Hearing loss has become one of the most common disabilities in the United States and can affect almost every age group. The number of people with hearing loss worldwide has been steadily increasing over recent years, reaching almost 49 million people in the US alone [1]. According to the National Institute on Deafness and Other Communication Disorders (NIDCD) 2010 statistics, approximately 17% of the American adult population experiences hearing loss and 3 out of every 1000 children are born deaf. The prevalence of hearing loss increases with age, as about 47% of adults over 75 years old have hearing impairment (NIDCD). Although hearing loss may not be life threatening, it can greatly influence the patient's quality of life, social interactions, and have a significant financial impact on society [1,2].

The ear is of tremendous importance in sensing the world around us. Aside from being the prime organ for the perception of sound, it also plays a crucial role in balance. The inner ear is a highly specialized sense organ with a complex structure and has been referred to as a labyrinth [3]. It contains hair cells arranged in a highly organized pattern and is innervated by sensory neurons. Acoustic energy, in the form of sound waves, is channeled into the ear canal where it strikes the tympanic membrane. As the energy hits the stapes located at the oval window, a pressure wave sets the cochlear fluid into motion. The hair bundles deflect as a result of the shearing force caused by the cochlear fluid and the stereocilia slide along one another. The movement on the stereocilia results in their depolarization and opens up gated calcium ion channels. This is followed by the release of neurotransmitters from the base of the hair cell. Primary auditory afferents then conduct the signal to the brainstem.

Sensorineural hearing loss (SNHL) involves damage to the cochlea (inner ear sensory hair cells) or the eighth nerve. It is irreversible and in most cases a hearing aid is required. Common causes for SNHL are aging, ototoxic drugs, noise induced trauma, inner ear concussion, and immune disorders [4-6]. Although only a small percentage of the cases can be treated medically and surgically, advances in molecular and stem cell therapies may provide tools to treat the irreparable damage of hair cells caused by SNHL [7].

Hair cells are sensory receptors located in the inner ear. They appear to be “hair-like” because of the numerous ciliary processes called stereocilia that extend from their surfaces. Hair cells are responsible for converting sound into electrical signals that are sent to the brain via the auditory nerve for processing. In the human cochlea, hair cell death can occur due to a variety of causes, such as age related deafness (presbycusis), a high dosage of ototoxic drugs (e.g., gentamycin, cisplatin, aminoglycosides), genetic disorders, infectious diseases, or high levels of noise exposure [3,8,9]. Patients exposed to high doses of aminoglycosides as a treatment regimen for bacterial infections often experience hair cell death. The aminoglycosides are known to induce cell death via activation of the intracellular caspase signaling pathway and trigger mitochondria to release cytochrome c into the cytoplasm, which consequently induces apoptosis via caspase-3 activation in hair cells. Noise induced cellular stress activates the JNK signaling pathway and causes neuronal cell death via necrosis. Necrotic hair cell death is less common than apoptotic cell death and is mainly induced by trauma or disease [3,9,10].

Over the past 30 years, several attempts have been made to understand the pathways involved in hair cell formation and death [11]. The process of hair cell regeneration was considered impossible to occur in higher vertebrates until two groups serendipitously discovered the amazing phenomenon in birds in the late 1980s [12]. At the same time, Jørgensen and Mathiesen showed that there were mitotic activity and continuous proliferation of hair cells in the vestibular epithelium of adult parakeets [13]. Ever since, there has been steady progress in understanding the mechanics and pathways of avian and mammalian (rat, mouse, guinea pig) hair cell regeneration [8,10,14-21]. Significant development in hair cell regeneration is outlined in the following reviews [7,22-29]. In this review, we explore how different genes, viral vectors and stem cell sources can be used in conjunction with each other to engineer inner ear hair cells and develop strategies in restoring the function of the inner ear.

It is important to identify and characterize genes that govern the ontogeny and differentiation of the cochlear epithelium; identifying such genes can lead to the design of several therapeutic approaches for sensory epithelial cell development and hair cell differentiation. A critical gene responsible for inner ear development is the Atonal gene—a protein belonging to the basic helix-loop-helix (bHLH) family of transcription factors that activates the E-box dependent transcription. Atoh1 has a unique auto regulatory enhancer element containing an E-box in the 3′ region of the gene [43]. Math1, also known as Atoh1, is the mouse homolog of the Drosophila melanogaster Atonal gene. The Math1 gene is essential for the differentiation of sensory hair cells from previously established sensory primordium and is limited to only a subpopulation of the non-sensory supporting cells, primarily the pillar cells [44,45]. Studies with embryonic Math1-null mice reported a failure to produce hair cells and deletion of Atoh1 using Pax2-Cre resulted in degeneration of cells in the organ of corti in mice [46], proving Math1 as a positive regulator in directing hair cell differentiation [47]. Gene delivery studies in guinea pigs, mice, and rats reported an over expression of Math1 in non-sensory cells, resulting in the production of ectopic immature hair cells outside the sensory epithelium via the transdifferentiation mechanism [16,44,48-52]. The non-sensory Math1 expressing cells attracted auditory nerve fibers and developed into mature hair cells [49,50].

The other homologues of the Atonal gene are Cath1 (chicken atonal homolog), Xath1 (Xenopus atonal homolog) and Hath1 (human atonal homolog), although Math1 is the most extensively studied and used transcription factor [53,54]. Studies with adenoviral expression of Hath1 in rats showed hair cell production without supporting cell proliferation [55].

