RasV12 Expression in Microglia Initiates Retinal Inflammation and Induces Photoreceptor Degeneration

Abstract

Purpose: The role of activated retinal microglia in driving retinal degeneration has been implicated in a number of in vivo disease models. Here, we investigated the primary consequences of microglial activation by the specific expression of constitutively active Ras in microglia in a transgenic mouse model before the onset of any degenerative changes in the retina.

Methods: The double transgenic lines CAG-LSL-RasV12-IRES-EGFP; Cx3cr1CreER/+ (Cx3cr1-RasV12 mice) and CAG-LSL-EGFP; Cx3cr1CreER_+ (control mice) were generated. The expression of RasV12 was induced in microglia by tamoxifen administration, and the retinas were examined by immunohistochemistry of frozen sections, RT-qPCR, and live imaging.

Results: RasV12 expression in retinal microglial cells promoted cell proliferation, cytokine expression, and phagocytosis. RasV12-expressing microglia migrated toward the inner and outer layers of the retina. Examination of glial fibrillary acidic protein (GFAP) expression revealed activation of Müller glia in the retina. We also observed loss of the photoreceptors in the outer nuclear layer in close proximity to microglial cells. However, no significant neurodegeneration was detected in the inner nuclear layer (INL) or ganglion cell layer (GCL). The morphology of RasV12-expressing microglia in the GCL and INL retained more ramified features compared with the predominantly-ameboid morphology found in outer retinal microglia.

Conclusions: The expression of RasV12 is sufficient to activate microglia and lead to photoreceptor degeneration. Neurons in the inner side of the retina were not damaged by the RasV12-activated microglia, suggesting that microenvironment cues may modulate the microglial phenotypic features and effects of microglial activation.

Figure 1.

Expression of RasV12-EGFP in the microglia in Cx3cr1-RasV12 mouse. Cx3cr1-RasV12 or control (Cx3cr1-EGFP) mice at P14 were administered with tamoxifen, and after day 1, 3, 5, or 7, the retinas were harvested. Isolated retinas were frozen sectioned, and stained with anti-EGFP and indicated antibodies. (A) EGFP and Iba1 double staining patterns of day 7 retinas. Scale bars: 50 µm. (B) The number of EGFP and Iba1 double positive cells at day 7. (C) The population of EGFP and Iba1 double positive cells in total EGFP positive cells. (D) Distribution of EGFP and Iba1 double positive cells was measured in each retinal layer. (E) EGFP and Tmem119 double staining patterns in day 1, 3, 5, and 7 after tamoxifen administration. Scale bars: 20 µm. (F) The populations of EGFP and Tmem119 doble positive cells in total EGFP positive cells are shown. (G) EGFP and phosphorylated Erk1/2 (pErk1/2) staining patterns at day 7. Scale bars: 20 µm. (H) The population of EGFP and pErk1/2 double positive cells in the total EGFP positive cells was calculated in each region of the retina. (I) EGFP and pErk1/2 double positive cells were segregated to those with high or low levels of pErk1/2 signals, and population was calculated. Graphs show mean of six samples from independent mice with SEM. P values were calculated by Mann-Whitney U test. * P < 0.05, ** P < 0.01, n.s., not significant.

Figure 2.

RasV12 expressing microglia accumulated in both inner and outer layers of the retina. (A) Number of EGFP and Tmem119 double positive cells after tamoxifen administration in retinas of control (Cx3cr1-EGFP) and Cx3cr1-RasV12 mice. Immunohistochemistry panels are shown as Figure 1E. (B) Distribution of RasV12/EGFP or control EGFP expressing-microglia in the retina. Quantification of number of EGFP and Tmem119 double positive cells in GCL, IPL, ONL, and SR are shown. (C–E) Proliferation of the microglia was examined by the immunohistochemical analysis of Ki67. (C) Immunostaining patterns of anti-EGFP and -Ki67 antibodies of day 7 retinas. Nuclei were visualized with staining by DAPI. Scale bars: 20 µm. (D) The population of EGFP and Ki67 double positive cells in total EGFP positive cells. (E) Distribution of EGFP and Ki67 double positive cells. (F) Expression of transcripts of various chemokines and cytokines was examined by RT-qPCR retinas of Cx3cr1-RasV12 or control mice at day 7 after tamoxifen administration. In (B) and (F), the data of graphs are mean with SEM from six samples of independent mice. P values were calculated by Mann-Whitney U test. * P < 0.05, ** P < 0.01, *** P < 0.001, n.s., not significant. In (D) and (E), graphs show mean with SEM from three samples of independent mice.

