Crystallin β-b2 promotes retinal ganglion cell protection in experimental autoimmune uveoretinitis

Abstract

Crystallin βb2 (crybb2) is upregulated in regenerating retinas and in various pathological conditions of the retina, including uveoretinitis. However, the role of crybb2 in this disease is largely unknown. Therefore, we used recombinant crybb2 (rcrybb2) as intravitreal treatment of B10.RIII mice prior to immunization with human interphotoreceptor retinoid-binding protein peptide 161-180 (hIRBPp161-180) in complete Freund's adjuvant (CFA) and concomitant injection of pertussis toxin (PTX) to induce experimental autoimmune uveoretinitis (EAU). In naïve mice, more beta III-tubulin (TUBB3) + and RNA-binding protein with multiple splicing (RBPMS) + cells were found in the ganglion cell layer of the retina than in EAU eyes, suggesting a loss of retinal ganglion cells (RGC) during the development of EAU. At the same time, the number of glial fibrillary acidic protein (GFAP) + cells increased in EAU eyes. RGCs were better protected in EAU eyes treated with rcrybb2, while the number of GFAP+ cells decreased. However, in retinal flatmounts, both retinal ganglion cells and retinal endothelial cells stained positive for TUBB3, indicating that TUBB3 is present in naïve B10.RIII mouse eyes not exclusive to RGCs. A significant decline in the number of RBPMS-positive retinal ganglion cells was observed in retinal flatmounts from EAU retinas in comparison to naïve retinas or EAU retinas with intravitreal rcrybb2 treatment. Whereas no significant decrease in TUBB3 levels was detected using Western blot and RT-qPCR, GFAP level, as a marker for astrocytes, increased in EAU mice compared to naïve mice. Level of Bax and Bcl2 in the retina was altered by treatment, suggesting better cell survival and inhibition of apoptosis. Furthermore, our histologic observations of the eyes showed no change in the incidence and severity of EAU, nor was the immune response affected by intravitreal rcrybb2 treatment. Taken together, these results suggest that intravitreal injection of rcrybb2 reduces retinal RGC death during the course of EAU, independent of local or systemic autoimmune responses. In the future, treating posterior uveitis with rcrybb2 to protect RGCs may offer a promising novel therapeutic strategy.

Keywords: apoptosis; crystallin β-b2; experimental autoimmune uveitis; neuroprotection; retinal ganglion cell.

Figure 1

Localization of rcrybb2 in the eyes after injection into naïve mice. Paraffin-embedded sections of eyes (N = 3 at each time point) after injection of 3 μg/2 μL rcrybb2 and immunofluorescence staining with an antibody directed against crybb2 showing pronounced fluorescent staining. Paraffin-embedded sections of mouse eyes at 0 (A), 10 (B), and 21 days (C) after injection of 3 μg/2 μL rcrybb2 showing that rcrybb2 could still be detected after 21 days. (D) Paraffin-embedded section of an untreated mouse eye. Scale bar: 100 μm.

Figure 2

Expression and localization of TUBB3 (green), RBPMS (red) and glial fibrillary acidic protein (GFAP; red) in naïve retinas or in retinas with EAU and prophylactic treatment with 2 μL rcrybb2 (3 μg/2 μL) or PBS as control. Some mice were also treated with a single intralenticular injection of PBS as a positive control. Mice were immunized 3 days later and eyes were collected 21 days after immunization. The expression of retinal (A,D) TUBB3 (green), (B,E) RBPMS (red) and (C,F) GFAP (red) was determined by immunofluorescence staining of sections (7 μm) from mice that were naive, with EAU (untreated), EAU + PBS (vitreous), EAU + PBS (lens), or EAU + rcrybb2 (vitreous) on day 21 p.i. Positive cells were counted in 3 different section per eye (N = 5 eyes per group) and data are expressed as mean ± SD. Scale bar: 100 μm. (G,H,I) The number of retinal RBPMS+ cells (red) in naïve, EAU (untreated) and EAU(rcrybb2)-treated retinas was determined by flatmount staining (N = 4 per group). Scale bar: 50 μm. (D–F,J) ANOVA with Tukey’s post-hoc test: ∗p < 0.05. ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001; ∗∗∗∗p < 0.0001.

