Failure of fluid absorption in the endolymphatic sac initiates cochlear enlargement that leads to deafness in mice lacking pendrin expression

Abstract

Mutations of SLC26A4 are among the most prevalent causes of hereditary deafness. Deafness in the corresponding mouse model, Slc26a4(-/-), results from an abnormally enlarged cochlear lumen. The goal of this study was to determine whether the cochlear enlargement originates with defective cochlear fluid transport or with a malfunction of fluid transport in the connected compartments, which are the vestibular labyrinth and the endolymphatic sac. Embryonic inner ears from Slc26a4(+/-) and Slc26a4(-/-) mice were examined by confocal microscopy ex vivo or after 2 days of organ culture. Culture allowed observations of intact, ligated or partially resected inner ears. Cochlear lumen formation was found to begin at the base of the cochlea between embryonic day (E) 13.5 and 14.5. Enlargement was immediately evident in Slc26a4(-/-) compared to Slc26a4(+/-) mice. In Slc26a4(+/-) and Slc26a4(-/-) mice, separation of the cochlea from the vestibular labyrinth by ligation at E14.5 resulted in a reduced cochlear lumen. Resection of the endolymphatic sacs at E14.5 led to an enlarged cochlear lumen in Slc26a4(+/-) mice but caused no further enlargement of the already enlarged cochlear lumen in Slc26a4(-/-) mice. Ligation or resection performed later, at E17.5, did not alter the cochlea lumen. In conclusion, the data suggest that cochlear lumen formation is initiated by fluid secretion in the vestibular labyrinth and temporarily controlled by fluid absorption in the endolymphatic sac. Failure of fluid absorption in the endolymphatic sac due to lack of Slc26a4 expression appears to initiate cochlear enlargement in mice, and possibly humans, lacking functional Slc26a4 expression.

Figure 1. Schematic diagram of the cochlea.

A) Diagram based on a cochlea obtained from an E18.5 Slc26a4^+/− mouse. B) Diagram based on a cochlea obtained from an E18.5 Slc26a4^−/− mouse, see Fig. 3. Abbreviations: C, otic capsule; S, stria vascularis; H, sensory hair cells; M, modiolus; N, cochlear nerve. Note, that spaces occupied by mesenchymal cell (green) are compressed in Slc26a4^−/− mice. Particular, fibrocytes in the modiolus (M) and between the otic capsule (C) and stria vascularis (S) are displaced.

Figure 2. Cochlear development from E12.5 to P3 in Slc26a4^+/− mice.

Na⁺/K⁺ ATPase (red) was visualized by immunocytochemistry. F-actin (green) and nuclei (blue) were labeled. A-F: Six different stages of development ranging from E12.5 to P3 are shown. Abbreviations: SV, stria vascularis; OS, outer sulcus; K, Kölliker's organ; RM, Reissner's membrane; Ca, otic capsule; SL, spiral ligament. The thickness of the otic capsule is marked by dashed lines.

Figure 3. Cochlear lumen formation progressed from the base to the apex.

A: Base of the cochlea. B: Apex of the same cochlea. F-actin (green) and nuclei (red) were labeled. Abbreviations: SV, stria vascularis; OS, outer sulcus; K, Kölliker's organ; RM, Reissner's membrane.

Figure 4. Cochlear lumen formation in Slc26a4^+/− and Slc26a4^−/− mice.

A, D and G: Luminal area measurements of cross-section #1 in cochleae obtained from E14.5, E16.5 and E18.5 littermates, respectively. The number next to the bars represents the N number of animals. Significant differences are marked (*). B and C, cochlear cross-sections of E14.5 Slc26a4^+/− and Slc26a4^−/− littermates. E and F, cochlear cross-sections of E16.5 Slc26a4^+/− and Slc26a4^−/− littermates. H and I, cochlear cross-sections of E18.5 Slc26a4^+/− and Slc26a4^−/− littermates. Scala media was highlighted in pink. Images without the pink overlay are available in Figure S1 (Supporting information). Arrows point to cross-section #1, which was used for digital area measurements. F-actin (green) was labeled. Cochlear cross-sections were imaged by laser-scanning microscopy.

