Potent anti-inflammatory effects of two quinolinedione compounds, OQ1 and OQ21, mediated by dual inhibition of inducible NO synthase and cyclooxygenase-2

Abstract

Background and purpose: Inducible nitric oxide synthase (iNOS) and cyclooxygenase-2 (COX-2) have been suggested as key components in various inflammatory diseases. Here we examined the effects of new quinolinedione derivatives, 6-(4-fluorophenyl)-amino-5,8-quinolinedione (OQ1) and 6-(2,3,4-trifluorophenyl)-amino-5,8-quinolinedione (OQ21) on activity and expression of iNOS and COX-2 to explore their anti-inflammatory properties.

Experimental approach: The effects of OQ1 and OQ21 were assessed on lipopolysaccharide (LPS)-induced iNOS and COX-2 in murine macrophage cell line (RAW264.7), along with isolated enzyme assays to measure enzyme inhibition. Nuclear factor-kappaB (NFkappaB) activation pathways were investigated to elucidate mechanisms underlying OQ-mediated suppression of the expression of iNOS and COX-2. In vivo anti-inflammatory activities of OQ compounds were evaluated in mouse ear oedema, induced by topical 12-O-tetradecanoylphorbol-13-acetate (TPA).

Key results: LPS-induced NO production in RAW264.7 cells was inhibited by OQ1 and OQ21 through the attenuation of iNOS expression as well as iNOS activity. Down-regulation of iNOS followed blocking of NFkappaB activation, as assessed by inhibitory kappaB degradation and electrophoretic mobility shift assay for NFkappaB. Synthesis and accumulation of prostaglandin E(2) were also suppressed by OQ1 and OQ21. LPS-induced COX-2 expression and cellular COX-2 activities were attenuated by OQ1 and OQ21. Consistent with these results, OQ1 showed potent anti-inflammatory effects in mouse ear oedema induced by TPA.

Conclusions and implications: The novel quinolinedione derivatives, OQ1 and OQ21, showed potent anti-inflammatory activity through dual inhibitory effects on iNOS and COX-2, suggesting that OQ derivatives might provide a new therapeutic modality for chronic inflammatory diseases, refractory to conventional drug therapies.

Figure 1

The suppressive effects of OQ1 and OQ21 on production of nitric oxide (NO) and enzyme activity of iNOS. A. RAW264.7 cells were incubated with LPS (0.1 µg·mL⁻¹) in the presence of OQ1 or OQ21 (1, 5, 10 and 25 µmol·L⁻¹) for 24 h. Accumulated nitrite was measured using Griess reagent. B. RAW264.7 cells were pre-incubated with LPS (0.1 µg·mL⁻¹) for 16 h to induce iNOS. After cells were washed with fresh media to remove LPS, cells were further treated with OQ1 or OQ21 for 8 h and nitrite generation was determined. C. Purified murine iNOS was incubated with OQ1 or OQ21 for 5 min and iNOS activity was measured by NADPH-initiated conversion of oxyHb to metHb. Values are means ± SEM (n = 4–5). * represents a significant difference from LPS-treated control by one-way anova followed by Duncan's multiple range test (P < 0.05). OQ1, 6-(4-fluorophenyl)-amino-5,8-quinolinedione; OQ21, 6-(2,3,4-trifluorophenyl)-amino-5,8-quinolinedione; iNOS, inducible nitric oxide synthase; LPS, lipopolysaccharide.

Figure 2

Suppression of iNOS expression and upstream signalling pathways by OQ1 and OQ21 in LPS-stimulated macrophages. A. RAW264.7 cells were incubated with LPS (0.1 µg·mL⁻¹) and various concentrations of OQ1 or OQ21 for 8 h. iNOS expression was determined by immunoblot assay with α-tubulin as a loading control. B. LPS-induced activation of NFκB was determined by electrophoretic mobility shift assay from nuclear extracts prepared from the cells stimulated with LPS (0.1 µg·mL⁻¹) in the presence of OQ1 or OQ21 (25 µmol·L⁻¹) for 3 h. Competition experiments were performed by adding 20-fold excess amount of unlabelled oligonucleotides containing the consensus binding sequence for NFκB or SP-1. C and D. RAW264.7 cells incubated with LPS (0.1 µg·mL⁻¹) and various concentrations of OQ1 or OQ21 for 30 min. Cell lysates were analysed for IκBα, phospho-Erk, and phospho-p38 by immunoblot assay. Representative results are shown from more than three separate experiments for each assay. OQ1, 6-(4-fluorophenyl)-amino-5,8-quinolinedione; OQ21, 6-(2,3,4-trifluorophenyl)-amino-5,8-quinolinedione; iNOS, inducible nitric oxide synthase; LPS, lipopolysaccharide; NFκB, nuclear factor-κB.

