data are presented as mean ± SEM (n = 4-6). ***P < 0.001 compared to the corresponding vehicle-injected mice; ^#P < 0.05 compared to the LPS-injected COX-2^+/+ mice.

Figure 5

Effects of COX-2 deficiency on LPS-induced expression of ROS-generating enzymes. Quantitative real time-PCR analysis of iNOS (A), gp91^phox (B) and p67^phox mRNA (C) for COX-2^+/+ and COX-2^-/- mice that received icv injection of LPS or vehicle 24 h before sacrifice. (D) Quantification of p67^phox protein levels, relative to GAPDH internal loading control in the whole brain. (E) Representative immunoblot of p67^phox expression in COX-2^+/+ and COX-2^-/- mice 24 h after icv injection of LPS or vehicle. Data are presented as mean ± SEM (n = 4-6). ***P < 0.001 compared to the corresponding vehicle-injected mice; ^#P < 0.05, ^##P < 0.01 compared to the LPS-injected COX-2^+/+ mice.

Figure 6

Effects of COX-2 deficiency on LPS-induced expression of transcription factors NF-κB and STAT3, and SOCS3. Quantitative real time-PCR analysis of NF-κB (A), SOCS3 (B) and STAT3 mRNA (C) for COX-2^+/+ and COX-2^-/- mice 24 h after icv injection of LPS or vehicle. (D) Quantification of phosphorylated STAT3 (Tyr 705) protein levels, relative to the total STAT3 in the brain nuclear fraction. (E) Representative immunoblot of P-STAT3 expression in COX-2^+/+ and COX-2^-/- mice 24 h after icv injection of LPS or vehicle. Data are presented as mean ± SEM (n = 4-6). *P < 0.05, **P < 0.01, ***P < 0.001 compared to the corresponding vehicle-injected mice; ^##P < 0.01 compared to the LPS-injected COX-2^+/+ mice.

Figure 7

Effects of chronic pretreatment with celecoxib (6000 ppm for 6 weeks) on LPS-induced brain IL-1β levels, the expression of COX-1, p67^phox NADPH oxidase subunit and the phosphorylated STAT3. (A) Representative immunoblot of COX-1 protein levels, relative to GAPDH internal loading control, in the whole brain of COX-2^+/+ and celecoxib-treated COX-2^+/+ mice 24 h after icv injection of LPS or vehicle. (B) ELISA based immunoassay for brain IL-1β in celecoxib-treated mice (n = 7) and untreated COX-2^+/+ mice (n = 4). (C) Quantification of p67^phox protein levels, relative to GAPDH internal loading control in the whole brain with representative immunoblot (D) in celecoxib-treated mice and untreated COX-2^+/+ mice (n = 4). (E) Quantification of phosphorylated STAT3 (Tyr 705) protein levels, relative to the total STAT3 in the brain nuclear fraction and the representative immunoblot (F) in celecoxib-treated mice and untreated COX-2^+/+ mice (n = 4). All data are expressed as mean ± SEM. *P <0.05, **P < 0.01, ***P < 0.001 compared to the corresponding vehicle-injected mice; ^#P < 0.05, ^###P < 0.001 compared to the LPS-injected COX-2^+/+ mice.