Fig 2: CASPR1 and CNTN1 co-immunoprecipitate with RIBEYE ACoomassie-stained proteins from bovine retina that were co-immunoprecipitated by the anti-RIBEYE antibody 2D9 (lane 3). Lane 2, control precipitation by irrelevant mouse IgGs; lane 1, input (10% of the total loaded in each lane).BWestern blot of the immunoprecipitated proteins (as in A, 1% of the total protein was loaded per lane) probed with anti-RIBEYE (Ba), anti-CASPR1 (Bb), and anti-CNTN1 (Bc). The 2D9 antibody immunoprecipitated RIBEYE (Ba, lane 3), CASPR1 (Bb, lane 3), and CNTN1 (Bc, lane 3); these proteins were not detectable in the control immunoprecipitations. (Bd) As a loading control, heavy chains of the precipitating antibody (IgG hc) were visualized by the goat anti-mouse IgG antibody conjugated to peroxidase. Source data are available online for this figure.

Fig 3: The amount of CASPR1 protein in the synapse is also altered in the PLP mouse model of multiple sclerosis A, BSemi-thin sections of PLP/CFA- and CFA-injected mice were probed with the indicated antibodies (see also Appendix Fig S13).CThe synaptic amount of CASPR1 is strongly reduced both in the OPL and IPL. RIBEYE is only reduced in the OPL but not significantly in the IPL. CNTN1 is strongly reduced in both OPL and IPL (see also Appendix Fig S13). N = 4 PLP/CFA-injected mice; N = 4 CFA-injected mice; n = 140 slices (CFA, RIBEYE); n = 143 slices (PLP/CFA, RIBEYE); n = 47 slices (CFA, CASPR1); n = 53 slices (PLP/CFA, CASPR1); n = 49 slices (CFA, CNTN1); n = 46 slices (PLP/CFA, CNTN1); n = 44 slices (CFA, v-Arrestin); n = 44 slices (PLP/CFA, v-Arrestin).D, ECryostat sections of the optic nerve from PLP and control mice were stained with anti-Nav-channel antibody to label the node of Ranvier and with anti-CASPR1 antibody to stain the paranodal region.FTransmission electron micrographs of the optic nerves from PLP/CFA- and CFA-injected animals (F1–F4).G, HLength of the node of Ranvier and length of the paranodal region were quantified.I, JAxon numbers and myelin thickness were quantified.

Fig 4: Localization of CNTN1 in semi-thin (0.5-µm-thin) sections of retinas from either MOG/CFA-injected mice (EAE mice) or CFA-injected control mice (9 days after injection) A, BSections were double-immunolabeled with antibodies against CNTN1 and RIBEYE. Scale bars: 20 µm.CQuantification of the immunosignals of the indicated proteins in the OPL and IPL. ONL, outer nuclear layer; OPL, outer plexiform layer; INL, nuclear layer; IPL, inner plexiform layer; n.s., non-significant; ***P < 0.001 (Mann–Whitney U-test, actual P-values are given in the figure) N = 6 independent set of experiments; n = 295 images (CFA); n = 274 images (MOG/CFA) for quantification of RIBEYE immunosignals; n = 95 (CFA); n = 92 (MOG/CFA) for quantification of CASPR1 immunosignals; n = 91 (CFA), n = 82 (MOG/CFA) for CNTN1 quantification; n = 109 (CFA); n = 100 (MOG/CFA) for quantification of v-Arrestin. Error bars are ± SEM.

Fig 5: Retinal CASPR1, CNTN1 and RIBEYE are down-regulated after MOG/CFA injection ARetinal lysates from MOG/CFA-injected and CFA-injected control mice tested in Western blot with the indicated antibodies.BSemi-quantitative evaluation of the band intensity of the proteins recognized in Western blot. n.s., non-significant; *P < 0.05; **P < 0.01 (precise P-values are given in the figure). N = 5 retinas from CFA-injected mice; N = 7 retinas from MOG/CFA-injected mice. Error bars are ± SEM; statistical test: Mann-Whitney U-test (Origin Pro).