Fig 2: Human MCs show epithelial phenotype associated with TGFß/Notch under pathological condition. (A) Human H&E stained sections of retinas that show healthy cuboidal RPE (Neg) and retinas presenting drusen beneath the pigment epithelium (Drusen pos). Zoom-in view showing a healthy cuboidal RPE layer (Neg, top left corner) and drusen or micro drusen underneath the RPE layer (Drusen pos, top left corner). (B) Quantification of drusen as either hyalinized rounded deposits (> 25 µm) or micro drusen occurring singly or in a row (< 25 µm) in between the RPE and the Bruch’s membrane. The analyzed length of the retina was 950 µm retina sections corresponding to the field of view. (C) Table summarizing the eight selected samples (total analyzed sections = 52) selected for H&E and immunofluorescence analysis. (D-J) Analysis of MC reactivity and phenotype of healthy retinas (Neg) and retinas presenting drusen (Drusen pos). Detection of GFAP (D.i-D.iv, E.i-E.iv), PCNA (F.i-F.iv, G.i-G.iv) and E-cadherin (H.i-H.iv, I.i-I.iv) in GS+ MCs. Shown are representative sections stained for GS and SOX9 (red), GFAP, PCNA and E-Cadherin (green). (J) Histogram illustrating the mean ± SD of the number of GFAP+, PCNA+ and E-cadherin+ cells normalized by the total number of GS+ cells expressed in percentage. Significant differences (****p < 0.0001) between “Neg” and “Drusen pos” retinas were determined by using a post-hoc Bonferroni two-way ANOVA test (n = 8). (K-O) Analysis of regulators of MET induction in healthy retinas (Neg) and retinas presenting drusen (Drusen pos). Detection of TGFß1 (K.i-K.iv, L.i-L.iv) and Notch2 (M.i-M.iv, N.i-N.iv) in GS+ MCs. Shown are representative retinal sections stained for GS (red), TGFß1, and Notch2 (green). (O) Histogram illustrating the mean ± SD of the number of TGFß1+ and Notch2+ cells normalized by the total number of GS+ cells expressed in percentage. INL, inner nuclear layer; ONL, outer nuclear layer. Scale bar of all images equals 50 µm, while in the zoom-in view corresponding to 150 µm

Fig 3: Investigation of Notch pathway during MET in murine MCs. Flow cytometry analysis of Notch isoforms in reactive MCs in mouse at the baseline (Uninjured) and at different time points after injury (Day 1, 3 and 7). (A) Representative figures of MCs gated for GFAP and PCNA and further for Notch1/2 at Day 3 and 7. (B-C) Histograms illustrating the mean ± SD of the number of GFAP+/PCNA+/Notch1+ cells (B) and GFAP+/PCNA+/Notch2+ cells (C) normalized by the total number of GFAP+/PCNA+ cells expressed in percentage. Significant differences (***p < 0.001 and ****p < 0.0001) between uninjured, Day 1, 3 and 7 were determined by using a post-hoc Bonferroni one-way ANOVA test (n = 12). (D) Heatmaps of receptors, ligands and transcription factors and cofactors of Notch signaling differentially expressed in sorted cycling zebrafish and murine MCs. Data are expressed as fold-changes compared to negative controls (cycling Müller from uninjured retinas). The blue boxes group the most significant genes with the highest upregulation

Fig 4: Cross-species comparison of MC gliosis in response to laser injury. (A-E) Analysis of MC gliotic response in zebrafish and mice at baseline (Uninjured) and different time point after injury (Day 1, 3, 7 and 14). Detection of GFAP in GS+MCs after laser induction in zebrafish (A.i-A.v) and mice (B.i-B.v). Shown are sections for GS (red) and GFAP (green). Detection of phospho-Erk1/2 in GS+MCs after injury and in uninjured zebrafish (C.i-C.v) and mice (D.i-D.v). Shown are sections for GS (red) and phospho-Erk1/2 (green). Histograms illustrating mean ± SD of GS+, GFAP+ and pERK1/2+cells normalized by the total of DAPI+ or GS+ cells in percentage (G). Significant differences (***p < 0.001, ****p < 0.0001) between uninjured and injured retinas were determined by post-hoc Bonferroni one-way ANOVA test (n = 12). (E-F; H) Evaluation of the proliferative potential of zebrafish and murine MCs during injury response at baseline (Uninjured) and different time point after injury (Day 1, 3, 7 and 14). Detection of PCNA in GS+MCs after injury in zebrafish (E.i-E.v) and mice (F.i-F.v). Shown are sections for GS (red) and PCNA (green). Histograms illustrating mean ± SD of PCNA+cells normalized by the total of GS+cells in percentage (H). Significant differences (****p < 0.0001) between uninjured and injured retinas were determined by post-hoc Bonferroni one-way ANOVA test (n = 12)

Fig 5: Visualization of macroglia response during wound healing in zebrafish retina after laser damage. (A) S100ß (magenta) and GS (green) expression over time in zebrafish retinal samples from day 10, day 12, and day 17 post-injury. (B) GFAP (magenta) and GS (green) expression over time in zebrafish retinal samples during wound healing. (C) Vimentin (magenta) and GS (green) expression over time in zebrafish retina sections. (D–F) Histograms depicting the respective distribution of single- or double-positive cells (white signal; marked with an arrow). No S100ß or GFAP-positive cells were found in the control samples. CTRL, control; GCL, ganglion cell layer; INL, inner nuclear layer; ONL, outer nuclear layer. Data are presented as mean ± SD (n = 3 for each time point for each combination) and significance of * p < 0.05, ** p < 0.01, and *** p < 0.001 with a two-tailed Student’s t-test. ns = not significant. Scale bars: 50 µm.