Auditory hair cell replacement and hearing improvement by Atoh1 gene therapy in deaf mammals

Article (PDF Available)inNature Medicine 11(3):271-6 · April 2005with75 Reads
DOI: 10.1038/nm1193 · Source: PubMed
In the mammalian auditory system, sensory cell loss resulting from aging, ototoxic drugs, infections, overstimulation and other causes is irreversible and leads to permanent sensorineural hearing loss. To restore hearing, it is necessary to generate new functional hair cells. One potential way to regenerate hair cells is to induce a phenotypic transdifferentiation of nonsensory cells that remain in the deaf cochlea. Here we report that Atoh1, a gene also known as Math1 encoding a basic helix-loop-helix transcription factor and key regulator of hair cell development, induces regeneration of hair cells and substantially improves hearing thresholds in the mature deaf inner ear after delivery to nonsensory cells through adenovectors. This is the first demonstration of cellular and functional repair in the organ of Corti of a mature deaf mammal. The data suggest a new therapeutic approach based on expressing crucial developmental genes for cellular and functional restoration in the damaged auditory epithelium and other sensory systems.



Full-text (PDF)

Available from: Douglas E Brough
MARCH 2005 271
Auditory hair cell replacement and hearing improvement
by Atoh1 gene therapy in deaf mammals
Masahiko Izumikawa
, Ryosei Minoda
, Kohei Kawamoto
, Karen A Abrashkin
, Donald L Swiderski
David F Dolan
, Douglas E Brough
& Yehoash Raphael
In the mammalian auditory system, sensory cell loss resulting from aging, ototoxic drugs, infections, overstimulation and
other causes is irreversible and leads to permanent sensorineural hearing loss. To restore hearing, it is necessary to generate
new functional hair cells. One potential way to regenerate hair cells is to induce a phenotypic transdifferentiation of
nonsensory cells that remain in the deaf cochlea. Here we report that Atoh1, a gene also known as Math1 encoding a basic
helix-loop-helix transcription factor and key regulator of hair cell development, induces regeneration of hair cells and
substantially improves hearing thresholds in the mature deaf inner ear after delivery to nonsensory cells through adenovectors.
This is the first demonstration of cellular and functional repair in the organ of Corti of a mature deaf mammal. The data
suggest a new therapeutic approach based on expressing crucial developmental genes for cellular and functional restoration
in the damaged auditory epithelium and other sensory systems.
The most common reason for sensorineural hearing loss is degeneration
of cochlear sensory (hair) cells, resulting from overstimulation, ototoxic
drugs, infections, autoimmune disease or aging. Sensorineural hearing
loss affects millions of people worldwide. This impairment is irrevers-
ible because lost auditory hair cells cannot be spontaneously replaced.
Treatment for replacing lost hair cells is currently unavailable.
During mammalian embryogenesis, cochlear hair cells and supporting
cells have common cellular precursors
. Precursor cell differentiation is
regulated by several genes; key among them is the gene Atoh1 (also known
as Math1), a mouse homolog of the Drosophila gene atonal, that encodes
a basic helix-loop-helix transcription factor. Atoh1 has been shown to
act as a ‘pro-hair cell gene’ and is required for the differentiation of hair
cells from multipotent progenitors
. Experimental overexpression of
Atoh1 or ATOH1 (the human atonal gene) in nonsensory cells of the
normal cochlea generates new hair cells, both in vitro
and in vivo
. Here
we test the influence of Atoh1 overexpression on hair cell regeneration
and hearing restoration in the mature deaf cochlea.
Young adult guinea pigs with normal hearing were deafened by
systemic administration of ototoxic drugs, resulting in complete bila-
teral hair cell loss in the high- and mid-frequency regions of the cochlea.
In animals killed 3 d after the deafening procedure, scanning electron
microscopy (SEM) analysis of the organ of Corti surface showed com-
plete absence of hair cells (Fig. 1a). At this time point, whole mounts
of the organ of Corti stained with phalloidin showed a complete lack
of hair cells at the upper (luminal) surface of the epithelium (Fig. 1b).
Hair cell features such as stereocilia and cuticular plate were absent,
whereas pillar cells remained in the tissue (Fig. 1a,b). In normal cochleae
stained with bisbenzimide (Hoechst) and analyzed with fluorescence
microscopy, outer hair cell (OHC) nuclei were arranged in three distinct
rows (Fig. 1c) with supporting cell nuclei similarly arranged, at a lower
focal plane (data not shown). OHC nuclei were absent 3 d after deafe-
ning (Fig. 1d); however, supporting cell nuclei were present at the lower
level (data not shown), showing that the original OHCs were eliminated
by this deafening protocol. Cochlear whole mounts stained 3 d after
the ototoxic insult with antibodies to myosin VIIa, a hair cell–speci-
fic marker
, showed a complete lack of myosin VIIa (data not shown).