Additional genes involved in the control of supporting cell fate include Hes1, Hes5, BETA2/Neurod1, Jagged2 and Notch Signaling [18,19]. Hes1 and Hes5 have been shown to influence supporting cell fate through negative regulation of Math1 [56,57]. Certain cell cycle kinases also influence inner ear development by regulating cell cycle and inhibiting hair cell differentiation (Refer Table 2). BETA2/Neurod1 gene has been shown to regulate the formation of sensory and neuronal ganglions in both cochlear and vestibular systems [58]. Table 2 gives a list of the different genes involved in hair cell differentiation.

Figure 3 represents a schematic on the interaction of different genes and their contribution to positive and negative regulation of Math1 transcription factor in neonates and during the embryonic development of the cochlea. (A) Hair cells express hair cell-specific Math1 transcription factor and notch ligands—Delta1 and Jagged2. Notch receptor (N) binds to the ligands, cleaves with the help of γ-secretase and releases Notch Intracellular Domain (NICD). NICD enters the nucleus of supporting cells and activated Hes1/Hes5 transcription factors. Hes1/Hes5 proteins inhibit Math1 gene expression. Alternatively, expression of Cdkn1b (p27^kip1) and Cdkn2d (p19^Ink4d) in early progenitor supporting cells repress Math1 expression and maintain supporting cell fate. (B) In the presence of γ-secretase inhibitors, the notch receptor fails to cleave and release the NICD, thus inhibiting the activation of Hes1/Hes5 that would otherwise down regulate Math1 expression. Similarly, targeted deletion of p27^kip1 and p19^Ink4d genes allows ectopic expression of Math1 resulting in supernumerary hair cells. These pathways can be induced or inhibited via standard or molecular therapy and additionally can be used to control the differentiation of stem cells.

Current therapies for treating hearing loss involve the use of either hearing aids or cochlear implants. Cochlear implants are only available to patients with severe hair cell damage and profound loss of hearing ability. However, the implants are not absolutely efficient in restoring hearing; their performance varies from patient to patient and requires training to adapt to the device. With advances in regenerative medicine using stem cells and gene therapy, several new strategies have emerged with the hope of permanently curing deafness. Some of these strategies are discussed in the following paragraphs.

The success of current approaches unveils the possibility of using different viral vectors, genes, growth factors and stem cell types in regenerating inner ear hair cells. Nevertheless, more research is still required to overcome challenges involved before these approaches will be ready for clinical and commercial use. Although several studies have reported the generation of hair cell-like cells in vitro and in vivo (in animals), the outcome and characteristic properties of these cells must be further explored. A cell that simply resembles a hair cell morphologically and expresses hair cell markers would be immature without being able to respond to mechanical stimuli and transfer signals to the auditory neurons. A detailed investigation of their ultra structure and ability to connect to auditory neurons and produce transduction currents is essential. In terms of functional outcomes, crucial to any tissue engineering strategy, the mechanoelectrical transduction and maturity of transplanted hair cell-like cells can be evaluated by auditory brainstem response (ABR) and otoacoustic emissions tests (OAE) in animal studies.

Audiologists and scientists have come a long way toward reaching the goal for hair cell regeneration in mammals by developing inner ear gene therapy strategies in animal models. However, an improvement in the existing gene delivery techniques is required to suit clinical applications.

Some of the drawbacks associated with inner ear gene therapy are addressed below. For example, gene delivery studies have shown that over-expression of the Math1 gene led to the formation of ectopic hair cells [16,50]. Hair cells are extremely rare and the cochlea contains only about 14,000 hair cells that detect and amplify sound [103], and it is important to regulate the cell cycle for normal development of cochlear function. Studies in p27kip knockout mice resulted in production of supernumerary hair cells and later a massive degradation of the hair cells occurred, leading to a severe hearing impairment in these animals. This indicates the importance of precise control of cell cycle [69]. There are several pathways that control the expression of Math1 and drive a cell toward differentiating into a hair cell. Math1 inhibitors and down regulators can be used in addition to targeting the inner ear with only the Math1 gene. This approach can be used to produce an appropriate number of hair cells. The activation of a notch receptor can up regulate Hes and Hey genes that are potent inhibitors of Math1 [18,19,57]. Inclusion of notch receptor inhibitors in gene therapy can help in boosting Math1 gene expression. However, a balance of expression and inhibition between the Math1 gene and Math1 inhibitor is required for normal inner ear development and hearing.

Besides developing strategies to control the expression of the Math1 gene, the side effects caused by viral routes also need to be considered. Although viral methods have shown tremendous success in the animal model, there is always the risk of DNA mutations and cancer associated with viruses. Non-viral gene delivery techniques have recently gained attention due to the minimal risks associated in terms of clinical safety and reliability. Another drawback of Math1 gene therapy is that although it can be a potential cure for hearing loss caused by sound trauma or ototoxic damage, it may not cure hearing loss caused by genetic defects. Stem cell-based therapies in vivo have reported migration of these cells into different areas, and thus an ideal route for delivering cells to the right location in the cochlea is crucial to restoring hearing loss. The large majority of studies have used only animal models and animal stem cells. There is no doubt that these studies have played a major role in gathering valuable information about hair cell regeneration; however, the variation among species requires the need to explore the use of human stem cell lines like bone marrow derived mesenchymal and hematopoietic stem cells, umbilical cord blood and umbilical mesenchymal stromal stem cells, and induced pluripotent stem cells. Research in human stem cells lines will have more clinical relevance.

Gene delivery and stem cells are a potential cure to hearing loss; however, there are limitations that must be overcome. Gene therapies must consider using combinations of essential genes and cell cycle inhibitors to control the production of hair cells, and detailed quantification of hair cell functionality is necessary. Gene delivery vehicles with minimal risk of mutagenesis and immune response must also be developed. Finally, an ideal route to implant stem cells inside the cochlea would be essential for successful innervations of hair cells, thereby allowing better, if not pristine transmission of sound with the assistance of a cochlear implant.