The timing of auditory sensory deficits in Norrie disease has implications for therapeutic intervention

Abstract

Norrie disease is caused by mutation of the NDP gene, presenting as congenital blindness followed by later onset of hearing loss. Protecting patients from hearing loss is critical for maintaining their quality of life. This study aimed to understand the onset of pathology in cochlear structure and function. By investigating patients and juvenile Ndp-mutant mice, we elucidated the sequence of onset of physiological changes (in auditory brainstem responses, distortion product otoacoustic emissions, endocochlear potential, blood-labyrinth barrier integrity) and determined the cellular, histological, and ultrastructural events leading to hearing loss. We found that cochlear vascular pathology occurs earlier than previously reported and precedes sensorineural hearing loss. The work defines a disease mechanism whereby early malformation of the cochlear microvasculature precedes loss of vessel integrity and decline of endocochlear potential, leading to hearing loss and hair cell death while sparing spiral ganglion cells. This provides essential information on events defining the optimal therapeutic window and indicates that early intervention is needed. In an era of advancing gene therapy and small-molecule technologies, this study establishes Ndp-mutant mice as a platform to test such interventions and has important implications for understanding the progression of hearing loss in Norrie disease.

Keywords: Development; Endothelial cells; Genetic diseases; Mouse models; Otology.

Figure 1. Sequential audiograms of patients with Norrie disease showing onset of hearing loss.

(A–F) Hearing levels at a range of frequencies and multiple ages for each individual. (A and B) Audiograms of right and left ears of the patient (N05) between 3 and 8 years showing onset of mild hearing loss at low frequencies at age 6. Note a progression each year in high frequencies. (C and D) Audiograms of right and left ears of the patient (N08) between 4 and 19 years. Note the presence of hearing loss at 4 years of age, and the tented shape and gradual progression across frequencies over the years. (E and F) Audiograms of right and left ears of the patient (N07) between 6 and 36 years showing onset of hearing loss at age 8 years, normal levels at 9 years, and subsequent drop in hearing at 11 years, affecting the right ear first. Note the skew toward loss of high frequencies and the marked jump of thresholds in adolescence in this patient. (G and H) Change over time in hearing levels at selected, speech-relevant, frequencies in 4 individuals.

Figure 2. Early onset of hearing impairment in Ndp-KO mice.

ABR (A and B) and DPOAE (C and D) thresholds and endocochlear potentials (EPs) (E and F) for individual animals were plotted as mean ± SD from WT mice (blue) and Ndp-KO mice (red) aged 1 month and 2 months. Ck, click. (A) No difference in ABR thresholds between genotypes at 1 month (Mann-Whitney rank sum test; U = 1050, T = 2226, P = 0.4540). (B) Ndp-KO mice show hearing loss in low frequencies at 2 months (U = 2754.4, T = 5382.5, P = 0.0026). n = 11 WT, n = 12 Ndp-KO at 1 month; n = 9 WT, n = 13 Ndp-KO at 2 months. (C and D) DPOAEs were comparable between genotypes at 1 month (U = 1252.5, T = 2527.5, P = 0.1326). At 2 months, Ndp-KO thresholds increased over all frequencies compared with controls (U = 689.5, T = 1724.5, P < 0.00000001). n = 10 WT, n = 12 Ndp-KO at 1 month; n = 9 WT, n = 13 Ndp-KO at 2 months. (E) At 1 month, EP was significantly lower in Ndp-KO than in WT, but within normal range (>100 mV); n = 11 WT, n = 12 Ndp-KO. (F) At 2 months, Ndp-KO EP decreased further; n = 7 WT, n = 12 Ndp-KO (2-way ANOVA; significant effect of age, F = 4.3400, P = 0.0440; significant effect of genotype, F = 29.9970, P = 0.0000297). There was a significant reduction at 1 month (E) (Holm-Šidák method for multiple comparisons; t = 3.5946, P = 0.00092067) and at 2 months (F) (Holm-Šidák method for multiple comparisons; t = 4.1323, P = 0.00019031). n = 11 WT, n = 12 Ndp-KO at 1 month; n = 7 WT, n = 12 Ndp-KO at 2 months.

Figure 3. SGN survival in Ndp-KO mice.