Figure 3

The effects of OQ1 and OQ21 on LPS-induced PGE₂ production and enzyme activity of COX-2. A. RAW264.7 cells were incubated with LPS (0.1 µg·mL⁻¹) and various concentrations of OQ1 or OQ21 for 8 h. The levels of accumulated PGE₂ were determined in the medium by PGE₂ EIA. B. RAW264.7 cells were stimulated with LPS for 8 h to induce COX-2 and washed with fresh media to remove LPS. Cells were further treated with OQ1 or OQ21 for 30 min followed by the addition of arachidonic acid (30 µmol·L⁻¹) as a substrate for COX-2. After 15 min, the levels of PGE₂ in media were determined using PGE₂ EIA. C. After purified COX-1/-2 enzymes were incubated with OQ1, OQ21 or indomethacin (IND) for 30 min, COX activities were measured by colorimetric assessment. Values are means ± SEM (n = 4–5). * represents a significant difference from LPS-treated control by one-way anova followed by Duncan's multiple range test (P < 0.05). OQ1, 6-(4-fluorophenyl)-amino-5,8-quinolinedione; OQ21, 6-(2,3,4-trifluorophenyl)-amino-5,8-quinolinedione; LPS, lipopolysaccharide; PGE₂, prostaglandin E₂; COX, cyclooxygenase.

Figure 4

Effects of OQ1 or OQ21 on COX-2 expression in LPS-stimulated macrophages. A. RAW264.7 cells were incubated with LPS (0.1 µg·mL⁻¹) and various concentrations of OQ1 or OQ21 for 8 h. COX-2 expression was determined by immunoblot assay with α-tubulin as a loading control. Representative results from more than three separate experiments are shown. B. RAW264.7 cells were transfected with murine COX-2 promoter (−3.2 kb) luciferase reporter plasmid and β-galactosidase expression plasmid. Cells were further treated with LPS (0.1 µg·mL⁻¹) and various concentrations of OQ1 or OQ21 for 8 h. Cell lysates were analysed for luciferase and β-galactosidase enzyme activities. Fold induction was calculated after normalization with β-galactosidase activity. Values are mean ± SEM (n = 4). * represents a significant difference from LPS-treated group by one-way anova followed by Duncan's multiple range test (P < 0.05). OQ1, 6-(4-fluorophenyl)-amino-5,8-quinolinedione; OQ21, 6-(2,3,4-trifluorophenyl)-amino-5,8-quinolinedione; LPS, lipopolysaccharide; COX-2, cyclooxygenase-2.

Figure 5

Anti-inflammatory activity of OQ1 in TPA-induced ear oedema mouse model. A. Ear oedema was induced with topical application of TPA (2.5 µg per ear). OQ1, OQ21 or indomethacin (IND; 0.1 mg per ear, 10 animals per group) was painted on the ear pad, 30 min and 6 h after TPA treatment. Twenty-four hours after TPA application, ear biopsies were collected and weighed to determine the severity of oedema formation. Values are means ± SEM (n = 10). * represents a significant difference from TPA-treated group by one-way anova followed by Duncan's multiple range test (P < 0.05). B. Five ear biopsies were randomly selected per group and processed (haematoxylin and eosin staining) for histological examinations. Tailless arrowhead represents inflammatory cells and tailed arrow, hyperkeratosis. Representative microscopic photographs are presented. C. Histopathological changes were scored for dermis and epidermis. Epidermal changes were evaluated with the scores for thickness (0–5), stratum granulosum (0–3), hyperkeratosis (0–5), spongiosis (0–4) and intracellular oedema (0–1). Dermal changes were scored for infiltration (0–6). Total scores for the respective ear are presented. * represents a significant difference from TPA-treated group by the Mann–Whitney U tests (P < 0.05). OQ1, 6-(4-fluorophenyl)-amino-5,8-quinolinedione; OQ21, 6-(2,3,4-trifluorophenyl)-amino-5,8-quinolinedione; TPA, 12-O-tetradecanoylphorbol-13-acetate.