NMOSD IgG Impact Retinal Cells in Murine Retinal Explants

Abstract

Neuromyelitis optica spectrum disorders (NMOSD) are chronic inflammatory diseases of the central nervous system, characterized by autoantibodies against aquaporin-4. The symptoms primarily involve severe optic neuritis and longitudinally extensive transverse myelitis. Although the disease progression is typically relapse-dependent, recent studies revealed retinal neuroaxonal degeneration unrelated to relapse activity, potentially due to anti-aquaporin-4-positive antibodies interacting with retinal glial cells such as Müller cells. In this exploratory study, we analysed the response of mouse retinal explants to NMOSD immunoglobulins (IgG). Mouse retinal explants were treated with purified IgG from patient or control sera for one and three days. We characterized tissue response patterns through morphological changes, chemokine secretion, and complement expression. Mouse retinal explants exhibited a basic proinflammatory response ex vivo, modified by IgG addition. NMOSD IgG, unlike control IgG, increased gliosis and decreased chemokine release (CCL2, CCL3, CCL4, and CXCL-10). Complement component expression by retinal cells remained unaltered by either IgG fraction. We conclude that human NMOSD IgG can possibly bind in the mouse retina, altering the local cellular environment. This intraretinal stress may contribute to retinal degeneration independent of relapse activity in NMOSD, suggesting a primary retinopathy.

Keywords: Müller cell; NMOSD; autoantibodies; chemokine; complement; local; mouse retinal explants; retina.

Figure A1

Apoptosis was assessed in all retinal layers during seven days of in vitro cultivation using the TUNEL technique. While no fluorescence was detected in the ex vivo retina, increasing apoptotic activity was observed with prolonged cultivation time. Therefore, we selected three days of retinal explant cultivation for all subsequent experiments. Bars represent mean values (n = 3) ± SD, with statistical significance denoted as ** p < 0.01, **** p < 0.0001 as determined by ordinary one-way ANOVA and Dunnett’s multiple comparisons test.

Figure A2

Secondary antibodies used in immunocytochemical staining were tested for unspecific binding to retinal tissue. (a) Negative control staining of the secondary goat–anti human FITC-conjugated antibody on uncultivated, fixated murine retinal tissue. (b) Negative control stainings of the secondary goat–anti rabbit Cy3-conjugated antibody on uncultivated, fixated murine retinal tissue.

Figure 1

Retinal explant morphology, complement component transcription, and chemokine secretion during in vitro cultivation. (A) Retinal morphology of freshly fixed (ex vivo) and cultivated retinas of C57BL/6J mice was assessed using a HE staining to visualize cell nuclei, extracellular matrix, and cytoplasm, respectively. Retinal thinning was observed in all layers after one and three days of cultivation compared to uncultured retinas. Cell death increased during cultivation with no fluorescence detected in the ex vivo retina. Scale bar, 50 µm. (B) Retinal transcription of complement components changed depending on the cultivation time. C1qb, c3, and cfh mRNA expression significantly increased during cultivation. Aqp4 mRNA decreased, and no trend was detectable for c1s mRNA. Bars represent mean values (n = 3) ± SD. Compared to an uncultivated control: * p < 0.05, ** p < 0.01 (ordinary one-way ANOVA, Dunnett’s multiple comparisons test). (C) Supernatants of cultivated retinas were collected at one and three days of untreated retinal cultivation and analysed for chemokines using a multiplex cytokine assay. CXCL10 secretion increased during in vitro cultivation. CCL2, CCL3, CCL4, CCL5, and CXCL2 secretion remained unchanged. GCL: Ganglion cell layer, INL: inner nuclear layer, ONL: outer nuclear layer. Bars represent mean values (n = 6) ± SD. * p < 0.05 (multiple paired t-tests, Holm–Sidak method).

Figure 2

Human purified NMOSD IgG bound to the mouse retina. (A) AQP4-IgG reactivity was analysed in serum of controls (n = 5) and NMOSD patients (n = 5) as well as purified IgG pools (n = 1 each). Control serum samples and the control IgG pool tested negative for AQP4-IgG, while positive anti-AQP4 IgG titers were confirmed for NMOSD patient sera and the purified NMOSD IgG pool. Bars represent mean values ± SD. * p < 0.05. (unpaired t-test assuming Gaussian distribution). (B) The Coomassie staining of NMOSD patient serum before and after immunoaffinity purification was used to verify IgG purification. (C) Mouse retina sections were incubated with serum and purified IgG from controls and NMOSD patients. Human IgG binding was confirmed for serum and purified IgG from NMOSD patients, while no binding was observed for serum and IgG from controls. Scale bar, 50 μm.

Figure 2

Human purified NMOSD IgG bound to the mouse retina. (A) AQP4-IgG reactivity was analysed in serum of controls (n = 5) and NMOSD patients (n = 5) as well as purified IgG pools (n = 1 each). Control serum samples and the control IgG pool tested negative for AQP4-IgG, while positive anti-AQP4 IgG titers were confirmed for NMOSD patient sera and the purified NMOSD IgG pool. Bars represent mean values ± SD. * p < 0.05. (unpaired t-test assuming Gaussian distribution). (B) The Coomassie staining of NMOSD patient serum before and after immunoaffinity purification was used to verify IgG purification. (C) Mouse retina sections were incubated with serum and purified IgG from controls and NMOSD patients. Human IgG binding was confirmed for serum and purified IgG from NMOSD patients, while no binding was observed for serum and IgG from controls. Scale bar, 50 μm.

Figure 3

NMOSD IgG-enhanced Müller cell gliosis in retinal explant cultures. (A) GFAP immunoreactivity (red) was assessed by immunohistochemistry staining in retinas treated with control IgG or NMOSD IgG pool. (B) Quantification of GFAP immunoreactivity after one and three days of treatment with respective IgG fractions in retinal explant culture demonstrated increased GFAP levels in NMOSD IgG pool-treated retinas after three days of cultivation compared to control IgG-treated retinas. Bars represent mean values ± SD. (control—one day (n = 3); three days (n = 2); NMOSD—one day (n = 3); three days (n = 3)). (C) Retinal aqp4 mRNA expression was analysed after one and three days of cultivation with control IgG or with NMOSD IgG pool (one day (n = 3), three days (n = 2)). The mRNA levels were compared to untreated, cultivated retinas. No significant changes in mRNA levels were observed due to different treatments. Bars represent mean values ± SD. (two-way ANOVA with Tukey’s multiple comparisons test).

Figure 4

Mouse retinal explants treated with NMOSD IgG showed decreased release of CCL2, CCL3, CCL4, and CXCL10. Chemokine levels were assessed in the supernatants of control IgG or NMOSD IgG-treated cultivated retinas. CCL2, CCL3, CCL4, and CXCL10 were significantly reduced after one day of cultivation with NMOSD IgG compared to control IgG-treated retinas. Bars represent mean values (n = 10–11) ± SD. * p < 0.05, ** p < 0.01 (two-way ANOVA with Sidak’s multiple comparisons test).

Figure 5

The effect of NMOSD IgG on complement mRNA expression in mouse retinal explants was investigated. The mRNA levels of complement components c1qb, c1s, c3, and cfh were analysed after one and three days of cultivation with control IgG (one day n = 4, three days n = 3) or with NMOSD IgG pool (one day n = 3, three days n = 2). The mRNA levels were compared to untreated, cultivated retinas. No significant changes in mRNA levels between the different antibody treatments were observed. Bars represent mean values ± SD (two-way ANOVA with Tukey’s multiple comparisons test).