Researchers in Boston and Japan have devised a pharmacological method to transform inner ear epithelial cells into working hair cells and have used the method to restore hearing in deaf mice.1 The finding provides proof of concept for treating hearing loss by manipulating developmental pathways in the adult ear, but figuring out the precise pathway to hit in humans will require further work.

Hair cells in the inner ear transmit sound waves into electromechanical signals sensed by the nervous system. The cells can be damaged by chronic exposure to loud noise and die off in aging adults, leading to age-related hearing loss that affects up to 50 million Americans.

The inner ears of adults with the condition are thought to be functional aside from the dead hair cells. Regenerating those cells has been a goal for the hearing field. Most strategies have focused on restoring hair cells through transplantation or reactivation of latent stem cells.

Now, a Boston-Japan team has proposed an alternative strategy-transdifferentiation of existing epithelial tissue.

In mice, the researchers blocked the activity of notch 1 (Notch1), a developmental factor that influences the differentiation of embryonic stem cells into specialized structures of the cochlea.

"Notch signaling works in the embryo to influence hair cell development," said team leader Albert Edge, associate professor of otology and laryngology at Harvard Medical School and investigator at the Massachusetts Eye and Ear Infirmary. "Embryologists had previously found that development of the cochlear structure uses Notch signaling to turn a layer of undifferentiated epithelial cells into the organ of Corti, which consists of alternating layers of hair cells and supporting cells."

Edge said there was prior evidence suggesting "that after the first few weeks of life, this signaling pathway was turned down, so it was previously thought that this pathway wasn't active in adults and thus this wouldn't be relevant to hearing loss."

His team turned off the activity of Notch1 using a small molecule inhibitor of g-secretase, a proteolytic complex that activates Notch1. In cell culture and in mice, inhibition of g-secretase led to the growth of new hair cells out of a population of supporting epithelial cells.

"We tested this in adult mice and to our surprise found that a g-secretase inhibitor was effectively making new hair cells," said Edge.

Edge's team screened cultured murine inner ear stem cells with a panel of g-secretase inhibitors and found at least one compound that caused the cells to express markers associated with hair cell identity.

The next step was to test whether such treatment could yield hair cells in adult ears. Indeed, in cultured organ of Corti explants from mice engineered to lose their hair cells, the g-secretase inhibitor increased the number of hair cells and decreased the number of supporting epithelial cells compared with vehicle-treated controls.

The team then surgically introduced the molecule into the inner ear of experimentally deafened mice and observed greater numbers of new hair cells growing out of the epithelial cell layer of the organ of Corti than those seen in vehicle-treated controls. Electrophysiological experiments revealed that these new hair cells restored some degree of hearing in deafened mice.

Results were reported in Neuron, and the team included researchers at Keio University School of Medicine.

Edge's study shows "for the first time that modulation of the Notch signaling system provides a measureable degree of auditory recovery," said John Brigande, associate professor of otolaryngology at Oregon Health & Science University.

Brigande noted that hair cell loss also underlies vestibular system disorders for regulating the sense of balance. Thus, he thinks Edge's technique could be useful for a number of conditions besides acute hearing loss.

"If you're thinking about treatments for hearing loss and even vestibular disorders, this is a pharmacological system that could be used to regenerate hair cells and preserve their functions in the inner ear," said Brigande.

Edge said the findings imply that Notch signaling is active in the ears of recently deafened adult animals, which was not previously thought to be the case.

Hear here

Despite the positive findings, questions remain about the relevance of the model used in the study to common forms of hearing loss, as well as the suitability of g-secretase as a clinical target.

For example, the in vivo hearing loss model used by Edge involved acute damage by extremely loud noise rather than the chronic exposure to moderately loud noise that is thought to underlie most cases of deafness.

It thus is possible that Notch signaling is reactivated as an immediate result of noise-induced damage. If so, it remains unclear how long the pathway would remain targetable in patients with long-term hearing loss.