Auditory brain-stem response (ABR) thresholds in all deafened ani-
mals were either unmeasurable or extremely high (>95 dB), indicative
of profound deafness. The battery of post-deafening analyses we have
used showed absence of stereocilia bundles and apical surfaces of hair
cells with SEM, absence of actin-rich hair cell features with phalloi-
din staining, lack of myosin VIIa staining with immunocytochemistry
and lack of nuclei at the OHC plane. These analyses were performed
in cochlear turns one through three, and showed that these three turns
were consistently devoid of inner hair cells (IHCs) and OHCs by day 3
after the ototoxic insult.
To deliver the Atoh1 transgene into nonsensory cells in the deafe-
ned auditory epithelium, an adenoviral vector was infused into the
left cochlea 4 d after the ototoxic lesion. Animals received adenovi-
rus alone (Ad.empty), with a GFP cassette (Ad.GFP), with Atoh1
(Ad.Atoh1) or with a dual cassette (Ad.Atoh1-GFP). To test the pat-
tern of Atoh1 expression, animals were killed 4 d after Ad.Atoh1 ino-
culation and processed for Atoh1 immunocytochemistry. We found
numerous Atoh1-positive cells among the nonsensory cells that
Kresge Hearing Research Institute, Department of Otolaryngology, University of Michigan, MSRB 3, Room 9303, 1150 W. Medical Center Drive, Ann Arbor,
Michigan 48109-0648, USA.
Department of Otolaryngology, Kansai Medical University, 10-15 Fumizono-Cho, Moriguchi, Osaka 570–8506, Japan.
of Otolaryngology–Head and Neck Surgery, Kumamoto University School of Medicine, 1-1-1 Honjo, Kumamoto 860–8556, Japan.
Department of Vector Sciences,
GenVec Inc., 65 W. Watkins Mill Road, Gaithersburg, Maryland 20879-4021, USA. Correspondence should be addressed to Y.R. (
Published online 13 February 2005; doi:10.1038/nm1193
© 2005 Nature Publishing Group
272 VOLUME 11
remained in the organ of Corti (Fig. 1e,f). We observed additional
Atoh1-positive nuclei in other focal planes (data not shown). We
observed the highest efficiency of Atoh1 transduction in the first and
second cochlear turns, within the normal boundaries of the organ of
Corti. We determined that in 2 mm flanking the inoculation site, an
average of 865 Atoh1-positive cells were found in the organ of Corti
4 d after Ad.Atoh1 inoculation (n = 10 ears, s.d. = 196.6). Right (con-
tralateral) ears and left ears inoculated with Ad.empty or Ad.GFP were
Atoh1-negative (data not shown). Deafened animals killed 4 d after ino-
culation with Ad.Atoh1-GFP showed widespread expression of both genes
(Fig. 1g). These data show efficient transduction and Atoh1 transgene
expression in nonsensory cells of the deaf cochlea and absence of Atoh1
expression in control ears.
To determine the effects of Atoh1 overexpression in deafened ears,
ABR thresholds were recorded and animals killed 4, 5 or 8 weeks after
inoculation. We prepared cochleae for SEM or whole-mount fluores-
cence microscopy. Ad.Atoh1-treated cochleae showed large numbers
of IHCs and OHCs 8 weeks after inoculation (Fig. 2ac). All right ears
(Fig. 2d) and all left ears treated with Ad.empty (Fig. 2e) or Ad.GFP
(not shown) were devoid of hair cells. Higher-magnification SEM
analysis showed that the surface morphology of the hair cells in
Atoh1-treated ears was relatively normal (Fig. 2f,g) but the supporting
cells that separated one hair cell from its neighbor were narrow and
not well defined.
Hair cells detected in Ad.Atoh1-treated deafened cochleae showed
normal surface morphology and orientation in the organ of Corti
(Fig. 2a,f,g), suggesting that the positional cues for cellular organization
remain in the mature traumatized tissue. Organization of third-row
OHCs was poorer than other rows, suggesting that cues for organizing
the tissue are closer to the first row of OHCs. Ectopic hair cells found
outside the organ of Corti (Fig. 2a) were neither well differentiated nor
correctly oriented. Based on their location in the organ of Corti and
their highly polarized morphology, we speculate that pillar cells provide
cues for reorganizing the auditory epithelium.
Hair cells in Atoh1-treated ears expressed myosin VIIa, providing
further evidence for hair-cell phenotype (Fig. 3a). Ad.empty- inoculated
ears (data not shown) and contralateral ears (Fig. 3b) were negative for
myosin VIIa, confirming that all original hair cells have been eliminated.
In the normal auditory epithelium there is a 1:1 ratio of OHCs to
Deiters cells (supporting cells), with OHC nuclei positioned above
the supporting-cell nuclei. In deafened ears, OHC nuclei were absent
(Fig. 1d) and the total number of nuclei in this area was half that of
normal ears. Two months after Ad.Atoh1 inoculation, nuclei reappeared
at a plane above the supporting cells (Fig. 3c). An average number
of 1,548 nuclei was observed in the same area of Atoh1-treated ears,
compared to an average of 830 nuclei in contralateral ears (n = 5
each, s.d. = 209 and 51, respectively), a statistically significant increase
(P =
0.0006). The increase in nuclei number probably is not a direct
result of Atoh1 expression, because Atoh1 acts as a differentiation factor.
