AAV-Mediated Inner Ear Gene Delivery

According to the World Health Organization, more than 430 million people worldwide live with disabling hearing loss, including an estimated 34 million children, with that number projected to approach 700 million by 2050. Genetic factors account for up to 60 percent of congenital and early-onset hearing loss. Conventional interventions such as hearing aids and cochlear implants can only compensate for lost function, whereas gene therapy mediated by adeno-associated virus (AAV) is beginning to make an outright cure possible. This guide condenses the core elements of AAV-based inner ear therapeutics — target-cell biology, serotype tropism, promoter strategy, delivery routes, and representative clinical translation case studies — into a practical, decision-oriented reference for experimental design.

1. Inner Ear Architecture and Target-Cell Biology

The ear is organized into three compartments: the outer ear (pinna and canal, which collects sound), the middle ear (tympanic membrane and three ossicles, which amplify vibration), and the inner ear, embedded deep in the temporal bone, which houses the cochlea (hearing) and the vestibular system (balance). The cochlea is a spiral structure divided into three fluid-filled compartments: the perilymph-filled scala vestibuli and scala tympani, and the endolymph-filled scala media, whose high potassium concentration (roughly 150 mM) sustains the +80 mV endocochlear potential. Because perilymph and endolymph differ markedly in ionic composition, an AAV vector delivered through the round window membrane enters the perilymph-filled scala tympani and must then cross the spiral lamina to reach hair cells bathed in endolymph — a physical barrier that limits transduction efficiency and must be factored into vector design.

Precise gene delivery depends on understanding four principal cell populations. Hair cells (HCs) are the sensory effectors: roughly 3,500 inner hair cells (IHCs) per human ear convert mechanical vibration into glutamate release that drives spiral ganglion neurons (SGNs) — mutations in OTOF disrupt this synaptic step and cause DFNB9 deafness — while about 12,000 outer hair cells (OHCs) per ear actively amplify sound via the motor protein prestin (SLC26A5) and are especially vulnerable to noise, ototoxic drugs, and genetic defects. Critically, mammalian hair cells do not regenerate once mature, which is precisely why gene therapy delivered before hair cell death offers a form of intervention no other treatment can replicate. Supporting cells (SCs) maintain cochlear ionic homeostasis and are the key target of hair-cell regeneration strategies such as Atoh1-driven trans-differentiation; GJB2 (connexin 26), expressed predominantly in SCs, is the single most common cause of hereditary hearing loss worldwide. The stria vascularis secretes potassium into the endolymph to maintain the endocochlear potential, and mutations in KCNQ1/KCNE1 or KCNQ4 disrupt this process. SGNs are the primary afferent auditory neurons — type I (about 95 percent) synapse with IHCs, and type II (about 5 percent) synapse with OHCs — and are an important target for neuroprotective gene therapy.

2. Choosing the Delivery Vehicle: AAV Serotype Tropism in the Inner Ear

Different AAV serotypes show markedly different affinities for inner ear cell types, and this tropism is strongly age- and species-dependent. The table below synthesizes the most clinically and experimentally relevant serotypes (most data from mouse models).

Selecting a serotype means weighing target cell type, animal age/species, and clinical-translation goals together. For broad hair-cell coverage, Anc80L65 (neonatal) or AAV2.7m8 (all ages) are the leading choices; for IHC-only targeting, AAV-ie or AAV9 suffice; for supporting cells, AAV-ie or AAV-KP1 are preferred; for SGNs, Anc80L65 is suitable; and for the stria vascularis, AAV8BP2 is the recommended option. Neonatal mice (P0–P5) respond well to Anc80L65, AAV-ie, or AAV2.7m8, while nonhuman primate work favors AAV-S. For clinical translation, AAV2.7m8 offers ophthalmic precedent (NCT03326336), AAV-ie has the strongest human-tissue validation, and AAV1 is already advanced in otologic trials (see Case 3). Promoter strategy, addressed next, should be integrated into this decision from the outset.

