13-Series resolvins mediate the leukocyte-platelet actions of atorvastatin and pravastatin in inflammatory arthritis

Abstract

Rheumatoid arthritis is an inflammatory condition characterized by overzealous inflammation that leads to joint damage and is associated with an increased incidence of cardiovascular disease. Statins are frontline therapeutics for patients with cardiovascular disease and exert beneficial actions in rheumatoid arthritis. The mechanism that mediates the beneficial actions of statins in rheumatoid arthritis remains of interest. In the present study, we found that the administration of 2 clinically relevant statins-atorvastatin (0.2 mg/kg) or pravastatin (0.2 mg/kg)-to mice during inflammatory arthritis up-regulated systemic and tissue amounts of a novel family of proresolving mediators, termed 13-series resolvins (RvTs), and significantly reduced joint disease. Of note, administration of simvastatin (0.2 mg/kg) did not significantly up-regulate RvTs or reduce joint inflammation. We also found that atorvastatin and pravastatin each reduced systemic leukocyte activation, including platelet-monocyte aggregates (∼25-60%). These statins decreased neutrophil trafficking to the joint as well as joint monocyte and macrophage numbers. Atorvastatin and pravastatin produced significant reductions (∼30-50%) in expression of CD11b and major histocompatibility complex class II on both monocytes and monocyte-derived macrophages in joints. Administration of an inhibitor to cyclooxygenase-2, the initiating enzyme in the RvT pathway, reversed the protective actions of these statins on both joint and systemic inflammation. Together, these findings provide evidence for the role of RvTs in mediating the protective actions of atorvastatin and pravastatin in reducing local and vascular inflammation, and suggest that RvTs may be useful in measuring the anti-inflammatory actions of statins.-Walker, M. E., Souza, P. R., Colas, R. A., Dalli, J. 13-Series resolvins mediate the leukocyte-platelet actions of atorvastatin and pravastatin in inflammatory arthritis.

Keywords: eicosanoids; pharmacology; resolution; vascular inflammation; ω-3.

Figure 1.

Increased RvTs in paws from mice that were administered atorvastatin and pravastatin during inflammatory arthritis. A) Arthritogenic K/BxN serum (100 μl, i.p.) was administered to mice to initiate disease, and disease progression was monitored daily using clinical scores. Arrows denote days when mice were given statins or vehicle (see Materials and Methods for details). B–D) Doses of 0.2 mg/kg atorvastatin, pravastatin, simvastatin, or vehicle (DPBS that contained 0.05% ethanol) were administered intravenously on d 3, 5, and 7 after disease onset. Paws were collected on d 8, and lipid mediators were identified and quantified by using lipid mediator profiling. B) Representative multiple reaction monitoring chromatograms of identified lipid mediators derived from docosahexaenoic acid, n-3 docosapentaenoic acid, eicosapentaenoic acid, and arachidonic acid. C) Tandem mass spectrometry spectra employed in the identification of RvT2 and RvT3. D) Percentage regulation of RvT1, RvT2, RvT3, and RvT4 compared with vehicle (D). Results are means ± sem; n = 9 for vehicle, 11 for atorvastatin, 11 for pravastatin, and 9 for simvastatin-treated mice from 4 independent experiments. *P < 0.05 vs. vehicle.

Figure 2.

Atorvastatin and pravastatin reduced disease severity and protected joint architecture. Arthritogenic K/BxN serum was administered to mice on d 0 and 2. A–C) Disease progression was monitored by using a 26-point clinical score in mice given atorvastatin (0.2 mg/kg; A), pravastatin (0.2 mg/kg; B), simvastatin (0.2 mg/kg; C), or vehicle (DPBS that contained 0.05% ethanol) on d 3, 5, and 7. Results are means ± sem; n = 8 for vehicle, 10 for atorvastatin, 10 for pravastatin, and 6 for simvastatin-treated mice from 3 independent experiments. D) Maximum percentage increase in midfoot pad thickness. Results are means ± sem; n = 8 for vehicle, 10 for atorvastatin, 10 for pravastatin, and 6 for simvastatin-treated mice from 3 independent experiments. E) Representative hematoxylin and eosin–stained knee sections of mice collected on d 8 using EVOS FL imaging system. Original magnification, ×4. Results are representative of n = 8 for vehicle, 10 for atorvastatin, 10 for pravastatin, and 6 for simvastatin-treated mice from 3 independent experiments. Arrows denote leukocyte infiltration. F, femur; IFP, infrapatellar fat pad; m, meniscus; PF, pannus formation; T, tibia. *P < 0.05 vs. vehicle using 1-way ANOVA with post hoc Dunnett’s multiple comparisons test; **P < 0.01, ***P < 0.05 vs. vehicle using ordinary 2-way ANOVA.