Therefore, we hypothesize that new hair cells recruit nonsensory cells
from outside the organ of Corti to migrate and become supporting cells,
as described in vitro
. Supporting cells may also induce proliferation of
other nonsensory cells as part of a self-organizational capacity of the
regenerating cochlea
. The Atoh1-induced transdifferentiation from a
nonsensory to sensory phenotype can be either direct, with no mitosis, or
indirect, with mitosis preceding the appearance of the hair cell features.
The numbers and appearance of hair cells were similar in Ad.Atoh1
and Ad.Atoh1-GFP inoculated ears. Some Ad.Atoh1-GFP– inoculated left
ears contained hair cells positive for GFP, establishing a direct relationship
between transgene expression in surviving nonsensory cells and transdif-
ferentiation into hair cells (Fig. 3d). Although adenoviral-mediated gene
expression is not expected to last for 2 months, the GFP gene product
probably persisted in transduced cells because of slow degradation.
ef g
Figure 1 Hair cell elimination and Atoh1 expression in deaf cochleae.
SEM (a) or epifluorescence (bg) showing ototoxic lesions with complete
hair cell loss (a,b and d), nuclei in the organ of Corti (c,d) and Atoh1
expression after Ad.Atoh1 inoculation (eg) in the second cochlear turn.
(a) Complete hair-cell loss was observed 3 d after systemic (bilateral)
deafening, with nonsensory cells replacing degenerated hair cells in the
IHC (I) and OHC (O) region. Pillar cells (P) survive the insult. (b) Distribution
of F-actin shows lack of hair cells 3 d after deafening. Ovals encircle
representative sites of missing IHCs and O points to cells missing in OHC
area. Pillar cells (P) survive the insult. (c) In the normal organ of Corti,
Hoechst stain depicts nuclei of the three rows of OHCs. (d) All OHC nuclei
are missing 3 d after an ototoxic insult. (e) Antibody to Atoh1 is seen in
nuclei of nonsensory cells that replaced hair cells (first turn) 4 d after
Ad.Atoh1 inoculation. (f) Atoh1 (red) in nonsensory cells in second cochlear
turn, with phalloidin (green) showing actin in cell-cell junctions. (g) Atoh1
(red, nuclear stain) and GFP (green, cytoplasmic stain) are coexpressed
in nonsensory cells (Deiters and pillar cells) 4 d after Ad.Atoh1-GFP
inoculation. Additional positive nuclei are present out of the focal plane
shown in eg. Scale bars, 25 µm in e and 10 µm in all other micrographs.
© 2005 Nature Publishing Group
MARCH 2005 273
Phalloidin epifluorescence facilitated identification of hair cells based
on the presence of stereocilia (data not shown) and the distribution of
actin in the cell border areas and the cuticular plate, and the absence of
actin in the centrosomal region (Fig. 3d). Quantitative analysis of these
phalloidin-stained cochleae 2 months after Atoh1 treatment showed the
number of new hair cells in the 2 mm of the organ of Corti flanking
the inoculation site to be 256 IHCs and 691 OHCs (n = 5 animals, s.d.
= 35 and 159 for IHCs and OHCs, respectively). The number of hair
cells in the Atoh1-treated ears was significantly greater than contrala-
teral ears, which were devoid of hair cells (P < 0.0006). The restoration
of hair cells in Atoh1-treated ears was best near the site of inoculation
(Figs. 2a and 3a,d). These data show that many new hair cells appear in
the 8 weeks after Atoh1 expression in deafened mature cochleae.
Cross-sections of Atoh1-treated cochleae showed that some cells in
the OHC area displayed a mixed phenotype with features of both OHC
and supporting cell. These mixed-phenotype cells showed a luminal
projection similar to stereocilia on their apical surface and a promi-
nent cuticular plate, yet spanned the distance from the luminal surface
to the basement membrane, suggesting that they also retained several
supporting-cell features (Fig. 3e). In some sections, two layers of nuclei
were found above the basement membrane (Fig. 3f), in agreement with
data obtained with Hoechst staining in whole mounts (Fig. 3c). IHC
morphology appeared relatively normal (Fig. 3g). The presence of cells
with a mixed phenotype suggests that, at least in some cases, the process
of transdifferentiation can occur directly, without a preceding step of
de-differentiation or mitosis. The increase in the number of nuclei com-
pared to the contralateral ears may be a result of mitosis or of migration
from adjacent regions in the epithelium.
We analyzed Ad.Atoh1-treated cochleae at intermediate time points
between deafening and 2 months. Cochleae analyzed with SEM
4 weeks after Ad.Atoh1 inoculation showed numerous bundles of very
tall stereocilia that resembled the apical surfaces of immature hair cells
(Fig. 4a). The high density of immature bundles observed at 4 weeks
corroborates the high efficiency of Atoh1 transduction shown at day
4. In the contralateral cochleae there were no surviving hair cells and
no immature bundles, and the area where OHCs used to reside often
appeared narrow and greatly diminished (Fig. 4b). The pattern of
myosin VIIa distribution in Atoh1-treated cochlea analyzed 5 weeks
after the inoculation of Ad.Atoh1 delineated both IHCs and OHCs
(Fig. 4c). The myosin VIIa–positive cells in the OHC area were rather
disorganized, in agreement with the SEM image obtained 1 week earlier
(Fig. 4a). Contralateral ears were negative for myosin VIIa (Fig. 4d).