3. Installing the Precision Switch: Promoter and Regulatory Element Selection

If the serotype determines which door the vector can enter, the promoter determines which lights turn on once inside. AAV’s limited packaging capacity of roughly 4.7 kb, including the inverted terminal repeats (ITRs), makes promoter choice a critical design constraint.

Natural promoters force a trade-off between specificity and strength: broad, strong promoters like CAG are efficient but risk off-target expression, while cell-specific promoters such as native Myo15a are precise but often too weak. Enhancer engineering resolves this trade-off by recombining conserved non-coding elements into synthetic constructs — the B8 enhancer (Case 2, below) is a leading example that combines high specificity with high expression strength. As a rule of thumb, use CBA or EF1α for basic proof-of-concept work, Myo15a (full-length or truncated) for hair-cell-specific therapeutic constructs, GFAP for supporting-cell targeting, and Syn1 for SGN work. Packaging capacity remains the binding constraint throughout: AAV holds roughly 4.4 kb of usable space after the ITRs, so large genes such as OTOF (~5.9 kb) or MYO15A (~11 kb) require a dual-AAV system and compact promoters, since oversized cargo sharply reduces both production yield and transduction efficiency.

4. Breaking Through the Final Barrier: Inner Ear Delivery Routes

The chosen delivery route directly affects both the distribution and transduction efficiency of an AAV vector, and because the inner ear is deeply embedded and structurally fragile, the method must balance efficacy, safety, and clinical translatability.

Typical injection volumes range from 0.5–1.5 µL in neonatal/juvenile mice (P0–P7) to 1.0–2.0 µL in adults (>P21), with viral doses of 1×10¹⁰–1×10¹¹ vg/ear for reporter studies and 5×10¹⁰–5×10¹¹ vg/ear for functional rescue; the bilateral DFNB9 trial described in Case 3 used 1.5×10¹² vg per ear in 50 µL at 120 nL/min. All procedures require aseptic technique, appropriate anesthesia (hypothermia for P0–P3 pups, isoflurane from P4 onward), injection rates of 50–150 nL/min to avoid pressure injury, and postoperative analgesia. ABR and DPOAE should be assessed at baseline, 2–4 weeks, and 4–8 weeks post-injection, and a graded-dose pilot study is recommended before scaling up, since excessive dose or volume can itself damage hair cells.

5. Representative Translational Case Studies

Case 1: Base Editing for Hereditary Hearing Loss

A team led by Yilai Shu (Fudan University Eye & ENT Hospital), with Sangsu Bae and Sang-Yeon Lee (Seoul National University Hospital), used PAM-flexible adenine base editing to correct the MPZL2 c.220C>T founder mutation — identified in a 1,437-family cohort as the cause of roughly 15.5 percent of pediatric DFNB111 hearing loss cases in East Asian populations — in a humanized mouse model. After screening 14 base-editor/sgRNA combinations, the team selected ABE8eWQ-SpRY with sgRNA3 (over 50 percent in vitro editing efficiency, low bystander and off-target activity) and delivered it via a split-intein dual-AAV-ie system (CMV/U6 promoters, 5×10¹³ GC/mL, round window microinjection in P2 mice, 2 µL total). Treated animals showed ABR thresholds approaching wild-type levels at 8 kHz for at least 20 weeks, restored outer hair cell and Deiters’ cell counts, normalized MPZL2 mRNA/protein expression, and no detectable off-target editing or ototoxicity — demonstrating that base editing combined with AAV-ie delivery can durably reverse a clinically significant founder mutation.

Figure 1. Dual AAV-ie-ABE8eWQ-SpRY:sgRNA3 vectors correct the MPZL2 c.220C>T mutation in vivo, showing the split-intein delivery schematic and on-target editing efficiency with minimal bystander or off-target activity.