Figure 3.

Perturbed ONL structure and activated Müller glia in Cx3cr1-RasV12 retina. (A) Paraffin sectioned retinas of control (Cx3cr1-EGFP) or Cx3cr1-RasV12 mice were subjected to H&E staining. Right panel of Cx3cr1-RasV12 section is enlarged view of black squared region in the middle panel. White arrowheads indicate spotted missing of photoreceptors. Scale bars: 20 µm. (B) Frozen sectioned retinas of control or Cx3cr1-RasV12 were subjected to immunohistochemistry using anti-EGFP, -Rhodopsin antibody, and nuclei were visualized by staining with DAPI. DIC images are shown. Scale bars: 20 µm. (C) Thickness of outer-segment was measured from DIC pictures in control and Cx3cr1 mice retina. (D, E) Flow cytometrical analysis of retinal cells of control and Cx3cr1-RasV12 retinas by staining with Annexin V, CD73, and PI. Dot blot patterns of Annexin V and propidium iodide (PI) of CD73 negative (upper panels) and CD73 positive (lower panels) are shown. PI negative/Annexin V positive cell population in CD73 negative or positive population of Cx3cr1-RasV12 retina are shown. (F) Expression levels of Edn2 and Fgf2 transcripts by RT-qPCR in control and Cx3cr1-RasV12 retinas at day 7. (G) Frozen sections of control or Cx3cr1-RasV12 mice derived retinas were stained with anti-EGFP, -GFAP, and -Tmem119 antibodies. Nuclei were visualized with DAPI staining. (H) The expression level of Gfap transcripts of control or Cx3cr1-RasV12 retina was examined by RT-qPCR. In (C), (F), and (H), the graphs are mean of six independent samples with SEM. In (E), values in the graph are mean with SEM from four samples of independent mice. P values were calculated by Mann-Whitney U test. *P < 0.05, ** P < 0.01.

Figure 4.

The expression of phagocytosis related molecules in Cx3cr1-RasV12 mice retina. (A) Orthological view of EGFP and Recoverin antibodies stained retina of Cx3cr1-RasV12 mice at day 7. Scale bar: 10 µm. (B–D) Frozen sections of retinas of control (Cx3cr1-EGFP) and Cx3cr1-RasV12 mice at day 7 after tamoxifen administration were subjected to immunohistochemistry using anti-EGFP, -CD68, and -Tmem119 antibodies. Nuclei were visualized with DAPI in some panels. White arrowheads indicate EGFP, CD68 and Tmem119 triple positive cells in ONL in (B). Scale bars: 20 µm. In (C), number of EGFP, CD68, and Tmem119 triple positive cells in subregions of the retina was counted, and cell number in each sub-region in 1mm width of the section. In (D), the population of the triple positive cells in EGFP and Tmem119 double positive cells (%) in subregions of the retina. (E) Expression of transcripts of various phagocytosis related genes in control (Cx3cr1-EGFP) and Cx3cr1-RasV12 retina were examined by RT-qPCR. In (C), (D), and (E), graphs show mean of counting from six independent samples with SEM. P values were calculated by Mann-Whitney U test. *P < 0.05, n.s., not significant.

Figure 5.

Phagocytosis of photoreceptors by RasV12-expressing microglia were observed in ONL but not observed in IPL. Time-lapse analysis in Cx3cr1-RasV12 retinal explant. EGFP positive cell phagocytosed nuclear in ONL (upper panels), but did not show phagocytotic feature in IPL (lower panels). White arrowheads indicate phagocytosed nuclei by EGFP positive cell. Panels show pictures in two-minute intervals. Scale bars: 10 µm. Movies are up-loaded as supporting information.