To test the influence of Atoh1 treatment on hearing, we measured
ABR thresholds at 4, 8 and 10 weeks after Ad.Atoh1 inoculation.
Measurements at 4 weeks indicated profound deafness with thresholds
high or unmeasurable, similar to post-deafening thresholds (not
shown). These data are consistent with the absence of mature hair cells
at 4 weeks after Atoh1 treatment (Fig. 4a). At 8 weeks, the average ABR
thresholds in a group of five Ad.Atoh1-treated ears was lower (better)
than contralateral ears at all frequencies, and ABR waveforms contained
the normal four to five vertex-positive peaks (Fig. 5). The considerable
improvement in thresholds (in some cases to near baseline) at the high
frequency region of the guinea pig cochlea (4–24 kHz) is in agreement
with the restoration of IHC morphology in this area
. Nevertheless,
in the absence of a normal OHC population, frequency selectivity is
unlikely to be normal
. When present, the thresholds recorded in the
contralateral (right) ears probably represent hearing from Atoh1-trea-
ted left ears and reflect interaural crossover
. ABR thresholds measured
in four animals at 10 weeks after Ad.Atoh1 inoculation appeared similar
to their 8-week thresholds (data not shown) suggesting that functional
recovery is stable, at least up to 10 weeks. In Ad.empty-treated ears, no
thresholds could be recorded (data not shown). These data provide the
first demonstration of a therapeutic approach leading to substantial
recovery of hearing in deaf mammalian ears.
Our results show generation of new hair cells and improvement of
hearing in deaf animals treated with Atoh1. New OHCs are incom-
pletely differentiated and unlikely to provide the functions of the
active cochlear amplifier. In contrast, regenerated IHCs seem nor-
mal and the restoration of thresholds attests to their functionality.
We show that differentiated cells (nonsensory cells of the auditory
epithelium) can be induced to alter their phenotype by expression of
the developmental gene, Atoh1. Together with loss-of-function and
in vitro experiments
, our data suggest that Atoh1 is a master
regulatory gene that is both necessary and sufficient for producing
hair cells in the mammalian cochlea. The competence of mature and
differentiated nonsensory cells to respond to Atoh1 is notable because
the structural and functional changes that accompany differentiation
of mammalian cells are usually irreversible. Our data, therefore, sug-
ef g
Figure 2 Hair cells reappear in deaf ears treated with Ad.Atoh1. SEM view
of deafened cochleae 2 months after Atoh1 inoculation (ac), contralateral
cochlea (d), Ad.empty-inoculated ear (e), and higher magnification of IHC (f)
and OHC (g) 2 months after Atoh1 inoculation. (a) The site of inoculation in
the second cochlear turn (asterisk) is shown along with numerous stereocilia
bundles at the normal sites of IHCs (I) and OHCs (rows 1–3). Pillar cells
(P) are present between IHCs and OHCs. Ectopic bundles (arrowheads) are
seen lateral to the third row of OHCs. (b) In some Atoh1-treated ears the
morphology of IHC (I) and OHCs (O) is less well differentiated. (c) In other
Atoh1-treated ears, hair cell reappearance is incomplete and third row OHCs
are missing. (d,e) second cochlear turns of right (d, contralateral to a) or
Ad.empty-inoculated left cochlea (e), showing complete absence of hair
cells. (f,g) Stereocilia bundle organization is relatively normal in IHCs (f) and
OHCs (g) but supporting cells between neighboring hair cells are narrow and
not well defined. Scale bars, 25 µm in a; 50 µm in b, c and e; 10 µm in d
and 5 µm in f and g.
© 2005 Nature Publishing Group
274 VOLUME 11
gest that re-expression of developmental regulatory genes in mature
tissues is a potential strategy for cell replacement therapy in the cochlea
and elsewhere.
The increase in number of nuclei in the sensory epithelium after Atoh1
treatment can be explained either by cell migration from areas flanking
the organ of Corti, as seen in birds
, or by a proliferative response of
nonsensory cells that occurs secondary to the generation of new hair
cells. Atoh1 by itself is unlikely to induce mitosis
. The cochlea may also
have a population of unidentified stem cells that divide after deafening,
a possibility suggested by recent findings in the utricle
. It is unknown
whether specific subpopulations of supporting cells are capable of
(i) generating new hair cells following Atoh1 expression (ii) migrating
into the organ of Corti and/or (iii) dividing and repopulating the
The Atoh1-induced restoration of the luminal surface of the
auditory epithelium generates both hair cell and supporting cell-like
areas which, in some cases, resemble the normal cellular mosaic in
the organ of Corti. We speculate that dual-phenotype cells contribute
in part to the generation of areas with surface features of supporting
cells. Newly added cells that derive from proliferation or migration
probably also contribute to the restoration of the surface morphology.
a b
Figure 4 Cochleae analyzed with SEM at 4 weeks after Atoh1 treatment
(a) or myosin VIIa immunocytochemistry at 5 weeks after treatment (c) and
the respective contralateral ears (b and d). (a) Several immature stereocilia
bundles (some colored to highlight the morphology) are observed lateral to
the pillar cells (P). The OHC region is wide and the tall surface projections
cover much of the surface. (b) In the contralateral cochlea the distance
between pillar cells (P) and Hensen cells (H) where OHCs had resided before
deafening (O, vertical double-arrow) is greatly diminished. Neither OHCs nor
IHCs (I and horizontal double-arrow) are present. (c) In a Atoh1-treated ear,
a well-organized single row of IHCs and relatively disorganized rows of
OHCs are seen with cytoplasmic staining of myosin VIIa in the second turn.