Case 2: Enhancer Engineering for Specificity and Strength

A team led by Guisheng Zhong (ShanghaiTech University) developed ARBITER, a platform for dissecting regulatory elements in a tissue too small for conventional ChIP-seq or ATAC-seq, and applied it to Slc26a5 (prestin), whose loss causes severe OHC-driven hearing loss. Two conserved intronic elements, E1 (247 bp) and E2 (525 bp), act cooperatively: E2 alone drives OHC-specific expression and E1 substantially boosts it, while combined knockout of both abolished prestin expression and caused profound deafness. Native E1+E2-driven gene therapy restored prestin to only about 25–50 percent of wild-type levels with incomplete hearing recovery, but a synthetic enhancer built from duplicated regulatory modules (B8) drove near-wild-type prestin expression — 8 of 11 treated mice showed complete ABR/DPOAE recovery with normal OHC morphology, and B8 caused no hearing damage in adult mice where CAG did. The study shows that specificity and expression strength, often treated as a trade-off, can be jointly optimized through modular enhancer engineering.

Figure 2. The synthetic B8 enhancer achieves safe, OHC-specific transduction in adult mice, avoiding the hearing damage seen with the broadly active CAG promoter.

Case 3: Bilateral OTOF Gene Therapy Restores Binaural Hearing

A team led by Yilai Shu and Wuqing Wang (Fudan University Eye & ENT Hospital) ran a single-arm trial of bilateral AAV1-hOTOF gene therapy in five children (ages 1–11) with DFNB9, a form of auditory neuropathy caused by biallelic OTOF mutations that disable otoferlin-dependent synaptic transmission despite normal hair-cell morphology. Because full-length OTOF cDNA (~5.9 kb) exceeds AAV’s packaging limit, the team used a dual-vector overlapping-recombination strategy — N-terminal and C-terminal AAV1 vectors under a hair-cell-specific Myo15a promoter — that reconstitutes full-length otoferlin intracellularly once both vectors enter the same cell, delivered via simultaneous bilateral round-window injection (1.5×10¹² vg per ear, 50 µL, 120 nL/min) to avoid the neutralizing-antibody barrier that staggered dosing would create. All five patients tolerated treatment with only mild, self-limited adverse events and no dose-limiting toxicity; ABR thresholds improved from no response (>95 dB) at baseline to 50–85 dB by 13–26 weeks, speech-perception scores rose substantially (one patient’s MAIS score rose from 1 to 28 of 40), and — uniquely enabled by bilateral treatment — sound-source localization was restored, with one patient’s localization error improving from 92.8° to 40.0°. The trial demonstrates both the feasibility of a dual-AAV strategy for oversized genes and the added functional value of treating both ears simultaneously.

Figure 3. Auditory brainstem response (ABR) and auditory steady-state response (ASSR) thresholds for patients 1–5, showing the shift from no response at baseline to measurable thresholds after bilateral AAV1-hOTOF treatment.

6. A Quick-Reference Decision Guide for Inner Ear AAV Applications

Four considerations cut across every row of this table. Age is the single most important variable — transduction efficiency for the same serotype can differ by an order of magnitude between neonatal and adult ears, particularly for OHCs — so use an age-appropriate model or results may fail to replicate. Promoter size shapes vector strategy, since gene plus promoter plus enhancer must stay within roughly 4.4 kb or a dual-vector approach becomes necessary. Surgical quality is the largest source of inter-experiment variability, so include a positive control (a standard serotype with known efficiency, plus CAG-GFP) and a consistent ABR baseline protocol. Finally, build immune-response testing into the plan — check peripheral blood for anti-AAV neutralizing antibodies one to two weeks after local injection to rule out pre-existing immunity as a confound.

7. Frequently Asked Questions

Q1: IHC transduction looks good, but OHC or supporting-cell transduction is poor. What should I check? This usually reflects a mismatch between serotype and target cell population, and the right fix depends on age, species, and target. For hair cells, try Anc80L65 or AAV2.7m8, and consider PSCC delivery in adults; for supporting cells, AAV-ie and its optimized variants are preferable. Run a small pilot screen of candidate serotypes before scaling up.

