Single-cell RNA sequencing of GER cell-derived organoids.

(A) Fluorescent reporter gene expression in a P2 organ of Corti vibratome section of an Fgfr3-Cre-ERT2/Ai14-tdTomato/Sox2-GFP transgenic mouse. Scale bar: 20 µm. DCs: Deiters’ cells; OPC: outer pillar cell; IPC: inner pillar cell; IPhC: inner phalangeal cell; IBC: inner border cell; IHC: inner hair cell; OHCs: outer hair cells; GER: greater epithelial ridge. (B) Enrichment strategy of GER cells from Fgfr3-Cre-ERT2/Ai14-tdTomato/Sox2-GFP transgenic mouse. Dissociated P2 cochlear floor cells were sorted from individual FACS gates (Gates 1-4) into defined positions in 96-well plates and subjected to single-cell RNA sequencing (scRNA-seq). Well-position metadata were then matched with cell type identities defined by clustering analysis to quantify the proportion of GER cells in each gate. A cell population expressing high GFP and lacking tdTomato fluorescence (Gate 1) contained GER cells with 93% purity 9,13. (C) Single-cell harvesting and sequencing from GER-derived organoids. FACS-enriched GER cells from Gate 1 were cultured for organoid growth in suspension culture for 7 days. Organoids were harvested on culture days 1 (D1), 3 (D3), and 7 (D7). A representative image of D7 organoids is shown (scale bar: 200 µm). Manual harvesting of single cells from dissociated organoids using a glass pipette of 50-60 µm diameter yielded a total library of 966 cells. Of these, scRNA-seq data were obtained from 768 cells, and quality control filtering retained 658 cells for downstream analysis. GER-derived organoids, harvested and transferred to an adherent culture system on day 7, underwent expansion for 14 days, resulting in otic colonies that harbor epithelial structures with nascent hair cells and supporting cells on culture day 21 (D21) 9. For quantification at D7, we used organoids obtained in 7 independent experiments.

Single-cell RNA sequencing of GER-derived organoids reveals proliferating cell clusters.

(A) Seurat data analysis of 658 cells, visualized by Uniform Manifold Approximation and Projection (UMAP), identified seven distinct clusters. Each dot represents a cell, colored by cluster. (B) UMAP projection, visualized by timepoint, showing the origin of cells from D1, D3, and D7 organoids. (C) Robust expression of mRNA encoding proliferation marker Kiel 67 (Mki67) was detected in the cells represented by clusters 1 and 3. (D) Volcano plot showing differentially expressed genes between the cells at the onset of proliferation (clusters 4, 5, and 6) and the remaining cells (clusters 1, 2, 3, and 7). Thresholds are indicated with dotted lines and were set to an average log2FC > |1| and adjusted p-value < 1E-5. The Wilcoxon Rank Sum test was used for the comparison.

Expression of galectin-1, galectin-3, and Myc in organoids.

(A, B) Day 3 organoids generated from whole cochlear duct cells. Two representative organoids are shown for each labeling. (A) Immunofluorescence labeling for galectin-1 (grey), Ki67 (green), and nuclei (DAPI, blue). Scale bars: 20 µm. (B) Immunofluorescence for galectin-3 (grey), Ki67 (green), and nuclei (DAPI, blue). Scale bars: 10 µm. (C-E) Single-cell RNA-sequencing data analysis comparing the mRNA expression levels of galectin-1 (C), galectin-3 (D), and Myc (E) between organoids and the P2 cochlear floor. (Left) Violin plots. Yellow bars: median lines. (Center) mRNA expression levels projected on the UMAP plots for GER-organoid cells and (right) for cochlear floor cells. DC: Deiters’ cells; GER: greater epithelialridge; IBC: inner border cells; IPhC: inner phalangeal cells; OPC: outer pillar cells; IPC: inner pillar cells; HC: hair cells.

Inhibition of galectin-1, galectin-3, and Myc reduces organoid growth.