(d) Myosin VIIa–positive cells are absent in contralateral ear. Scale bars,
5 µm in a and b and 10 µm in c and d.
Figure 3 New hair cells and nuclei in the deafened cochleae inoculated with Ad.Atoh1. Atoh1-treated ears (a and cg) and contralateral cochlea
(b) processed 2 months later for myosin VIIa immunocytochemistry (a,b), Hoechst stain (c), epifluorescence for actin and GFP (d) or plastic sections (eg).
(a) Pillar cells (P) are flanked by IHCs (I) and OHCs (rows 1–3) positive for myosin VIIa in the second cochlear turn. (b) Lack of myosin VIIa–positive hair
cells in the contralateral (right) ear. (c) Nuclei are present in rows at the supporting cell focal plane (S), and above, closer to the OHC focal plane (rows
1–3). (d) Two months after inoculation with Ad.Atoh1-GFP, actin (red), is seen in junctional complexes delineating the contour of OHCs (dotted circles) and
in cuticular plates (asterisks), but not in the area of the centrosomes which is actin-free (arrowheads). GFP (green, cytoplasmic) is seen in all three rows of
OHCs (rows 1–3). First row OHCs are shown in focal plane beneath apical junctions. Nuclei visible in this focal plane are unstained for GFP (arrows).
(e) A cell in the OHC area of the second turn has a cuticular plate (white arrowhead) and a lumenal projection (black arrow) protruding from the cuticular
plate. The cell body extends from the lumenal surface (black arrowhead) to the basement membrane (white arrow), and the nucleus is basal. (f) In some
regions of OHC area a dual layer of nuclei can be seen above the basement membrane (arrow). (g) An IHC featuring a luminal projection (arrow) and normal
morphology. Scale bars, 10 µm in a, b, and dg, and 5 µm in c.
© 2005 Nature Publishing Group
MARCH 2005 275
The organization of the tissue is likely to depend on the ability of hair
cells to dictate the phenotype of their neighbors and to attract sup-
porting cells, as shown in vitro
. The ability of the tissue to respond to
Atoh1 overexpression and to reorganize the new hair cells generated
by Atoh1 may depend on the mechanism of deafening, the timing
of treatment after deafening and the degree of differentiation of the
surviving nonsensory cells. Better understanding of the response of
supporting cells to Atoh1 expression will also shed light on the pro-
cess of post-trauma repair of the mosaic-like cytoarchitecture at the
luminal surface of the auditory epithelium.
A clinical restoration of threshold sensitivity similar to that shown
here in guinea pigs would be an extremely attractive therapeutic
outcome for patients with profound deafness. Addition of new hair
cells may also enhance the outcome of cochlear implantation, because
presence of some hair cells improves implant function
. Our findings
support the feasibility of genetic manipulation for cell replacement
therapies based on inducing transdifferentiation of endogenous cells
in the inner ear and in other systems.
Animals. All animal experiments were approved by the University of
Michigan Institutional Committee on the Use and Care of Animals and per-
formed using accepted veterinary standards. We used young adult guinea
pigs (Elm Hill Breeding Laboratory) weighing 250–400 g with normal ABR
baseline thresholds of 15–50 dB sound pressure level (SPL). Animal groups
were Ad.Atoh1 (n = 30), Ad.empty (n = 12), Ad.Atoh1-GFP (n = 5), Ad.GFP
(n = 5) and deafening alone (n = 4). Inner ears were analyzed using epifluores-
cence, SEM or plastic cross-sections of the organ of Corti. Animals that did not
have a complete loss of hair cells in the three lower cochlear turns, as determined
by observing the right (contralateral) ear, were excluded from the study.
Deafening surgery and viral inoculation. Guinea pigs were
deafened bilaterally with a single systemic dose of kanamycin
(500 mg/kg, subcutaneous) followed 2 h later by ethacrynic acid
(50 mg/kg, intravenous). We verified deafening by measuring ABR thresholds.
This ototoxic drug regimen is designed to consistently eliminate hair cells
but spare the pillar cells. Hair-cell elimination is complete in the first three
cochlear turns, although a few hair cells may survive in the apical turn. We
excluded animals that did not show a threshold greater than 95 dB SPL
(numbers of animals given above do not include excluded animals). Viral
vectors were inoculated into the left cochlea 4 d after the deafening surgery,
using a procedure previously described
, except that the inoculation was into
the second cochlear turn.
Adenovirus vector. Replication-deficient recombinant adenoviruses with deleted
E1, E3, and E4 regions
were Ad.Atoh1, Ad.Atoh1-GFP, Ad.empty and Ad.GFP.