Figure 3

Expression and localization of TUBB3, RBPMS and CD31/PECAM1 in naïve retinas and in retinas with EAU. The expression of TUBB3 (green), RBPMS (red) in (A) naïve retinas and (B) retinas with EAU was determined by flatmount staining. The results shows that TUBB3 is expressed in retinal ganglion cells (RGCs) but also in (C) CD31+ endothelial cells. Scale bar: 50 μm.

Figure 4

Western blot analysis and RT-qPCR graphs of correlated relative densities. Blots of (A) TUBB3 and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) and their correlated relative densities, each with the corresponding control containing GAPDH. (B) mRNA level of Tubb3 as measured by RT-qPCR analysis. Blots of (C) GFAP and GAPDH and their correlated relative densities, each with the corresponding control with GAPDH. (D) mRNA level of Gfap as measured by RT-qPCR analysis. Retinal samples were obtained from naïve eyes, EAU (untreated), EAU + PBS (vitreous), or EAU + rcrybb2 (vitreous). Mean ± SEM. Student’s t-test, ∗p < 0.05.

Figure 5

mRNA level of the pro- or anti-apoptotic proteins Bax and Bcl-2 in the retina after treatment with rcrybb2. Gene transcription of Bax and Bcl-2 was measured by RT-qPCR analysis. Retinal samples were isolated from naïve mice, EAU (untreated), EAU + PBS (vitreous), or EAU + rcrybb2 (vitreous). Mean ± SEM. Student’s t-test, ∗p < 0.05.

Figure 6

Severity of EAU after intravitreal injection of 2 μL rcrybb2 (3 μg/2 μL) or PBS as control. The lens is a site of crystallin origin; therefore, some mice were treated with a single intralenticular injection of PBS as a positive control. Mice were immunized 3 days later and eyes were collected 21 days after immunization. Statistical differences of EAU scores of the eyes between groups were calculated by using the Kruskal-Wallis test with post-hoc analysis. The analysis showed that there was no significant difference between the different treated groups. In addition, the differences between the left (treated) and right (untreated) eyes were not statistically significant as determined by U-test. (A) Histologic EAU scores of the different groups: EAU (untreated), EAU + PBS (vitreous), EAU + PBS (lens), EAU + rcrybb2 (vitreous; N = 20 each group). (B) Representative section of a mouse retina with EAU (clinical score: 2) showing retinal folding. (C) Section of a normal mouse retina. Hematoxylin–eosin staining. Scale bar: 100 μm.

Figure 7

Influence of intravitreal rcrybb2 injection on hIRBPp161-180 specific immune responses. Splenocytes were cocultured with medium, hIRBPp161-180, CD3, or ConA. Splenocyte supernatants from immunized mice were collected 24 h after stimulation and assayed for the indicated cytokines using ELISA kits. Levels of (A) IFN-γ, (B) IL-17, (C) IL-6, and (D) IL-10 after stimulation. Data are expressed as mean ± SEM. ANOVA with Tukey’s post-hoc test: ∗p < 0.05.

Figure 8

Increased level of Tubb3 and Gfap in retinal endothelial cells in mice with EAU. Single-cell RNA sequencing (scRNAseq) analysis of endothelial cells from eyes with EAU (DE, diseased endothelium, N = 3) or healthy control eyes [NE, naïve endothelium, N = 4; GSE144168, reported by Lipski et al. (2020)]. (A,B) Level of Tubb3 and Gfap calculated using the classical Bayesian algorithm of the “Limma” R package of Network Analyst 3.0. Mean ± SEM, Adj. P: ∗∗p < 0.01.