Gene Therapy Restores Balance and Auditory Functions in a Mouse Model of Usher Syndrome

Abstract

Dizziness and hearing loss are among the most common disabilities. Many forms of hereditary balance and hearing disorders are caused by abnormal development of stereocilia, mechanosensory organelles on the apical surface of hair cells in the inner ear. The deaf whirler mouse, a model of human Usher syndrome (manifested by hearing loss, dizziness, and blindness), has a recessive mutation in the whirlin gene, which renders hair cell stereocilia short and dysfunctional. In this study, wild-type whirlin cDNA was delivered to the inner ears of neonatal whirler mice using adeno-associated virus serotype 2/8 (AAV8-whirlin) by injection into the posterior semicircular canal. Unilateral whirlin gene therapy injection was able to restore balance function as well as improve hearing in whirler mice for at least 4 months. Our data indicate that gene therapy is likely to become a treatment option for hereditary disorders of balance and hearing.

Keywords: vestibular dysfunction; whirler; whirlin.

Figure 1

AAV8-Whirlin Gene Therapy Restores Whirlin Expression in Infected Utricular Hair Cells (A) Schematic of the study. The inner ear is important for balance function. Vestibular hair cells detect rotational inputs as well as linear and gravitational acceleration by the deflection of their stereocilia bundles. This information is transmitted by the superior and inferior vestibular nerves to the central vestibular system for processing. In this study, we examined whether AAV8-whirlin gene therapy delivered through the posterior semicircular canal can restore vestibular hair cell stereocilia morphology and balance function in whirler mice. The predominant long isoform of whirlin used in this study encodes a scaffolding protein with three PDZ domains and one proline-rich domain. The region recognized by the anti-whirlin antibody is indicated by the purple line. ab, antibody. (B) Utricular hair cells in a phenotypically normal control mouse (whrn^+/wi, P30) show immunolocalization of whirlin (green) at the tips of the stereocilia (red). Examples of whirlin expression are shown with white arrows. (C) Utricular hair cells in a whirler untreated control mouse (whrn^wi/wi, no gene therapy, P30) lack whirlin expression at the stereocilia tips. In addition, whirler utricular hair cells have aberrantly short stereocilia bundles. (D) AAV8-whirlin gene therapy delivered via the posterior semicircular canal at P4 restored whirlin expression at the stereocilia tips and initiated elongation of stereocilia in infected utricular hair cells in a P30 whirler mouse. Examples of whirlin expression are indicated with white arrows. (E) Distribution of utricular hair cell infection efficiency in all whirler ears that received AAV8-whirlin gene therapy.

Figure 2

AAV8-Whirlin Gene Therapy Restores Stereocilia Length in Whirler Utricles (A) Stereocilia length measurements of utricular hair cells. The “normal control” group includes data from wild-type and heterozygous littermates (whrn^+/+ and whrn^+/wi, n = 4). The “whirler untreated control” group includes data from whirler mice (whrn^wi/wi) that did not receive AAV8-whirlin gene therapy (n = 5). The “whirler w/ AAV8-whirlin” group includes data from all whirler ears (whrn^wi/wi) that received AAV8-whirlin gene therapy, including ones with low or no infection (n = 28). The “AAV8-whirlin contralateral” group includes data from the contralateral whirler ears (whrn^wi/wi) that did not receive AAV8-whirlin gene therapy (n = 28). The “>40% infection” group includes data from whirlers with >40% AAV8-whirlin infection in utricular hair cells (n = 18). The “>40% contralateral” group includes data from the contralateral whirler ears that did not receive AAV8-whirlin gene therapy in the >40% infection group (n = 18). The “whirler w/ AAV8-GFP” group includes data from whirlers that received AAV8-GFP as a sham surgery control. All stereocilia length measurements were made using confocal images with 3D rendering. Error bars represent SEs. *p < 0.05, ***p < 0.001. (B) Scanning electron microscopy images showing whirler untreated control (whrn^wi/wi, no gene therapy) vestibular hair cells that have aberrantly short stereocilia bundles (top panel), and whirler vestibular hair cells with elongated stereocilia bundles after AAV8-whirlin gene therapy (bottom panel). Examples of stereocilia length are shown with red brackets. (C) There was a strong positive correlation (R² = 0.67) between utricular hair cell infection efficiency and stereocilia length.