The Atoh1 cDNA was obtained from H. Zoghbi (Baylor College of Medicine).
The insert in Ad.Atoh1 was driven by the human cytomegalovirus promoter and
the GFP by the chicken beta-actin promoter. We used undiluted vectors at a con-
centration of 1 × 10
total particles purified virus per milliliter. Viral suspensions
were kept at –80 °C until thawed for use.
ABR measurement. To assess auditory thresholds, we recorded ABRs at 4, 8,
16, and 24 kHz (tone bursts, 15 ms duration, 1 ms cos
-shaped rise-fall times)
as previously described
. Measured frequencies correspond to the basal and
lower second turn of the guinea pig cochlea, close to the site of inoculation.
When present, the thresholds recorded in the contralateral (right) ears probably
represent hearing from Atoh1-treated left ears. For this reason, an animal with
lesser threshold recovery in the left ear was selected for showing ABR waveforms
in the contralateral ear in Fig. 5. Lower frequency thresholds are not reliably
measurable with ABR audiometry
. The researcher who measured ABRs was
blinded to the identity of the animal.
Immunocytochemistry and SEM. We stained whole mounts of the auditory
epithelium with antibodies to myosin VIIa or Atoh1 as described
, except that
tissues were double stained with phalloidin (Molecular Probes) to stain actin
. To
augment GFP fluorescence, we stained tissues with a GFP-specific rabbit antibody
(Chemicon, diluted 1:500). Antibody to myosin VIIa was purchased from the
University of California San Diego. Antibody to Atoh1 was purchased from the
Tissue Culture Hybridoma Core at the University of Iowa. To stain nuclei, we used
the DNA-specific label Hoechst as previously described
. The specimens were
examined and photographed using a Leica DMRB epifluorescence microscope
(Leica) with a CCD Cooled SPOT-RT digital camera (Diagnostic Instruments).
For SEM, samples were prepared as described
. The samples were mounted on
stubs using silver paste and photographed digitally using a Philips XL30 Field
Emission Gun SEM (FEI).
Plastic sections. We decalcified cochleae, embedded them in Epon as described
and sectioned (1 µm) them with a glass knife. Sections were photographed using
a Leica DMRB epifluorescence microscope and a 100× objective lens.
Data analysis. We performed the statistical analysis of the ABR data using Excel.
Conventional t-tests were performed to determine the significance of the dif-
ference between treated and nontreated ears at each tested frequency. Because
we performed multiple tests, we applied sequential Bonferroni adjustment
Figure 5 Atoh1 treatment improves ABR thresholds. (a) ABR waves for
Ad.Atoh1-treated ear represented in c by orange circle with asterisk. Waves
show marked peaks in response to input as low as 65 dB sound pressure
level (SPL). The waveform shows that the latency of the marked peak
increases with decreasing level of the stimulus, as expected in normal
ABR response
. (b) In right (contralateral) ear no consistent peaks are
seen. (c) Data points for thresholds at each frequency for the left (Atoh1-
inoculated) and right (contralateral) ears of five animals. ABRs could not be
obtained from the right ear of the orange animal at the limits of the sound
delivery system (105 dB SPL). Average dB SPL thresholds in treated ears
are significantly lower (better) than in contralateral ears at each frequency
(P < 0.004). The number of IHCs per 100 µm is given for each animal.
Thresholds are significantly correlated with the number of IHCs in Atoh1-
treated ears. The weakest correlation was at 4 kHz (r
= 0.78, P < 0.05).
© 2005 Nature Publishing Group
276 VOLUME 11
to keep the table-wide error rate below 0.05. ABR thresholds for each animal
were regressed on the number of IHC to obtain the correlation for each tested
frequency. A t-test was performed on the 4 kHz data to determine whether the
correlation was significant (slope > 0).
Evaluation of the morphology of Atoh1-inoculated ears 8 weeks after the
inoculation showed three animals that were completely devoid of hair cells. These
animals were assumed to have had failed inoculation surgeries and were excluded
from these statistical analyses.
We performed quantitative analysis of whole mounts of the organ of Corti to
determine the number of hair cells (phalloidin-stained cochleae) or positively
stained nuclei (Atoh1 or Hoechst stains). Nuclei counts in Hoechst-stained whole
mounts were performed in the OHC area, where nuclei of hair and supporting
cells can be distinguished from each other based on their organization in distinct
focal planes. We counted a segment of 1 mm on each side of the inoculation
site for a total of 2 mm. Conventional t-tests were performed to determine the
significance of differences between treated and nontreated cochleae.
We thank L.A. Beyer and L.L. Kabara for technical assistance. We thank
P.F. Hitchcock and D.M. Martin for discussions and comments on the
manuscript. This work was supported by a gift from B. and A. Hirschfield, by
GenVec and by US National Institutes of Health–National Institute on Deafness
and Other Communication Disorders grants R01 DC01634, DC05401 and
P30 DC05188. We thank P.J. Olynyk for graphics work on the cover illustration.
The authors declare competing financial interests (see the Nature Medicine website
for details).