To answer this question, Edge plans to examine whether blocking Notch signaling can cause hair cell regeneration long after the initial loss of hearing.

Another concern is whether hitting Notch signaling will be safe in humans. g-Secretase inhibitors such as Eli Lilly and Co.'s semagacestat have encountered safety and tolerability problems in clinical trials in Alzheimer's disease (AD), which is caused by abnormal processing of amyloid precursor protein (APP), a Notch-like protein.

Some of the toxic effects of g-secretase inhibitors are thought to result from the compounds' interference with Notch signaling in the skin and intestine, which use that pathway for normal functions in adult tissue.

Indeed, Edge's team used surgical delivery of a g-secretase inhibitor to deafen mice after an oral formulation caused tolerability issues.

David Weber, president and CEO of ear drug delivery company Otonomy Inc., thinks Edge's approach will require localized delivery technology. Weber suspects that repeated or long-term application of a g-secretase inhibitor will be needed to regenerate hair cells because the compound is likely to have a short half-life inside the ear.

"This paper clearly illustrates that it's one thing to have a therapeutic agent but another thing to make it an effective therapy and bring it to the market," said Weber. "They tried to deliver it orally but encountered systemic toxicity issues and had to go to direct injection. However, the local delivery technique in this paper is not practical for clinical development in humans, as it would require surgery."

Instead, Weber advocated formulating g-secretase inhibitors in a sustained-release gel for transtympanic injection, which is Otonomy's core technology.

The company's lead candidate is OTO-104, a formulation of dexamethasone for Meniere's disease. It will enter a pivotal Phase IIb trial this year. OTO-201, a formulation of ciprofloxacin, is in Phase I testing for otitis media.

Edge said his team used surgical injection because of the difficulty of transtympanic injection into mouse ears, which are very small.

Brigande and Edge both said other developmental pathways besides Notch signaling are likely to be at play in hair cell development, citing wingless-type MMTV integration site (WNT), sonic hedgehog homolog (SHH), transforming growth factor-b (TGFB; TGF-b) and fibroblast growth factors (FGFs) as other potential targets.

Edge now plans to use a combination of mouse genetic and pharmacological assays to figure out whether hitting any of these other pathways can improve on the safety and efficacy of inhibiting g-secretase.

Edge is a cofounder of Audion Therapeutics B.V., which is working with Sanofi to develop technology to stimulate regeneration of hair cells to treat hearing loss. Edge said Audion's work is related to prior discoveries from his laboratory and does not concern the findings in the Neuron study. The status of IP related to the Neuron paper is undisclosed.

Another company, Inception 3 Inc., has licensed technology for hair cell protection and regeneration developed at Stanford University. Roche has rights to acquire Inception 3 at the time the biotech completes an IND package.

Massachusetts Eye and Ear Infirmary has filed patents based on the discoveries.

Osherovich, L. SciBX 6(2); doi:10.1038/scibx.2013.29
Published online Jan. 17, 2013


1.   Mitzutari, K. et al. Neuron; published online Jan. 9, 2013; doi:10.1016/j.neuron.2012.10.032
Contact: Albert S.B. Edge, Harvard Medical School, Boston, Mass.


Audion Therapeutics B.V., Amsterdam, the Netherlands

Eli Lilly and Co. (NYSE:LLY), Indianapolis, Ind.

Harvard Medical School, Boston, Mass.

Inception 3 Inc., San Diego, Calif.

Keio University School of Medicine, Tokyo, Japan

Massachusetts Eye and Ear Infirmary, Boston, Mass.

Oregon Health & Science University, Portland, Ore.

Otonomy Inc., San Diego, Calif.

Roche (SIX:ROG; OTCQX:RHHBY), Basel, Switzerland

Sanofi (Euronext:SAN; NYSE:SNY), Paris, France

Stanford University, Stanford, Calif.