Q2: AAV packaging efficiency and titer are low, and expression is unstable. What’s going wrong? Keep the ITR-to-ITR sequence well under the 4.7 kb ceiling — closer to 3.0–3.5 kb is optimal. For large genes, split the coding sequence across a dual- or triple-AAV system or use a validated functional truncation, and confirm the final construct by full-length sequencing.

Q3: Transgene expression peaks early, then declines. How can I prevent this? Choose a stable promoter such as EF1α or an optimized cell-specific promoter, and avoid promoters prone to methylation-driven silencing. Adding enhancer elements or insulator sequences (such as cHS4) can also help.

Q4: A serotype works well in neonates but fails in adults. Why, and what should I do about it? In the neonatal window (P0–P7), RWM delivery is generally most effective, while PSCC delivery is worth considering in adults for OHC transduction. Track the natural progression of the target disease and intervene as early as feasible — once hair cells have already degenerated, a regenerative (supporting-cell trans-differentiation) strategy may be more appropriate than direct gene replacement.

8.Practical AAV Design Principles for Inner Ear Research

For researchers developing AAV-based inner ear studies, several principles can improve experimental success:

• Define the primary target cell before selecting a capsid.
• Consider animal age, species, and disease stage early in study design.
• Use human-relevant or translational models when possible.
• Match the promoter or enhancer to the target cell and payload.
• Avoid unnecessary overexpression in fragile cochlear cells.
• Confirm that the total cassette size is compatible with AAV packaging.
• Use dual-AAV strategies when therapeutic genes exceed AAV capacity.
• Include appropriate control AAV vectors.
• Evaluate both molecular expression and functional hearing outcomes.
• Monitor auditory and vestibular safety.
• Use well-characterized AAV preparations with strong analytical support.

Conclusion

AAV-mediated inner ear gene delivery is transforming the study and treatment of hereditary hearing loss. The field has moved from early proof-of-concept experiments toward clinically validated and regulatory-recognized applications, particularly for OTOF-related deafness. At the same time, the inner ear remains one of the most technically demanding targets for gene delivery.

Successful AAV inner ear research requires an integrated strategy that considers target-cell biology, capsid tropism, promoter and enhancer design, payload size, delivery route, vector quality, and functional safety. No single serotype, promoter, or delivery method is optimal for all applications. Instead, each project should be designed around the biology of the target gene, the affected cell type, the disease stage, and the translational goal.

As capsid engineering, enhancer design, dual-vector systems, base editing, and high-quality AAV manufacturing continue to advance, AAV-based inner ear gene therapy is positioned to expand beyond a small number of monogenic indications and support broader research into hearing restoration, cochlear protection, and auditory system repair.

How PackGene Supports AAV Inner Ear Gene Therapy Research

PackGene provides customized AAV vector design, packaging, production, purification, serotype selection, and analytical testing services to support inner ear gene delivery research. For hearing loss and cochlear gene therapy projects, PackGene can help researchers design AAV strategies that consider target cell type, capsid selection, promoter or enhancer choice, payload size, dual-AAV requirements, expression goals, and downstream quality control.

PackGene supports a broad AAV serotype portfolio, including commonly used wild-type and engineered capsids for gene delivery research. Combined with quality-focused AAV production and analytical characterization, PackGene helps researchers develop AAV-based tools for inner hair cell targeting, outer hair cell targeting, supporting cell studies, spiral ganglion neuron research, disease modeling, gene replacement, gene editing, and preclinical inner ear gene therapy development.