Received 10 September 2004; accepted 25 January 2005
Published online at
1. Fekete, D.M. Cell fate specification in the inner ear. Current Opin. Neurobiol. 6, 533–
541 (1996).
2. Fekete, D.M. Making sense of making hair cells. Trends Neurosci. 23, 386 (2000).
3. Torres, M. & Giraldez, F. The development of the vertebrate inner ear. Mech. Dev. 71,
5–21 (1998).
4. Bermingham, N.A. et al. Math1: an essential gene for the generation of inner ear hair
cells. Science 284, 1837–1841 (1999).
5. Chen, P., Johnson, J.E., Zoghbi, H.Y. & Segil, N. The role of Math1 in inner ear devel-
opment: Uncoupling the establishment of the sensory primordium from hair cell fate
determination. Development 129, 2495–2505 (2002).
6. Shou, J., Zheng, J.L. & Gao, W.Q. Robust generation of new hair cells in the mature
mammalian inner ear by adenoviral expression of Hath1. Mol. Cell Neurosci. 23,
169–179 (2003).
7. Kawamoto, K., Ishimoto, S., Minoda, R., Brough, D.E. & Raphael, Y. Math1 gene
transfer generates new cochlear hair cells in mature guinea pigs in vivo. J. Neurosci.
23, 4395–4400 (2003).
8. Hasson, T., Heintzelman, M.B., Santos-Sacchi, J., Corey, D.P. & Mooseker, M.S.
Expression in cochlea and retina of myosin VIIa, the gene product defective in Usher
syndrome type 1B. Proc. Nat. Acad. Sci. USA 92, 9815–9819 (1995).
9. Woods, C., Montcouquiol, M. & Kelley, M.W. Math1 regulates development of the sen-
sory epithelium in the mammalian cochlea. Nat. Neurosci. 7, 1310–1318 (2004).
10. Greenwood, D.D. Critical bandwidth and the frequency coordinates of the basilar mem-
brane. J. Acoust. Soc. Am. 33, 1344–1356 (1961).
11. Dallos, P. & Harris, D. Properties of auditory nerve responses in absence of outer hair
cells. J. Neurophysiol. 41, 365–383 (1978).
12. Teas, D.C. & Nielsen, D.W. Interaural attenuation versus frequency for guinea pig and
chinchilla CM response. J. Acoust. Soc. Am. 58, 1066–1072 (1975).
13. Zheng, J.L. & Gao, W.Q. Overexpression of Math1 induces robust production of extra
hair cells in postnatal rat inner ears. Nat. Neurosci. 3, 580–586 (2000).
14. Cotanche, D.A., Messana, E.P. & Ofsie, M.S. Migration of hyaline cells into the chick
basilar papilla during severe noise damage. Hear. Res. 91, 148–159 (1995).
15. Li, H., Liu, H. & Heller, S. Pluripotent stem cells from the adult mouse inner ear.
Nat. Med. 9, 1293–1299 (2003).
16. Turner, C.W., Gantz, B.J., Vidal, C., Behrens, A. & Henry, B.A. Speech recognition in
noise for cochlear implant listeners: benefits of residual acoustic hearing. J. Acoust.
Soc. Am. 115, 1729–1735 (2004).
17. Ishimoto, S., Kawamoto, K., Kanzaki, S. & Raphael, Y. Gene transfer into supporting
cells of the organ of Corti.
Hear. Res. 173, 187–197 (2002).
18. Brough, D.E. et al. Activation of transgene expression by early region 4 is responsible
for a high level of persistent transgene expression from adenovirus vectors in vivo.
J. Virol. 71, 9206–9213. (1997).
19. Yamasoba, T. & Dolan, D.F. Chronic strychnine administration into the cochlea poten-
tiates permanent threshold shift following noise exposure. Hear. Res. 112, 13–20
20. Davis, H. Principles of electric response audiometry. Ann. Otol. Rhinol. Laryngol. 85
Suppl 28, 1–96 (1976).
21. Raphael, Y. & Altschuler, R.A. Reorganization of cytoskeletal and junctional pro-
teins during cochlear hair cell degeneration. Cell Motil. Cytoskeleton 18, 215–227
22. Raphael, Y. Reorganization of the chick basilar papilla after acoustic trauma. J. Comp.
Neurol. 330, 521–532 (1993).
23. Beyer, L.A. et al. Hair cells in the inner ear of the pirouette and shaker 2 mutant mice.
J. Neurocytol. 29, 227–240 (2000).
24. Yagi, M. et al. Spiral ganglion neurons are protected from degeneration by GDNF gene
therapy. J. Assoc. Res. Otolaryngol. 1, 315–325. (2000).
