Ccl2, Cx3cr1 and Ccl2/Cx3cr1 chemokine deficiencies are not sufficient to cause age-related retinal degeneration

Abstract

Monocytes, macrophages, dendritic cells and microglia play critical roles in the local immune response to acute and chronic tissue injury and have been implicated in the pathogenesis of age-related macular degeneration. Defects in Ccl2-Ccr2 and Cx3cl1-Cx3cr1 chemokine signalling cause enhanced accumulation of bloated subretinal microglia/macrophages in senescent mice and this phenomenon is reported to result in the acceleration of age-related retinal degeneration. The purpose of this study was to determine whether defects in CCL2-CCR2 and CX3CL1-CX3CR1 signalling pathways, alone or in combination, cause age-dependent retinal degeneration. We tested whether three chemokine knockout mouse lines, Ccl2(-/-), Cx3cr1(-/-) and Ccl2(-/-)/Cx3cr1(-/-), in comparison to age-matched C57Bl/6 control mice show differences in subretinal macrophage accumulation and loss of adjacent photoreceptor cells at 12-14 months of age. All mouse lines are derived from common parental strains and do not carry the homozygous rd8 mutation in the Crb1 gene that has been a major confounding factor in previous reports. We quantified subretinal macrophages by counting autofluorescent lesions in fundus images obtained by scanning laser ophthalmoscopy (AF-SLO) and by immunohistochemistry for Iba1 positive cells. The accumulation of subretinal macrophages was enhanced in Ccl2(-/-), but not in Cx3cr1(-/-) or Ccl2(-/-)/Cx3cr1(-/-) mice. We identified no evidence of retinal degeneration in any of these mouse lines by TUNEL staining or semithin histology. In conclusion, CCL2-CCR2 and/or CX3CL1-CX3CR1 signalling defects may differentially affect the trafficking of microglia and macrophages in the retina during ageing, but do not appear to cause age-related retinal degeneration in mice.

Fig. 1

A. Autofluorescent scanning laser ophthalmoscopy fundus images (AF-SLO) of C57Bl/6, Ccl2^−/−, Cx3cr1^−/− and Ccl2^−/−/Cx3cr1^−/− knockout mice. Images are focused on the outer retina and reveal discrete autofluorescent lesions (white arrowheads) indicative of subretinal macrophages in all genotypes at the age of 12–14 months. B. Quantification of discrete autofluorescent spots in AF-SLO fundus images. The individual data points in this graph represent the mean number of autofluorescent spots between the left and the right eye of each animal (n). The mean number of autofluorescent spots ± standard deviation for each group (genotype) is indicated by a small horizontal line and the error bars. Ccl2^−/− mice show significantly higher numbers of autofluorescent spots per fundus image than the other two chemokine deficient lines or the C57Bl/6 age-matched control. Stars indicate statistically significant changes (p = 0.0017, Oneway ANOVA with Tukey's multiple comparison p < 0.05; n (C57Bl/6) = 7, n (Ccl2^−/−) = 5, n (Cx3cr1^−/−) = 8, n (Ccl2^−/−/Cx3cr1^−/−) = 7). C. Immunohistochemistry for the macrophage/microglia marker Iba1 (green) on sagittal, superior-inferior oriented retinal sections obtained from 12 to 14 months old C57Bl/6, Ccl2^−/−, Cx3cr1^−/− and Ccl2^−/−/Cx3cr1^−/− knockout mice. White arrowheads indicate the position of Iba1 positive subretinal macrophages. Sections were counterstained with DAPI (blue) to label the nuclei. INL: Inner nuclear layer, ONL: Outer nuclear layer, RPE/Ch: retinal pigment epithelium and choroid. Scale bar: 50 μm. D. Quantification of Iba1 positive macrophages (white arrowheads, Fig. 1C) in the subretinal space of C57Bl/6, Ccl2^−/−, Cx3cr1^−/− and Ccl2^−/−/Cx3cr1^−/− mice. Subretinal macrophages were counted in 10 superior to inferior oriented sagittal sections that were about 225 μm apart and equally distributed across the eyes. One eye per animal was evaluated and the results for each group are shown as the sum of Iba1⁺ macrophages/10 sections ± standard deviation. The pattern observed is consistent with that in the AF-SLO counts (Fig. 1B), although no statistically significant differences between any of the four mouse lines were detected (p = 0.1653 Oneway ANOVA with Tukey's multiple comparison p < 0.05; n (C57Bl/6) = 7, n (Ccl2^−/−) = 5, n (Cx3cr1^−/−) = 7, n (Ccl2^−/−/Cx3cr1^−/−) = 7).

Fig. 2

A. Semithin histology from the central retina of C57Bl/6, Ccl2^−/−, Cx3cr1^−/− and Ccl2^−/−/Cx3cr1^−/− mice. No gross alterations in retinal structure or RPE morphology were observed. B. Assessment of photoreceptor death by using TUNEL staining. Independent of the chemokine genotypes, all animals showed occasional TUNEL positive nuclei (white arrowheads) in the outer nuclear layer indicating a similar basal level of photoreceptor apoptosis at the age of 12–14 months. C. Quantification of photoreceptor nuclei in the outer nuclear layer (ONL) on semithin histological sections. We determined the number of rows of nuclei in the outer nuclear layer of all 4 mouse lines and did not observe statistically significant differences. All lines showed mean ± SD of 11 ± 1 rows of photoreceptor nuclei in the ONL, independent of the respective genotype (p = 0.6045 Oneway ANOVA with Tukey's multiple comparison p < 0.05, n (C57Bl/6) = 6, n (Ccl2^−/−) = 5, n (Cx3cr1^−/−) = 7, n (Ccl2^−/−/Cx3cr1^−/−) = 7). D. Quantification of TUNEL positive photoreceptor nuclei in the outer nuclear layer of C57Bl/6, Ccl2^−/−, Cx3cr1^−/− and Ccl2^−/−/Cx3cr1^−/− mice. 10 superior to inferior oriented sagittal retinal sections that were about 225 μm apart and equally distributed across the eye were evaluated and the sum of TUNEL positive photoreceptor nuclei in the outer nuclear layer for each animal are shown as individual data points. The mean ± standard deviation for each group is also indicated. No significant differences were observed between any of the three chemokine deficient mice as well as in comparison to C57Bl/6 mice (p = 0.9976 Oneway ANOVA with Tukey's multiple comparison p < 0.05, n (C57Bl/6) = 7, n (Ccl2^−/−) = 5, n (Cx3cr1^−/−) = 7, n (Ccl2^−/−/Cx3cr1^−/−) = 7). INL: inner nuclear layer, ONL: outer nuclear layer, RPE/Ch: retinal pigment epithelium/Choroid; scale bars 50 μm.