One of the greatest challenges in the treatment of
inner-ear disorders is to find a cure for the hearing loss
that is caused by the loss of cochlear hair cells or spiral
ganglion neurons. The recent discovery of stem cells in
the adult inner ear that are capable of differentiating
into hair cells, as well as the finding that embryonic
stem cells can be converted into hair cells, raise hope
for the future development of stem-cell-based treatment
regimens. Here, we propose different approaches
for using stem cells to regenerate the damaged inner
ear and we describe the potential obstacles that translational
approaches must overcome for the development
of stem-cell-based cell-replacement therapies for
the damaged inner ear.
The sense of hearing is integral to our interaction with our
environment, yet the cellular components of the inner ear
are vulnerable to a variety of damaging agents. All hearing
sensation is derived from the output of a remarkably small
number of sensory cells: fewer than 15 000 per inner ear.
These hair cells are the mechanoelectrical transducers of
the ear; deflections of the stereociliary bundles (the hairlike
structures that give hair cells their name) on their
apical surfaces lead to transmitter release from their
basolateral poles, leading, in turn, to action potentials in
the auditory nerve fibers.
Most types of congenital and acquired hearing loss arise
from damage to, or loss of, cochlear hair cells or their
associated neurons. The incidence of heritable deafness is
high: one child in a thousand is born deaf; another one in a
thousand becomes deaf before adulthood [1,2]. Depending
on the age of onset, hearing impairment can affect oral
language acquisition, cognitive development and psychosocial
development. The prevalence of acquired hearing
loss is rising, as the population of the world increases
and ages and as noise pollution steadily increases. It is
estimated that one in three adults over the age of 65
has a handicapping hearing loss, making this condition
one of the most common chronic disorders, with more
than 250 million affected people worldwide in 2001
(http://www.who.org).
Underlying the irreversibility of hearing loss in
mammals is the inability to replace lost hair cells by cell
division and by regeneration from endogenous cells in the
inner-ear epithelia. Hearing impairment in humans is, in
most cases, a direct consequence of hair-cell loss. Clinically,
the functionality of lost hair cells can be partially
restored by the electrical stimulation of the auditory
nerve, which is achieved by the implantation of electronic
devices. For example, cochlear implants can provide a
subset of suitable patients with improved hearing. Other
avenues of therapy are being explored to find more
biologically based and more widely applicable treatments.
The past year has introduced stem cells into the search
for new approaches to hair-cell regeneration in mammals.
A major advance in the prospects for the use of stem cells
for the replacement of inner-ear cells came with the recent
discovery that hair cells could be generated in vivo from
embryonic stem (ES) cells, from adult inner-ear stem cells
and from neural stem cells [3–5]. These stem cells are
pluripotent, such that all cell types in the inner ear can, in
theory, be regenerated from these cells. We propose that
stem-cell-based treatment regimens might be applicable to
the damaged inner ear as part of future clinical applications
for the treatment of hearing loss.
Stem cells for the treatment of degenerative diseases
Cell-replacement therapywith stem cellshas the potential to
have a massive impact on human health during the coming
decades. The first targets for therapeutic stem-cell applications
are degenerative ailments, such as heart disease,
diabetes, Parkinson’s disease and other neurodegenerative
disorders. The initial results using stem-cell-based generation
of replacement cells for these disorders indicate that
stem cells can be developed into highly specialized cell types
and that these new cells can function in animal models, even
improving the underlying organ function [6–9].
There are three principal sources for stem cells that
have been used to (re-)generate organ-specific cell types:
ES cells, stem cells that are isolated from the organ to be
generated and stem cells from other organs. Consequently,
the regeneration of lost hair cells can, theoretically, involve
ES cells, inner-ear stem cells or stem cells from brain, skin
or the hematopoietic system.
Hair cells from embryonic stem cells
ES cells are derived from the inner cell mass of the
blastocyst. Because they are the precursors for all other
embryonic cells, ES cells have the greatest capacity for
differentiation into multiple cell types, which is termed
pluripotency. ES cells also have the capacity for selfrenewal
and can, therefore, be expanded to large numbers.
The generation of specific cell types by directing ES-cell
differentiation hypothetically offers an extensive resource
for developing clinical applications to replace diseased
or injured cells. Successful application of this strategy
appears to validate this hypothesis, because it has led to
Corresponding author: Stefan Heller (hellers@epl.meei.harvard.edu).