© 2005 Nature Publishing Group
    • "By indicating which TFs are present in IHCs and in OHCs we provide a starting point for future studies into the activity of individual TFs and how groups of TFs mediate HC phenotype and regulate differentiation, cell cycle control and survival. Understanding HC fate determination, cell cycle regulation and long-term maintenance are essential for developing strategies for HC repair, regeneration and maintenance , all of which are critical for maintenance of HCs as well as restoring lost hearing using gene therapy [58,59]. "
    [Show abstract] [Hide abstract] ABSTRACT: Regulation of gene expression is essential to determining the functional complexity and morphological diversity seen among different cells. Transcriptional regulation is a crucial step in gene expression regulation because the genetic information is directly read from DNA by sequence-specific transcription factors (TFs). Although several mouse TF databases created from genome sequences and transcriptomes are available, a cell type-specific TF database from any normal cell populations is still lacking. We identify cell type-specific TF genes expressed in cochlear inner hair cells (IHCs) and outer hair cells (OHCs) using hair cell-specific transcriptomes from adult mice. IHCs and OHCs are the two types of sensory receptor cells in the mammalian cochlea. We show that 1,563 and 1,616 TF genes are respectively expressed in IHCs and OHCs among 2,230 putative mouse TF genes. While 1,536 are commonly expressed in both populations, 73 genes are differentially expressed (with at least a twofold difference) in IHCs and 13 are differentially expressed in OHCs. Our datasets represent the first cell type-specific TF databases for two populations of sensory receptor cells and are key informational resources for understanding the molecular mechanism underlying the biological properties and phenotypical differences of these cells.
    Full-text · Article · Mar 2016
    • "gene acts as a " switch " to turn on hair cell growth and it is discovered that Atoh1 is artificially switched on in the cells that support hair cells (called " supporting cells " ); it instructs them to divide and form new hair cells. Atoh1 (Math1) plays an important role in the differentiation of hair cells of the developing inner ear and restore auditory function878889. Using the tools of gene therapy may activate Atoh1 to induce undamaged cells within the cochlea to develop into hair cells in an adult human ear and rebuild a damaged ear by replicating the steps that took place during embryonic development. There is still a lot of work to be done for human adult ear. "
    [Show abstract] [Hide abstract] ABSTRACT: Despite the significant advances in understanding the molecular basis of hearing loss, precise identification of genetic cause still presents some difficulties, owing to phenotypical variation. Gene discovery efforts for hearing disorders are complicated by extreme heterogeneity. Mutations in some of these genes, such as GJB2, MYO7A, CDH23, OTOF, SLC26A4, TMC1, are quite common and responsible for hearing loss. Clinical exome sequencing is a highly complex molecular test that analyzes the exons or coding regions of thousands of genes simultaneously, using next-generation sequencing techniques. The development of a biological method for the repair, regeneration, and replacement of hair cells of the damaged cochlea has the potential to restore normal hearing. At present, gene therapy and stem cells are two promising therapeutic applications for hearing disorders. Gene therapy and stem cell treatment have still a long way to go before these treatments will be available to use in humans. Therefore, existing measures must focus on the prevention of hearing loss to decrease the frequency of genetic hearing loss. Over time, genetic diagnostic tests will become available most rapidly, followed by targeted gene therapy or various permutations of progenitor cell transplantation, and eventually, the preventive interventions for a wider range of hearing impaired patients.
    Full-text · Chapter · Dec 2015 · The Journal of Comparative Neurology
    • "The spatiotemporal pattern of eya1 expression agrees with cell proliferation, which suggests eya1 may have a role in progenitor proliferation during ampullary organ regeneration . Along with SIX1, EYA1 also can induce hair cell differentiation through activating the preneural gene atoh1 (Ahmed et al., 2012), a key gene for hair cell development and regeneration in the mammalian inner ear (Izumikawa et al., 2005; Atkinson et al., 2014). We have tried to clone atoh1 in Siberian sturgeon using a homology cloning strategy and embryonic transcriptome sequencing, but failed. "
    [Show abstract] [Hide abstract] ABSTRACT: The lateral line found in some amphibians and fishes has two distinctive classes of sensory organs: mechanoreceptors (neuromasts) and electroreceptors (ampullary organs). Hair cells in neuromasts can be damaged by aminoglycoside antibiotics and they will regenerate rapidly afterward. Aminoglycoside sensitivity and the capacity for regeneration have not been investigated in ampullary organs. We treated Siberian sturgeon (Acipenser baerii) larvae with neomycin and observed loss and regeneration of sensory hair cells in both organs by labeling with DASPEI and scanning electron microscopy (SEM). The numbers of sensory hair cells in both organs were reduced to the lowest levels at 6 hours post-treatment (hpt). New sensory hair cells began to appear at 12 hpt and were regenerated completely in 7 days. To reveal the possible mechanism for ampullary hair cell regeneration, we analyzed cell proliferation and the expression of neural placodal gene eya1 during regeneration. Both cell proliferation and eya1 expression were concentrated in peripheral mantle cells and both increased to the highest level at 12 hpt, which is consistent with the time course for regeneration of the ampullary hair cells. Furthermore, we used Texas Red-conjugated gentamicin in an uptake assay following pre-treatment with a cation channel blocker (amiloride) and found that entry of the antibiotic was suppressed in both organs. Together, our results indicate that ampullary hair cells in Siberian sturgeon larvae can be damaged by neomycin exposure and they can regenerate rapidly. We suggest that the mechanisms for aminoglycoside uptake and hair cell regeneration are conserved for mechanoreceptors and electroreceptors. This article is protected by copyright. All rights reserved.
    Full-text · Article · Oct 2015
Show more

Recommended publications

Discover more