(A, D, G) Dose-dependent inhibition of organoid formation by OTX008 (galectin-1 inhibitor) (A), GB1107 (galectin-3 inhibitor) (D), and 10058-F4 (Myc inhibitor) (G). Organoids generated from whole cochlear duct cells of P2 FVB/NJ mice after 7 days in culture in the presence of increasing inhibitor concentrations. Scale bar: 200 µm. (B, E, H) Organoid number (left) and size (right) quantification after 7 days in culture at serial concentrations of inhibitors (OTX008 (B), GB1107 (E), and 10058-F4 (H)). Organoids were centered in the field of view of an inverted microscope in 48-well plates, and quantification was performed using acquired images. For organoid number quantification, medians are shown as solid lines. For organoid size, medians and quartiles are shown as dotted lines. Statistical analysis was performed using Dunnett’s multiple comparisons test in one-way ANOVA. *p <0.05, **p <0.01, ***p <0.001, **** p <0.0001, ns: not significant. (C, F, I) Live cell quantification of organoids using a colorimetric viable cell mass assay after 7 days in culture. Shown are the means ± SD of N = 3-4 independent experiments. Statistical analysis was performed using Dunnett’s multiple comparisons test in one-way ANOVA. *p <0.05, **p <0.01, ***p <0.001, **** p <0.0001, ns: not significant.

GER cell-derived organoid growth is strongly attenuated with inhibitors targeting galectin-1, galectin-3, and Myc.

(A) Organoids generated from FACS-enriched GER cells after seven days in culture. Dose-dependent inhibition of organoid formation was observed with OTX008 (galectin-1 inhibitor), GB1107 (galectin-3 inhibitor), and 10058-F4 (Myc inhibitor). Scale bars: 200 µm. (B) Live cell quantification of organoid cells using a colorimetric viable cell mass assay after seven days in culture in increasing concentrations of OTX008, GB1107, and 10058-F4. Means ± SDs of N = 3-4 independent experiments are shown. Statistical analysis was performed using Dunnett’s multiple comparisons test in one-way ANOVA. *p <0.1, **p <0.01, ***p <0.001, **** p <0.0001, ns: not significant.

AAV-mediated overexpression of galectin-1, galectin-3, and Myc in organoids generated from neonatal cochlear duct cells.

(A) Schematic illustration of the organoid formation assay using dissociated P2 FVB/NJ mouse cochlear duct cells infected with AAVs for each gene of interest (AAV-GOI). Effective gene overexpression was presumed to start within two days post-infection (dpi), and cell proliferation was assessed at 9 dpi. (B) qPCR for Mki67 mRNA in organoids infected with AAV-lgals1 or AAV-myc, compared to organoids infected with AAV-EGFP. Statistical analysis was performed using Dunnett’s multiple comparisons test in one-way ANOVA. *p <0.05, ***p <0.001, ****p <0.0001, ns: not significant. (C) mRNA expression changes for Lgals1 and Myc were quantified using qPCR in organoids infected with AAV-lgals1 (left) and AAV-myc (right) at 3, 5, 7, and 9 dpi, compared to organoids infected with AAV-EGFP for each respective time point. Statistical analysis was performed using Dunnett’s multiple comparisons test in one-way ANOVA. *p <0.05, **p <0.01, ***p <0.001, ns: not significant. (D) Cochlear duct cells were infected with AAV-EGFP, AAV-lgals1, or AAV-myc at titers of 4×109 or 8×109 viral genomes per milliliter (vg/mL). AAV-lgals1 and AAV-myc visibly enhanced organoid formation compared to the control (AAV-EGFP) at both viral titers at 9 dpi. Scale bar: 200 µm. (E) Quantification of organoid number (left) and size (right) at 9 dpi following infection with AAV-EGFP, AAV-lgals1, or AAV-myc at 4×109 or 8×109 vg/mL. Organoids were centered in the field of view of an inverted microscope in 48-well plates, and quantification was performed using acquired images. For organoid number quantification, medians are shown as solid lines. For organoid size, medians and quartiles are shown as dotted lines. Statistical analysis was performed using Dunnett’s multiple comparisons test in one-way ANOVA. *p <0.05, **p <0.01, **** p <0.0001, ns: not significant. (F) Live cell quantification of the organoids using a colorimetric viable cell mass assay. Means ± SDs are shown. N = 8 and N = 4 for the experiments with 4×109 and 8×109 vg/mL viral titers. Statistical analysis was performed using Dunnett’s multiple comparisons test in one-way ANOVA. *p <0.05, **p <0.01, ns: not significant. (G) FACS quantification of live cells in organoids at 9 dpi following infection with AAV-EGFP, AAV-lgals1, or AAV-myc at 4×109 vg/mL. (Left three panels) Individual live cell populations lacking dead cell stain cells (detected in the allophycocyanin (APC) channel), are gated in dot plots. (Right top two panels) Cells from AAV-EGFP-and AAV-lgals1-infected organoids, as well as cells from AAV-EGFP-and AAV-myc-infected organoids, are overlaid to show the expression of EGFP and mCherry encoded in AAV-EGFP and AAV-myc, respectively. (Right bottom panel) Percentage of live cells in organoids at 9 dpi following infection with AAV-EGFP, AAV-lgals1, or AAV-myc at 4×109 vg/mL. Statistical analysis was performed using Dunnett’s multiple comparisons test in one-way ANOVA. ***p <0.001, ns: not significant. (H) Cochlear duct cells were infected with AAV-EGFP or AAV-lgals3 at titers of 4×109 or 8×109 vg/mL. AAV-lgals3-infected organoids exhibited no significant difference in growth compared to the control (AAV-EGFP) at both viral titers at 9 dpi. Scale bar: 200 µm. (I) Quantification of organoid number (left) and size (right) at 9 dpi following infection with AAV-EGFP, or AAV-lgals3 at 4×109 or 8×109 vg/mL. Organoids were centered in the field of view of an inverted microscope in 48-well plates, and quantification was performed using acquired images. For organoid number quantification, medians are shown as solid lines. For organoid size, medians and quartiles are shown as dotted lines. Statistical analysis was performed using the two-tailed Mann-Whitney U test. ns: not significant. (J) Live cell quantification of organoids using a colorimetric viable cell mass assay. Means ± SDs of N = 3-4 experiments are shown. Statistical analysis was performed using a two-tailed, unpaired t-test. ns: not significant.