Figure 5. Cochlear lumen development in Slc26a4^+/− and Slc26a4^−/− mice.

A, lumen development in Slc26a4^+/− mice. B, lumen development in Slc26a4^−/− mice. Numbers next to symbols represent the N number of Slc26a4^+/− and Slc26a4^−/− littermates. Figure preceded by ‘x’ indicate the factor between measurements in Slc26a4^+/− and Slc26a4^−/− littermates.

Figure 6. Isolation of the cochlea from the vestibular labyrinth by ligation.

A: Diagram of the inner ear. Abbreviations: S, saccule; U, utricle; A, Ampulla; ED, endolymphatic duct; ES, endolymphatic sac. B: Photograph of a ligated inner ear obtained from an E14.5 Slc26a4^+/− mouse.

Figure 7. Fluid secretion in the vestibular labyrinth ‘pumps up’ the cochlea during lumen formation.

A, D and G: Luminal area measurements (average of the two basal cross-sections) in cochleae with and without vestibular labyrinth. Cochleae were harvested at E14.5 and E17.5 from Slc26a4^+/− and Slc26a4^−/− mice and maintained two days in organ culture. The number next to the bars represents the N number of animals. Significant differences are marked (*). B and C, cochlear cross-sections of E14.5 Slc26a4^+/− mice with and without vestibular labyrinth. E and F, cochlear cross-sections of E14.5 Slc26a4^−/− mice with and without vestibular labyrinth. H and I, cochlear cross-sections of E17.5 Slc26a4^+/− mice with and without vestibular labyrinth. Scala media was highlighted in pink. Images without the pink overlay are available in Figure S1 (Supporting information). F-actin (green) and nuclei (blue) were labeled. Cochlear cross-sections were imaged by laser-scanning microscopy. Arrows point to the two basal cross-sections, which were used for digital area measurements.

Figure 8. Resection of the endolymphatic sac.

A: Diagram of the inner ear. B: Photograph of an inner ear obtained from an E17.5 Slc26a4^−/− mouse. Abbreviations: S, saccule; U, utricle; A, Ampulla; ED, endolymphatic duct; ES, endolymphatic sac.

Figure 9. Fluid absorption in the endolymphatic sac ‘drains’ the cochlea during lumen formation.

A, D and G: Luminal area measurements (average of the two basal cross-sections) in cochleae with and without endolymphatic sac. Cochleae were harvested at E14.5 and E17.5 from Slc26a4^+/− and Slc26a4^−/− mice and maintained two days in organ culture. The number next to the bars represents the N number of animals. Significant differences are marked (*). B and C, cochlear cross-sections of E14.5 Slc26a4^+/− mice with and without endolymphatic sac. E and F, cochlear cross-sections of E14.5 Slc26a4^−/− mice with and without endolymphatic sac. H and I, cochlear cross-sections of E17.5 Slc26a4^+/− mice with and without endolymphatic sac. Scala media was highlighted in pink. Images without the pink overlay are available in Figure S1 (Supporting information). F-actin (green) and nuclei (blue) were labeled. Cochlear cross-sections were imaged by laser-scanning microscopy. Arrows point to the two basal cross-sections, which were used for digital area measurements.

Figure 10. Diagram of ion transport in the endolymphatic sac.

The endolymphatic sac epithelium consists mainly of ribosomal-rich cells that are interspersed by mitochondrial-rich cells. Mitochondrial-rich cells express H⁺ ATPase and the Cl⁻/HCO₃ ⁻ exchanger pendrin in their apical membrane. Ribosomal-rich cells express Na⁺ channels including ENaC. A current generated by the H⁺ ATPase drives Na⁺ reabsorption via Na⁺ channels. The role of the Cl⁻/HCO₃ ⁻ exchanger pendrin is to export HCO₃ ⁻ that is generated by carbonic anhydrase (CA) in the reaction that leads to the generation of H⁺.