(A) Apical to basal spiral ganglia of WT and Ndp-KO mice at 1 month showing that Tubb3 is expressed in the SGNs and Kir4.1 is expressed in the surrounding satellite glial cells. (B) Number of Tubb3-positive SGNs per 10,000 μm² in apical to basal spiral ganglia of WT and Ndp-KO mice at 1 month. (C) Average size (μm²) of SGNs in apical to basal spiral ganglia of WT and Ndp-KO mice at 1 month. (D) Apical to basal spiral ganglia of WT and Ndp-KO mice at 11–12 months showing that Nf200 is expressed in the spiral SGNs and Mbp is expressed in the surrounding satellite glial cells. (E) Number of Nf200-positive SGNs per 10,000 μm² in apical to basal spiral ganglia of WT and Ndp-KO mice at 11–12 months. (F) Average size (μm²) of SGNs in apical to basal spiral ganglia of WT and Ndp-KO mice at 11–12 months. High-magnification images of SGNs are shown in Supplemental Figure 3. n = 3 WT, n = 3 Ndp-KO analyzed at each time point; bars indicate mean ± SD. Analyzed with 1-way ANOVA, Holm-Šidák correction for multiple comparisons; *P ≤ 0.05, **P ≤ 0.01; NS, P ≥ 0.05. Scale bars: 100 μm.

Figure 4. OHC loss in Ndp-KO cochlea.

(A–F) Characteristic appearance of the organ of Corti of the WT and Ndp-KO at 2 months. White, anti–myosin VIIA (Myo7a) immunostaining; blue, DAPI. (A and B) In WT, IHCs (arrowheads) and OHCs (arrows) survive; example taken from middle region. (C and D) In Ndp-KO, IHCs survive (arrowheads), but the OHCs are degenerated in the middle region; arrows point to an example of a surviving OHC. (E and F) Low-magnification images of WT and Ndp-KO organ of Corti, showing the loss of OHCs; fractional distance from the apex is demarcated by lines; arrows point to comparable locations where they survive in WT (E). (G) Quantification of OHC loss across the length of the cochlea at 2 months (Myo7a staining). WT (blue) and Ndp-KO (red), mean ± SD, n = 7. Analysis with a mixed-model 2-way ANOVA and Šidák’s post hoc test indicated significant effects of both region (P < 0.0001) and genotype (P < 0.0001), and interaction of region and genotype (P < 0.0001). Šidák’s post hoc test indicated significant differences between comparable regions of WT and Ndp-KO in the apical regions at 2/8 to 4/8. n = 7 WT, n = 7 Ndp-KO analyzed; bars indicate mean ± SD. **P < 0.01. Regions of the organ of Corti defined as fractional distance from the apex (1/8 to 8/8): region 1/8 (3.1–6.1 kHz), region 2/8 (6.1–10.0 kHz), region 3/8 (10.0–15.0 kHz), region 4/8 (15.0–21.6 kHz), region 5/8 (21.6–30.2 kHz), region 6/8 (30.2–41.3 kHz), region 7/8 (41.3–55.9 kHz), region 8/8 (55.9–74.8 kHz). Scale bars: 20 μm (A and B), 250 μm (C and D).

Figure 5. Electron microscopy of the cochlear lateral wall shows abnormalities of capillaries and marginal cells in Ndp-KO.

Stria vascularis of 1-month-old WT (A) and Ndp-KO (B–E) mice and 2-month-old WT (F) and Ndp-KO (G–J) mice. (A) Middle turn of the cochlea in WT at 1 month shows normal architecture. MC, marginal cell; IC, intermediate cell; BC, basal cell; CAP, enclosed capillary. (B–E) In Ndp-KO at 1 month, large spaces surround the capillaries in apical coils (arrows in B) and extend to basal coils (arrows in C) in some animals. The capillaries are surrounded by shrunken and disrupted cells (D) and large extracellular spaces (D and E) that fill with material that stains with heavy metals to a density similar to that of plasma inside the capillaries (E). Loosely distributed fibrils (arrows in E) are present within the spaces close to the capillary. Most marginal cells are largely unaffected (D); they retain intense staining of the cytoplasm, appearing “dark,” and the extensive infoldings of the basolateral membrane with the characteristic elongated scalloped morphology of their nuclei located close beneath the luminal plasma membrane. D and E are higher magnifications of B and C, respectively. (F) Stria vascularis of WT at 2 months. (G–J) In Ndp-KO at 2 months, the capillaries are surrounded by intact cells filling the spaces apparent at 1 month (G and H). The stria appears thinner than at 1 month (H and I), and many capillaries appear enlarged (H; asterisks). (I) Marginal cells occupy a lower area, have lost their basal infoldings (I and J) and electron-dense staining, appear more rounded, and show changes in nuclear morphology (J). n = 6 WT, n = 6 Ndp-KO at 1 and 2 months. Scale bars: 10 μm (A, D, and I), 1 μm (E), 2 μm (J).

Figure 6. Marginal cell morphology and permeability of vascular barrier in Ndp-KO stria vascularis.