Overexpression of galectin-1 and Myc in organoids generated from postnatal day 14-17 cochlear epithelial cells.

(A) (a1-3) Immunofluorescence detection of Myosin 7a (Myo7a) and CD326/EpCAM in a P14 FVB/NJ mouse cochlea vibratome section. Hair cells are labeled for Myosin 7a. CD326/EpCAM was detected in cochlear floor epithelial cells and the cells of the stria vascularis. The GER was absent at this more mature age. SV: stria vascularis; TM: tectorial membrane. Scale bar: 50 µm. (b1-4) Magnified images of the areas surrounded by the dotted boxes in a2 (b1,2) and a3 (b3,4). Scale bar: 25 µm. (B) Schematic illustration of the organoid formation assay timeline using dissociated P14-17 FVB/NJ mouse cochlear duct cells infected with AAVs for each gene of interest (AAV-GOI). CD326-positive epithelial cells were enriched by Magnetic Activated Cell Sorting (MACS). Organoids were infected with AAV-GOIs. Gene overexpression was presumed to start within two days post-infection, and cell proliferation was assessed at 9 dpi. (C) MACS-enriched CD326-positive cochlear cells were infected with AAV-EGFP, AAV-lgals1, or AAV-myc at a titer of 4×109 vg/mL. AAV-myc infection visibly enhanced organoid formation, while AAV-lgals1-infected organoids exhibited no sifnificant difference in growth compared to the control (AAV-EGFP). Arrowheads: organoids. The cellular debris likely represents the remnants of cochlear cells that did not grow into organoids at 9 dpi. Scale bar: 100 µm. (D) Quantification of organoid number (left) and size (right) at 9 dpi following infection with AAV-EGFP, AAV-lgals1 or AAV-myc at 4×109 vg/mL. Organoids were centered in the field of view of an inverted microscope in 48-well plates, and quantification was performed using acquired images. For organoid number quantification, medians are shown as solid lines. For organoid size, medians and quartiles are shown as dotted lines. Statistical analysis was performed using Dunnett’s multiple comparisons test in one-way ANOVA. *** p <0.001, **** p <0.0001, ns: not significant. (E) Live cell quantification of the organoids using a colorimetric viable cell mass assay. Means ± SDs of three experiments are shown. Statistical analysis was performed using Dunnett’s multiple comparisons test in one-way ANOVA. *p <0.05, ns: not significant.

Galectin-1 upregulation in the GER in the Lgr5-DTR mouse damage model.

