12/15-lipoxygenase activity promotes efficient inflammation resolution in a murine model of Lyme arthritis

Abstract

Infection of C3H/HeJ (C3H) mice with Borrelia burgdorferi results in the development of a robust inflammatory arthritis that peaks around 3-4 weeks post-infection and then spontaneously resolves over the next few weeks. Mice lacking cyclooxygenase (COX)-2 or 5-lipoxygenase (5-LO) activity develop arthritis similar to wild-type mice but display delayed or prolonged joint resolution. Since 12/15-lipoxygenase (12/15-LO) activity is generally down-stream of both COX-2 and 5-LO activity and results in the production of pro-resolution lipids such as lipoxins and resolvins among others, we investigated the impact of 12/15-LO deficiency on the resolution of Lyme arthritis in mice on a C3H background. We found the expression of Alox15 (12/15-LO gene) peaked around 4-weeks post-infection in C3H mice suggesting a role for 12/15-LO in mediating arthritis resolution. A deficiency in 12/15-LO resulted in exacerbated ankle swelling and arthritis severity during the resolution phase without compromising anti-Borrelia antibody production and spirochete clearance. However, clearance of inflammatory cells was impeded. Therapeutic treatment of B. burgdorferi-infected C3H mice with lipoxin A4 (LXA₄) near the peak of disease resulted in significantly decreased ankle swelling and a switch of joint macrophages to a resolving phenotype but did not directly impact arthritis severity. These results demonstrate that 12/15-LO lipid metabolites are important components of inflammatory arthritis resolution in murine Lyme arthritis and may be a therapeutic target for treatment of joint edema and pain for Lyme arthritis patients without compromising spirochete clearance.

Keywords: Borrelia burgdorferi; Lyme disease; arthritis; eicosanoids; inflammation; resolution.

Figure 1

Expression of Alox15 mRNA over the time-course of mLA. WT C3H mice were infected with B. burgdorferi and groups of mice were sacrificed on the days indicated. The expression of Alox15 mRNA in ankle joints was determined and normalized to Nid1 and reported relative to D0 (uninfected) levels. n=10/group. *p<0.05 by one way-ANOVA with Dunnett’s test, compared to D0.

Figure 2

12/15-LO-/- C3H mice display defective mLA resolution. C3H WT and 12/15-LO-/- mice were infected with B burgdorferi and mLA was followed over time. (A) Ankle swelling curve throughout mLA displayed as diameter change from D0 (uninfected). Closed circles and solid line represent WT mice and open squares and dotted line represent 12/15-LO-/- mice. n=10/group. *p<0.05, **p<0.01, ***p<0.001 by t-test between strains per timepoint. ^##p<0.01, ^####p<0.0001 by Kruskal-Wallis one-way ANOVA with Dunnett’s test compared to D21 of the respective strain. (B) Arthritis severity scores as determined by histological analysis of H&E-stained ankle joints (scale 0-4) from mice sacrificed on the indicated days. Closed circles represent WT and open squares represent 12/15-LO-/- individual animals. Open bars represent median values from WT mice and grey bars represent 12/15-LO-/- mice. n=4-9/group. Data from two experiments were combined to increase the n value for D21, D28, and D35 groups. *p<0.05 Mann-Whitney U test. (C) Representative histology from WT and 12/15-LO-/- ankles at D35. Ankle joints were isolated and (D) macrophage (CD45.2+F4/80+) and (E) neutrophil (CD45.2+Ly-6Ghi) populations were quantified by flow cytometry. n=4-5/group. *p<0.05, **p<0.01 by t-test between strains per timepoint. ^#p<0.05, ^##p<0.01, ^###p<0.001 by one-way ANOVA with Dunnett’s test compared to D21 of the respective strain.

Figure 3

12/15-LO deficiency does not affect the anti-Borrelia immune response. C3H WT and 12/15-LO^-/- mice were infected with B. burgdorferi and sacrificed on the indicated dpi. Blood was collected and B. burgdorferi-specific serum IgM (A) and IgG (B) levels were determined by ELISA. n=7-10/group. *p<0.05 by t-test between strains per timepoint. (C) B. burgdorferi loads in ankles at D35 by qPCR. n=5/group.

Figure 4

Apoptosis and efferocytosis in 12/15-LO^-/- bone marrow neutrophils and macrophages. Bone marrow neutrophils (BMN) were isolated and cultured for 24hr to induce apoptosis. (A) Untreated APC-Cy7-stained apoptotic BMN (AN), either WT (black bars) or 12/15-LO^-/- (white bars), were cultured with WT or 12/15-LO^-/- BMDM (2:1) for 4hr, and efferocytosis, defined as F4/80⁺APC-Cy7⁺ cells, was determined by flow cytometry. Apoptosis of WT and 12/15-LO^-/- in 24hr culture +/- Bb (MOI 10) was measured by proportion (B) and MFI (C) of AnxV expression. (D) Late-stage apoptosis in cultures from B&C as measured by AnxV and 7-AAD co-staining. n=3/group, assayed in duplicate. (A–C) *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001 by t-test, (D) one-way ANOVA with Tukey’s posttest.

Figure 5

LXA₄:mFpr2 signaling axis components are upregulated in response to B. burgdorferi. (A) LXA₄ production by WT BMDMs after 2hr incubation with control (no Bb) or Bb (MOI 10). (B) mFpr2 mRNA expression by RT-qPCR after 24hr culture +/- Bb, normalized to Gapdh and relative to expression in unstimulated (no Bb) BMDM. (C) mFpr2 mRNA by RT-qPCR from WT ankle joints during mLA, normalized to Nid1 and relative to D0 levels. (A, B) n=3/treatment, assayed in duplicate, *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001 by t-test. (C) n=7-10/timepoint, one-way ANOVA with Dunnett’s posttest, compared to D0.

Figure 6

Exogenous LXA4 treatment reduces ankle edema and accelerates joint inflammatory cell removal. C3H WT mice were infected with B. burgdorferi and treated with either vehicle control (VC; closed circles, solid line) or LXA4 (open squares, dotted line) i.p. on days 18,19, and 20pi (arrows). Development of mLA was followed over time. (A) Ankle diameters were measured on the days indicated. n=5/group. **p<0.01, ****p<0.0001 by t-test between treatment groups per timepoint. (B) Arthritis severity scores of H&E-stained ankle joints from mice sacrificed on D21 and D35. VC (closed circles, open bars), LXA4 (open squares, grey bars). Bars represent median values. n=5-6/group. ^#p<0.05 by Mann-Whitney U test compared to D21 of the respective strain. Representative histology is in (C). Total macrophages (D) and neutrophils (E) quantified from the ankle joints of VC- (closed circles, open bars) or LXA4-treated (open squares, closed bars) WT mice at D21 and D35 by flow cytometry. Two macrophage subsets were further measured in D21 ankles: inflammatory macrophages (F; CD45.2+F4/80+Ly-6Chi) and reparative macrophages (G; CD45.2+F4/80+Ly-6Cint/lo). n=5/group. *p<0.05 by t-test between treatment groups per timepoint. Data from two experiments were combined to reach power in D35 groups (A, B, D, E). ***p<0.001.