References

1. Zhao, Y., Zhang, L., Wang, D., Chen, B., & Shu, Y. (2022). Approaches and Vectors for Efficient Cochlear Gene Transfer in Adult Mouse Models. Biomolecules, 13(1), 38.
2. Zhang, L., Wang, H., Xun, M., Tang, H., Wang, J., Lv, J., … Shu, Y. (2025). Preclinical evaluation of the efficacy and safety of AAV1-hOTOF in mice and nonhuman primates. Molecular Therapy Methods & Clinical Development, 33(3), 101575.
3. Lv, J., Wang, H., Cheng, X., Chen, Y., Wang, D., Zhang, L., … Shu, Y. (2024). AAV1-hOTOF gene therapy for autosomal recessive deafness 9: a single-arm trial. The Lancet, 403(10441). https://doi.org/10.1016/S0140-6736(23)02874-X
4. Iranfar, S., Cornille, M., Roldan, M. S., Plion, B., Lecomte, M.-J., Safieddine, S., & Lahlou, G. (2025). Cell tropism of adeno-associated viruses within the mouse inner ear in vivo: from embryonic to adult stages. Scientific Reports, 15(1).
5. Ballana, E., Wang, J., Venail, F., Estivill, X., Puel, J.-L., Arbonès, M. L., & Bosch, A. (2008). Efficient and specific transduction of cochlear supporting cells by adeno-associated virus serotype 5. Neuroscience Letters, 442(2).
6. Emptoz, A., Michel, V., Lelli, A., Akil, O., Boutet de Monvel, J., Lahlou, G., … Safieddine, S. (2017). Local gene therapy durably restores vestibular function in a mouse model of Usher syndrome type 1G. Proceedings of the National Academy of Sciences, 114(36).
7. Isgrig, K., Ishibashi, Y., Lee, H. J., Zhu, J., Grati, M., Bennett, J., … Chien, W. W. (2022). AAV8BP2 and AAV8 transduce the mammalian cochlear lateral wall and endolymphatic sac with high efficiency. Molecular Therapy – Methods & Clinical Development, 26.
8. Han, S., Xu, Z., Wang, S., Tang, H., Hu, S., Wang, H., … Shu, Y. (2024). Distributional comparison of different AAV vectors after unilateral cochlear administration. Gene Therapy, 31(3–4).
9. Isgrig, K., McDougald, D. S., Zhu, J., Wang, H. J., Bennett, J., & Chien, W. W. (2019). AAV2.7m8 is a powerful viral vector for inner ear gene therapy. Nature Communications, 10(1).
10. György, B., Meijer, E. J., Ivanchenko, M. V., Tenneson, K., Emond, F., Hanlon, K. S., … Corey, D. P. (2019). Gene Transfer with AAV9-PHP.B Rescues Hearing in a Mouse Model of Usher Syndrome 3A and Transduces Hair Cells in a Non-human Primate. Molecular Therapy – Methods & Clinical Development, 13.
11. Tao, Y., Liu, X., Yang, L., Chu, C., Tan, F., Yu, Z., … Wu, H. (2022). AAV-ie-K558R mediated cochlear gene therapy and hair cell regeneration. Signal Transduction and Targeted Therapy, 7(1).
12. Ivanchenko, M. V., Hanlon, K. S., Hathaway, D. M., Klein, A. J., Peters, C. W., Li, Y., … Corey, D. P. (2021). AAV-S: A versatile capsid variant for transduction of mouse and primate inner ear. Molecular Therapy – Methods & Clinical Development, 21.
13. Kim, M.-A., Ryu, N., Kim, H.-M., Kim, Y.-R., Lee, B., Kwon, T.-J., … Kim, U.-K. (2019). Targeted Gene Delivery into the Mammalian Inner Ear Using Synthetic Serotypes of Adeno-Associated Virus Vectors. Molecular Therapy – Methods & Clinical Development, 13.
14. Aaron, K. A., Pekrun, K., Atkinson, P. J., Billings, S. E., Abitbol, J. M., Lee, I. A., … Cheng, A. G. (2023). Selection of viral capsids and promoters affects the efficacy of rescue of Tmprss3-deficient cochlea. Molecular Therapy – Methods & Clinical Development, 30.