(A) Schematic illustration of the Lgr5-DTR mouse damage model. Intraperitoneal injection of diphtheria toxin (DT) in P1 mice induces Lgr5-driven cell ablation. Cell loss and concurrent cell proliferation in the lateral GER occur by P4. New IPhCs emerge by P7. (B) Seurat data analysis with 3,773 cells obtained from P4 Lgr5-DTR damage model mice (Ki67CreERT2/+; Lgr5DTR/+; Rosa26RtdTomato/+) and control (Ki67CreERT2/+; Rosa26RtdTomato/+). Eight clusters representing medial GER (MGER), lateral GER (LGER), IPhCs (this cluster also includes IBCs), and proliferating cells (Prolif) are visualized in a UMAP projection. (C) UMAP visualization of the clusters split by damage and control mouse models. Cluster 8 (arrow) is predominantly composed of cells after damage. (D) Violin plot showing the downregulation of Lgr5 expression in the damage model compared to the control. (E) Marker gene expression in GER (Sparcl1 and Gsn), proliferating cells (Mki67 and Top2a), MGER (Calb1 and Epyc), LGER (Igfbp3 and Tsen15), and IPhC (Lfng and Fabp7), projected onto the UMAP plot. A Log2 expression scale is shown for each UMAP. (F) Volcano plot showing differential gene expression in cluster 8 (proliferating cells) compared to the remaining clusters. (G) Lgals1 (left) and Myc (right) expression, projected onto the UMAP plot, and visualized individually for the damage model and control cells. The arrow points to cells with high expression of Lgals1. (H) Differentially expressed genes were analyzed in cells from the damage model compared to the control, individually for MGER (clusters 2, 3, 4, 5, and 6) (upper) and LGER (cluster 1) (lower) clusters. The fold change expression of selected genes is shown with lollipop plots for MGER and LGER clusters. (I) Immunohistology of P4 cochleae from control and Lgr5-DTR mice (Lgr5DTR/+) post-DT treatment revealed upregulation of galectin-1 in the medial and lateral GER. Proliferating cells labeled with Ki67 were observed in the lateral GER, accompanied by a loss of Fabp7-expressing cells (IPhC and IBC) along the apex-to-base gradient. Solid yellow arrows indicate the GER, while dotted yellow arrows mark areas with visible galectin-1 expression. IHC: inner hair cell; OHC: outer hair cell; IPhC: inner phalangeal cell; IBC: inner border cell. Scale bars: 20 µm.

In vivo administration of the galectin-1 inhibitor OTX008 prevents GER cell proliferation.

(A) A schematic illustrating the DT and OTX008 treatment schedule in Lgr5-DTR (Lgr5DTR/+) and control mice. (B) Immunohistology of P4 cochleae obtained from control and Lgr5-DTR mice treated with DT at P1, representing intact and damage controls, respectively, and from control and Lgr5-DTR mice treated with OTX008 (10 mg/kg per injection) at P2 and P3. Stacked images of the apical turns of the cochleae are presented at three levels: IPhC apical membrane, IPhC nuclei, and GER (scale bars: 20 µm). In the IPhC nucleus-level images, magnified views of the IBC/IPhC nuclei are shown for the areas outlined by squares (scale bars: 10 µm). Fabp7-positive debris lacking nuclei, along with pyknotic cells, were observed in Lgr5-DTR cochleae treated with DT (damage control) or DT+OTX008. Asterisks indicate IPhCs and IBCs. Solid yellow arrows indicate the GER. (C) Quantification of Fabp7-positive and Ki67-positive cells in the apical, middle, and basal turns of the cochleae, corresponding to Figures 7B and S5A, B. **** p<0.0001, *** p<0.001, ** p<0.01, * p<0.05, ns: not significant. (D) P4 cochleae obtained from control and Lgr5-DTR mice treated with DT at P1, followed by OTX008 (10 mg/kg per injection) only at P2. Stacked images of the apical turns of the cochleae are presented at three levels, as described in (B). Scale bars: 20 µm and 10 µm (magnified images). Fabp7-positive debris lacking nuclei, along with pyknotic cells, were observed in Lgr5-DTR cochleae treated with DT+OTX008 (P2). Asterisks indicate IPhCs and IBCs. Solid yellow arrows indicate GER. (E) Quantification of Fabp7-positive and Ki67-positive cells in the apical, middle, and basal turns of the cochleae, corresponding to Figures 7D and S6A, B. **** p<0.0001, *** p<0.001, ** p<0.01, * p<0.05, ns: not significant.