Available online 17 June 2004
Opinion TRENDS in Molecular Medicine Vol.10 No.7 July 2004
www.sciencedirect.com 1471-4914/$ - see front matter q 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.molmed.2004.05.008
the generation of dopaminergic neurons for Parkinson’s
disease [6,8], motor neurons for spinal-cord injuries [10]
and, apparently, insulin-secreting cells for diabetes [7]
(but see [11]). Recently, inner-ear progenitors have been
generated from murine ES cells in vitro [4]. These progenitors
express a set of marker genes that identify them
as cells in the lineage of the hair cells, because these
markers can only be found in this specific combination in
the developing inner ear. After differentiation in vitro, a
subpopulation of the ES-cell-derived progenitors exhibited
a hair-cell phenotype, as revealed by the expression of
characteristic markers such as the transcription factors
Math1 (murine atonal homologue 1) and Brn3.1, which are
important for the generation of, and for maintaining the
maturation of, hair cells [12,13]. Expression of these
transcriptional key regulators was accompanied by the
upregulation of structural hair-cell proteins, such as
the unconventional myosin VIIA [14,15], parvalbumin 3
[16,17] and espin [18,19].
The implantation of genetically labeled ES-cell-derived
inner-ear progenitors into the inner ear of chicken embryos
and following their fate through early otic development
showed that engrafted cells initiated the expression of
hair-cell markers when situated in developing inner-ear
sensory epithelia. Progenitor-derived cells that were
found elsewhere in the inner ear did not express haircell
markers. Consequently, it has been hypothesized that
grafted murine ES-cell-derived inner-ear progenitor cells
can respond to local cues that control (hair) cell-type
specification in the developing chicken inner ear [4].
Although the developing avian inner-ear sensory epithelia
are different from injured or diseased mammalian organ
of Corti or vestibular hair-cell-bearing epithelia, these
results are the first successful approach using ES cells to
generate hair cells in vivo.
Hair cells from adult stem cells
Stem cells have been isolated and propagated from
many adult organs, including the brain, bone marrow,
muscle, heart, skin, eye and, recently, from the inner ear
[3,9,20–23]. Neural stem cells, which have the ability to
differentiate into many neuronal cell types, have been
successfully grafted into the drug-injured mouse inner
ear; the cells survived for several weeks and expressed
markers of mature cell types, including glia, neurons and
hair cells, albeit not in the cochlea [5,24]. Comparison of
the in vitro potential of adult neural stem cells with stem
cells from the inner ear of adult mice revealed two
substantial differences in the potential of the cells to
differentiate into hair-cell-marker-positive cells [3]. First,
the upregulation of hair-cell markers was readily observed
in ,10% of all cells that were differentiated from innerear-derived
stem cells in vitro, whereas adult neural stem
cells that were isolated from the forebrain rarely (,0.1%)
gave rise to hair-cell-marker-positive cells in this assay.
Second, inner-ear stem cells appeared to differentiate
more completely into hair cells than the neural stem-cell
derivatives. This became apparent by the formation of
hair-bundle-like structures that were immunopositive
for specific stereociliary markers (Figure 1k) [3]. In vitro
inner-ear stem-cell-derived cells, after transplantation
into a developing chicken inner ear, upregulate hair-cellspecific
markers in a similar manner to grafted ES-cellderivatives
[3,4].
Adult inner-ear stem cells reside in the sensory epithelium
of the utricle [3] and are a plausible candidate for
the progenitor cells that have been postulated as the
source of hair-cell regeneration in the damaged utricular
sensory epithelium [25–29]. Inner-ear stem cells are
pluripotent because they can develop into many other
cell types outside of the inner ear that are derived from
either ectodermal, endodermal or mesodermal germ layers.
The defining stem-cell feature of inner-ear stem cells is
their high proliferative capacity, which makes it possible to
isolate these cells in the form of clonal floating colonies or
‘spheres’. Proliferation potential is crucial to developing
treatment strategies for hearing loss, because propagation
of these cells might become the foundation of a replacement
strategy for human inner-ear cells.
Do hair cells that are generated from stem cells follow the
native developmental program?
The variety of cellular interactions that have roles during
the complex development and morphogenesis of the inner
ear are