Figure 3.

Differential regulation of circulating leukocyte and platelet activation by each of the statins in inflammatory arthritis. Serum-induced arthritis was initiated in mice and atorvastatin, pravastatin, simvastatin (0.2 mg/kg each), or vehicle (DPBS that contained 0.05% ethanol) were administered on d 3, 5, and 7. On d 8, blood was collected. Leukocyte subsets and activation were identified by using fluorescently labeled Abs and flow cytometry. Activation markers on nonclassic monocytes (A), classic monocytes (B), neutrophils (C), and platelets (D) were assessed as percentage decrease from vehicle. Results are means ± sem; n = 9 for vehicle, 11 for atorvastatin, 11 for pravastatin, and 9 for simvastatin-treated mice from 4 independent experiments. *P < 0.05 vs. vehicle using 1-way ANOVA with post hoc Dunnett’s multiple comparisons test.

Figure 4.

Reduction of monocyte, neutrophil, and macrophage activation, as well as trafficking to the joint, by atorvastatin and pravastatin in inflammatory arthritis. Serum-induced arthritis was initiated in mice and atorvastatin, pravastatin, simvastatin (0.2 mg/kg each), or vehicle (DPBS that contained 0.05% ethanol) were administered on d 3, 5, and 7. Front paws were collected on d 8 and digested to liberate infiltrating leukocytes. Leukocyte subsets were defined by using Abs against specific markers and flow cytometry. Trafficking and activation of nonclassic monocytes (A), neutrophils (B), and monocyte-derived macrophages (C) were assessed. Results are means ± sem; n = 9 for vehicle, 11 for atorvastatin, 11 for pravastatin, and 9 for simvastatin-treated mice from 4 independent experiments. *P < 0.05 vs. vehicle using 1-way ANOVA with post hoc Dunnett’s multiple comparisons test.

Figure 5.

Inhibition of RvT production by celecoxib reverses the joint protective actions of atorvastatin and pravastatin. Inflammatory arthritis was initiated by using arthritogenic serum (see Materials and Methods for details). A, B) On d 3, 5, and 7, mice were administered celecoxib (10 mg/kg) or vehicle (DPBS that contained 0.05% ethanol) and after 1 h given atorvastatin (0.2 mg/kg; A), pravastatin (0.2 mg/kg; B), or vehicle (PBS that contained 0.05% ethanol). Disease activity was assessed daily. C) On d 8, paws were collected, and RvTs were identified and quantified by using liquid chromatography–tandem mass spectrometry–based lipid mediator profiling. *P < 0.05; **P < 0.01 vs. atorvastatin or pravastatin alone using 1-way ANOVA with post hoc Sidak’s multiple comparisons test. D) Representative hematoxylin and eosin–stained knee sections of mice collected on d 8 using EVOS FL imaging system. Original magnification, ×4. Results are means ± sem; n = 9 for vehicle, 11 for atorvastatin, 11 for pravastatin, 7 for celecoxib plus atorvastatin, and 6 for celecoxib plus pravastatin-treated mice per group from 2–3 independent experiments. *P < 0.05 vs. vehicle using ordinary 2-way ANOVA.

Figure 6.

COX-2 inhibition reverses the protective actions of atorvastatin on both systemic and joint leukocytes. Serum-induced arthritis was initiated on d 3, 5, and 7 and mice were administered celecoxib (10 mg/kg) or vehicle (DPBS that contained 0.05% ethanol) and, after 1 h, given atorvastatin (0.2 mg/kg) or vehicle (DPBS that contained 0.05% ethanol). Blood was collected on d 8 and leukocyte subsets and activation were identified by using fluorescently labeled Abs and flow cytometry. A–C) Activation markers on circulating nonclassic monocytes (A), neutrophils (B), and platelets (C). Results are presented as percentage decrease from vehicle. D–F) Leukocytes were recovered from the inflamed paws (see Materials and Methods for details) on d 8. Trafficking and activation profile for nonclassic monocytes (D) neutrophils (E), and monocyte-derived macrophages (F) were assessed by using flow cytometry. Results are means ± sem; n = 9 for vehicle, 11 for atorvastatin, 11 for pravastatin, and 7 for celecoxib plus atorvastatin-treated mice from 2 independent experiments. *P < 0.05 vs. vehicle, ^#P < 0.05 vs. atorvastatin using 1-way ANOVA with post hoc Dunnett’s multiple comparisons test.