15. Shubina-Oleinik, O., Nist-Lund, C., French, C., Rockowitz, S., Shearer, A. E., & Holt, J. R. (2021). Dual-vector gene therapy restores cochlear amplification and auditory sensitivity in a mouse model of DFNB16 hearing loss. Science Advances, 7(51).
16. Akil, O., Dyka, F., Calvet, C., Emptoz, A., Lahlou, G., Nouaille, S., … Lustig, L. R. (2019). Dual AAV-mediated gene therapy restores hearing in a DFNB9 mouse model. Proceedings of the National Academy of Sciences, 116(10).
17. Al‐Moyed, H., Cepeda, A. P., Jung, S., Moser, T., Kügler, S., & Reisinger, E. (2019). A dual‐AAV approach restores fast exocytosis and partially rescues auditory function in deaf otoferlin knock‐out mice. EMBO Molecular Medicine, 11(1).
18. US Patent 2021/0388045A1.
19. Hu, S. W., Lv, J., Wang, Z., Tang, H., Wang, H., Wang, F., … Li, H. (2024). Engineering of the AAV-Compatible Hair Cell-Specific Small-Size Myo15 Promoter for Gene Therapy in the Inner Ear. Research, 7.
20. Rio, C., Dikkes, P., Liberman, M. C., & Corfas, G. (2002). Glial fibrillary acidic protein expression and promoter activity in the inner ear of developing and adult mice. Journal of Comparative Neurology, 442(2).
21. Keppeler, D., Merino, R. M., Lopez de la Morena, D., Bali, B., Huet, A. T., Gehrt, A., … Moser, T. (2018). Ultrafast optogenetic stimulation of the auditory pathway by targeting‐optimized Chronos. The EMBO Journal, 37(24).
22. Gene Therapy vs Cochlear Implantation in Restoring Hearing Function and Speech Perception for Individuals With Congenital Deafness. JAMA Neurology. Published online July 21, 2025. https://doi.org/10.1001/jamaneurol.2025.2053
23. Hu, S. W., Jeong, S., Jiang, L., Koo, H., Wang, Z., Choi, W. H., … Shu, Y. (2025). PAM-flexible adenine base editing rescues hearing loss in a humanized MPZL2 mouse model harboring an East Asian founder mutation. Nature Communications, 16(1), 7186. https://doi.org/10.1038/s41467-025-62562-8
24. Zhao, S., Yang, Q., Yu, Z., Chu, C., Dai, S., Li, H., … Zhong, G. (2025). Deciphering enhancers of hearing loss genes for efficient and targeted gene therapy of hereditary deafness. Neuron, 113(10). https://doi.org/10.1016/j.neuron.2025.03.023
25. Wang, H., Chen, Y., Lv, J., Cheng, X., Cao, Q., Wang, D., … Shu, Y. (2024). Bilateral gene therapy in children with autosomal recessive deafness 9: single-arm trial results. Nature Medicine, 30(7), 1898–1904. https://doi.org/10.1038/s41591-024-03023-5
26. Akil O, et al. “Surgical Method for Virally Mediated Gene Delivery to the Mouse Inner Ear through the Round Window Membrane.” JoVE, Issue 97, 2015. PMCID: PMC4401361.
27. Isgrig K & Chien WW. “Posterior Semicircular Canal Approach for Inner Ear Gene Delivery in Neonatal Mouse.” JoVE, Issue 133, 2018. PMCID: PMC5931426.
28. World Health Organization. (2021). Deafness and hearing loss. https://www.who.int/news-room/fact-sheets/detail/deafness-and-hearing-loss
29. Shearer, A. E., Hildebrand, M. S., Odell, A. M., et al. Genetic Hearing Loss Overview. In: GeneReviews, NCBI Bookshelf. https://www.ncbi.nlm.nih.gov/books/NBK1434/
30. Wu, Z., Yang, H., & Colosi, P. (2010). Effect of genome size on AAV vector packaging. Molecular Therapy, 18(1), 80–86; see also overview at https://pmc.ncbi.nlm.nih.gov/articles/PMC9910337/