The role of lipoprotein-associated phospholipase A2 in a murine model of experimental autoimmune uveoretinitis

Abstract

Macrophage activation is, in part, regulated via hydrolysis of oxidised low density lipoproteins by Lipoprotein-Associated phospholipase A2 (Lp-PLA2), resulting in increased macrophage migration, pro-inflammatory cytokine release and chemokine expression. In uveitis, tissue damage is mediated as a result of macrophage activation; hence inhibition of Lp-PLA2 may limit macrophage activation and protect the tissue. Utilising Lp-PLA2 gene-deficient (KO) mice and a pharmacological inhibitor of Lp-PLA2 (SB-435495) we aimed to determine the effect of Lp-PLA2 suppression in mediating retinal protection in a model of autoimmune retinal inflammation, experimental autoimmune uveoretinitis (EAU). Following immunisation with RBP-3 (IRBP) 1-20 or 161-180 peptides, clinical disease was monitored and severity assessed, infiltrating leukocytes were enumerated by flow cytometry and tissue destruction quantified by histology. Despite ablation of Lp-PLA2 enzyme activity in Lp-PLA2 KO mice or wild-type mice treated with SB-435495, the number of infiltrating CD45+ cells in the retina was equivalent to control EAU animals, and there was no reduction in disease severity. Thus, despite the reported beneficial effects of therapeutic Lp-PLA2 depletion in a variety of vascular inflammatory conditions, we were unable to attenuate disease, show delayed disease onset or prevent progression of EAU in Lp-PLA2 KO mice. Although EAU exhibits inflammatory vasculopathy there is no overt defect in lipid metabolism and given the lack of effect following Lp-PLA2 suppression, these data support the hypothesis that sub-acute autoimmune inflammatory disease progresses independently of Lp-PLA2 activity.

Fig 1. Quantification of Lp-PLA2 enzyme activity and gene expression in murine plasmas and BMDM.

A) Enzyme analysis of Lp-PLA₂ activity from murine cardiac blood samples (n = 15–17) **** p<0.001. B) RT-qPCR quantification of BMDM Pla2g7 expression (n = 3). Lp-PLA₂ WT, HET and KO BMMO were stimulated with LPS or media only for 2 or 4 hours, and gene expression analysed. Pla2g7 mRNA expression normalized by GAPDH. * p<0.05 ** p<0.01 ***p<0.005

Fig 2. Effect of Lp-PLA2 depletion on macrophage activation and phenotype on BMDM IL-6 protein expression.

Supernatants from BMDM cultured for 24 hours with SB-435495 treated oxLDL plus 1ng/ml LPS or media control were used to quantify IL-6 protein expression (n = 3). A) SB-435495 was not able to significantly alter IL-6 expression in LPS treated BMDM. IL-6 was undetected in media only controls. BMDM derived from Lp-PLA₂ WT, HET and KO mice were cultured with a titration of LPS. Supernatants were collected at 24 hours and used to quantify IL-6 protein expression by ELISA. B) There was no significant difference in IL-6 expression between Lp-PLA₂ WT, HET or KO BMDM.

Fig 3. Effect of Lp-PLA2 enzyme ablation on macrophage activation markers as an indicator of macrophage activation status.

Expression of IL-1β, IL-10 and Arginase-1 in Lp-PLA₂ WT, HET and KO mice by RT-qPCR. Data from BMDM, following 2 or 4 hour stimulation with LPS (1ng/ml) or media control A) IL-1β (n = 3–4) B) IL-10(n = 3) C) Arginase-1 (n = 3–4). D) A nitrite assay performed as described, using supernatants of BMDM from Lp-PLA₂ WT, HET and KO mice cultured in presence of LPS/IFNγ for 24 hours, showed no significant difference between genotypes (n = 9 from 5 mice). E) Data collected from nitrite assays using supernatants of BMDM treated with either LPS or IFNγ plus SB-435495 (1μM or 0.1μM) showed no significant difference between genotypes (n = 3) ****p<0.001

Fig 4. Flow cytometric analysis of BMDM phenotype from cells cultured for 24 hours in the presence of a panel of cytokines.

Geometric mean fluorescence intensity values for A) intracellular CD68 expression (n = 4), B) cell surface MHCII expression (n = 3) and C) cell surface CD40 expression (n = 5) * p<0.05 ** p<0.01 ***p<0.005

Fig 5. Quantification of retinal cellular infiltrate and onset of clinical disease in a murine model of EAU; effective of SB-435454.

A) Flow cytometry data showing CD45⁺, CD11b⁺, CD4⁺ and CD8⁺ cellular infiltrate in retinal digests from animals at day 14 post immunisation, treated twice daily with either SB-435495 or controls (n = 9–10 eyes). B) Histological staining scores from 12μm ocular sections taken from eyes enucleated at day 14 post immunisation, stained for CD45⁺ (DAB/haematoxylin) (n = 9–10 eyes). C) TEFI clinical score, day 13 post immunisation, (average scores provided by two independent assessors from two identical repeat experiments. n = 18–20).

Fig 6. Quantification of retinal cellular infiltrate at day 26 post immunisation in a C57BL/6 murine model of EAU carrying Lp-PLA2 WT, heterozygous and homozygous gene-deletion genotypes.

Flow cytometry data showing CD45⁺, CD11b⁺, CD4⁺ and CD8⁺ cellular infiltrate in retinal digests from Lp-PLA₂ KO, HET, WT animals with EAU at day 26 post immunisation. (n = 14–15 eyes).

Fig 7. Ocular oxLDL in naive and uveitic mice.

Immunofluorescence staining of DAPI (blue) oxLDL (red) and CD31 (green) using 12μm ocular sections from A) naïve mice or B-C) those immunised for EAU (sequential slides). Ganglion cell layer (GCL), inner nuclear layer (INL), inner plexiform layer (IPL), outer nuclear layer (ONL).