Figure 3

AAV8-Whirlin Gene Therapy Reduces Circling Behavior in Whirler Mice (A) Representative track plots from a phenotypically normal control (whrn^+/wi), a whirler untreated control (whrn^wi/wi, no gene therapy), and a whirler mouse that received AAV8-whirlin gene therapy, demonstrating that AAV8-whirlin gene therapy significantly reduced circling behavior in whirler mice. (B) Quantification of circling behavior testing was performed at ∼P30. Error bars represent SEs. **p < 0.01, ***p < 0.001. (C) Linear regression analysis revealed a strong negative correlation (R² = 0.44) between utricular hair cell infection efficiency and circling behavior, indicating the greater the infection efficiency, the less the circling behavior.

Figure 4

AAV8-Whirlin Gene Therapy Improves Swimming and Rotarod Performance in Whirler Mice (A) A well-established swim scoring system was used to measure swim performance, with 0 representing normal swimming and 3 representing underwater tumbling requiring immediate rescue. Testing was performed between P90 and P120. (B) The duration of time each animal remained balanced on the rotarod apparatus was recorded. Testing was performed between P90 and P120. Error bars in both figures represent SEs. *p < 0.05, **p < 0.01.

Figure 5

AAV8-Whirlin Gene Therapy Improves Hearing in Whirler Mice (A) Representative ABR recordings from a phenotypically normal control (whrn^+/wi), a whirler untreated control (whrn^wi/wi, no gene therapy), and a whirler that received AAV8-whirlin gene therapy. In 8 of 29 whirler mutant mice, measurable ABR thresholds were obtained after AAV8-whirlin gene therapy. (B) ABR thresholds at the four measured frequencies (4, 8, 16, and 32 kHz). The ABR thresholds from the eight whirler mice that had improved hearing are shown individually (dashed lines). An ABR threshold of 100 dB represents no hearing. ABR testing was done at ∼P30. (C) Representative confocal images of the inner hair cells (IHCs) from the cochlear apex in a whirler mouse that received AAV8-whirlin gene therapy in the left ear (top panel) and the contralateral ear that did not receive AAV8-whirlin gene therapy (bottom panel). Images were taken at P120. The treated ear has robust whirlin expression (green) at the stereocilia tips in all IHCs, whereas the contralateral non-treated ear shows no whirlin expression. In addition, several IHCs have degenerated in the non-treated ear (white arrows). The stereocilia (red) lengths in infected whirler IHCs are significantly longer than the non-infected ones. (D) Stereocilia length measurements at different regions of the cochlea in AAV8-whirlin-treated and contralateral (non-treated) ears. The whirler ears that received AAV8-whirlin had significantly longer stereocilia compared to the contralateral non-treated ears. In one animal, the contralateral cochlea was damaged during dissection, leaving seven cochleas for analyses. The stereocilia length measurements from normal controls from a previous study are also shown for comparison (these measurements were made in the exact same way by the same investigators). Error bars represent SEs. ***p < 0.001.

Figure 6

The Effects of AAV8-Whirlin Gene Therapy Are Long Lasting (A) Comparison of circling behavior at 1 month and 4 months of age in seven whirler mice that received AAV8-whirlin gene therapy. Five of seven whirler mutants (71.4%) had sustained reductions in circling behavior. (B) Utricular hair cells in a whirler mouse at P120 showing robust whirlin expression (green) at the stereocilia tips with elongated stereocilia (red). The inset (right) shows stereocilia at higher magnification. (C) IHC counts from AAV8-whirlin-treated whirler ears and the contralateral non-treated ears at P120. The AAV8-whirlin-treated whirler ears had more surviving inner hair cells (IHCs) compared to the contralateral non-treated ears across the entire cochlea. Error bars represent SEs. (D) Representative confocal images of the IHCs from cochlear middle turn in a whirler mouse that received AAV8-whirlin gene therapy in the left ear (top panel) and the contralateral ear that did not receive AAV8-whirlin gene therapy (bottom panel). Images were taken at P120. The treated ear has robust whirlin expression (green) at the stereocilia tips in all IHCs, whereas the contralateral non-treated ear shows no whirlin expression. In addition, several IHCs have degenerated in the non-treated ear (white arrows). Stereocilia (red) lengths in infected whirler IHCs were significantly longer than the non-infected ones.