(A–D) Anti-ZO1 immunostaining demonstrates evenly sized marginal cells at 1 month in the WT (A) and Ndp-KO (B) lateral wall; n = 3 WT, n = 4 Ndp-KO. At 2 months, WT marginal cells remain evenly sized (C), while those in Ndp-KO (D) are unevenly sized; n = 3 WT, n = 3 Ndp-KO. (E–L) Permeability assessments at P20 and 1 and 2 months using a fluorescent tracer assay (green, FITC-BSA; blue, vessels counterstained for endomucin). (E–J) FITC-BSA is restricted within the vessels (arrowheads) in WT, but detectable outside them in Ndp-KO (arrows). (K and L) At 2 months, FITC-BSA signal is uneven in Ndp-KO stria with patches of weak (arrowheads) and strong (arrows) intensity. Scale bars: 30 μm (A–D), 50 μm (E–J), 500 μm (K and L). (M) Extravascular FITC intensity is elevated in Ndp-KO compared with WT (2-way ANOVA; Šidák’s post hoc test showed P < 0.05 at P20; see also Supplemental Figure 5). Mean ± SD is represented; n = 3 WT, n = 3 Ndp-KO at 1 and 2 months. (N–P) Quantitative reverse transcriptase PCR analyses of vascular barrier and permeability marker gene expression. In Ndp-KO cochlea, Cldn5 expression was significantly reduced from P20 (N), but Plvap and Cav1 increased progressively from 1 and 2 months, respectively (O and P), compared with WT. Normalized fold change ± SD; 2-way ANOVA with Šidák’s post hoc test. n = 6 WT, n = 3 KO at P10; n = 7 WT, n = 6 KO at P20; n = 8 WT, n = 6 KO at 1 month; n = 6 WT, n = 7 KO at 2 months. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, ****P ≤ 0.0001.

Figure 7. Postnatal onset of morphological abnormalities in stria vascularis capillaries in Ndp-KO.

The stria vascularis of 1-month-old mice was examined with an antibody targeting endomucin. (A and B) Panels display the middle-apical region of the stria vascularis cropped from a 3D image of whole cochleae from WT (A) and Ndp-KO (B) mice. The regions indicated by perforated boxes in A and B are displayed in adjacent panels. White arrows indicate dilated capillaries, white arrowheads indicate narrow capillaries and regions with sparse vessels. (C–H) The apical region of the stria vascularis was isolated and examined with GS-IB4. WT and Ndp-KO were examined at P10 (C and D), P20 (E and F), and 1 month (G and H). White arrowheads indicate intercapillary regions that are small and circular/oval in structure. (I) Quantification of vessel diameter and shape descriptor measurements of intercapillary regions for P10 stria vascularis; n = 5 WT, n = 5 Ndp-KO; bars indicate mean ± SD. (J) Quantification of vessel diameter in P20 and 1 month stria vascularis; n = 3 WT, n = 3 Ndp-KO; bars indicate mean ± SD. I (circularity) and J analyzed with an unpaired t test, I (vessel diameter and solidity) analyzed with Mann-Whitney test; *P < 0.05, **P < 0.01. Scale bars: 50 μm.

Figure 8. Abnormal pericyte coverage of capillaries in the lateral wall affecting the spiral ligament and stria vascularis.

(A–V) Stria vascularis (SV) (A–H) and spiral ligament (SL) capillaries (I–V), immunostained with vessel marker anti-endomucin (Emcn), pericyte marker anti-desmin (Des), and vascular tight junction marker anti–claudin-5 (Cldn5) antibodies. (A and E) At P10, pericytes (Des, white) are evenly distributed across the SV (Emcn, blue) in both WT (A) and Ndp-KO (E). (B–H) At 2 months, Ndp-KO strial capillaries (F–H) are either high-endomucin, low–claudin-5, low-pericyte-coverage vessels (white arrows, F–H) or low-endomucin, high–claudin-5, dense-pericyte-coverage vessels (red arrows, F–H), compared with the even staining patterns of WT strial capillaries (B–D). Scale bars in A–H: 30 μm. (I–O) At P10 and 2 months, WT SL capillaries (I–K) form regular branches (Emcn, I, J, and N) with even pericyte coverage (desmin, K, L, and N; white) and claudin-5 staining (M; cyan). J and K are a higher magnification of I. (P–V) In contrast, Ndp-KO SL capillaries are abnormal from P10, showing either irregular patterns of high-endomucin-staining meshworks (P, white arrows) with filopodia (Q, yellow arrows) and low pericyte coverage (R: Des, white; white arrows), or, alternatively, low endomucin staining (P–T, red arrows) and abnormally dense pericyte wrapping (R and U: Des, white; red arrows). Q and R are a higher magnification of P. Like in the SV, abnormal vessels with high pericyte coverage show claudin-5 (V, red arrow), whereas vessels loosely covered with pericytes show low or absent claudin-5 (V, white arrows), compared with the pattern in WT (N and O). Scale bars in I–V: 100 μm (I and P), 20 μm (all others). n = 9 WT, n = 8 Ndp-KO analyzed at P10, and n = 14 WT, n = 15 Ndp-KO analyzed at 2 months, for desmin/endomucin/claudin-5 costaining.