Specific expression of the retinoic acid-synthesizing enzyme RALDH2 during mouse inner ear development.

Retinoid binding proteins and nuclear receptors are expressed in the developing mouse inner ear. Here, we report that the retinaldehyde dehydrogenase 2 (Raldh2) gene, whose product is involved in the enzymatic generation of retinoic acid (RA), exhibits a restricted expression pattern during mouse inner ear ontogenesis. The Raldh2 gene is first expressed at embryonic day (E) 10.5 in a V-shaped medio-dorsal region of the otocyst outer epithelium, which evolves as two separate domains upon otocyst morphogenesis. At E14.5, Raldh2 is expressed in two areas of the utricle epithelium and specific regions of the saccule and cochlear mesenchyme. Later, Raldh2 transcripts are restricted to two cochlear areas, the stria vascularis and Reissner membrane. Raldh2 mesenchymal expression did not correlate with migrating neural crest-derived melanoblasts. These restricted expression domains may correspond to specific sites of RA synthesis during inner ear morphogenesis.

Spatio-temporal distribution of cellular retinoid binding protein gene transcripts in the developing and the adult cochlea. Morphological and functional consequences in CRABP- and CRBPI-null mutant mice.

The expression patterns of the mouse cellular retinoid binding protein genes were investigated by in situ hybridization analysis in the inner ear from 10.5 days post coïtum (dpc) up to the adult stage. The cellular retinoic acid binding protein II (CRABPII) and cellular retinol binding protein I (CRBPI) were present in a widespread and abundant pattern in cochlear structures during embryogenesis. Expression of the cellular retinoic acid binding protein I (CRABPI) is restricted during development in Kölliker's organ whilst cellular retinol binding protein II (CRBPII) is only visible after birth with a ubiquitous distribution in most regions of the cochlea including nervous components. No CRABP or CRBP transcripts were observed in the auditory receptors. Morphological observations of CRBPI- and CRABPI/CRABPII-null mutant fetus at 18.5 dpc do not show any structural modification at the level of the organ of Corti. Furthermore, electrophysiological tests performed by measuring distorsion-product otoacoustic emissions and auditory brainstem evoked responses did not present significant alteration of the auditory function for the different types of mutants. The expression of retinoid binding proteins in cochlear structures during embryogenesis could suggest important roles for these proteins during ontogenesis and morphogenesis of the inner ear. Despite these observations, morphological and functional data from mutant mice did not present obvious modifications of the cochlear structures and auditory thresholds. It is therefore unlikely that CRABPs and CRBPI are directly involved in development of the cochlea and hair cell differentiation.

Effects of growth factors on the hair cells after ototoxic treatment of the neonatal mammalian cochlea in vitro.

The aim of this study was to test the possible regenerative potential of several molecules and growth factors such as retinoic acid (RA), insulin, epidermal growth factor (EGF) and transforming growth factors alpha (TGFalpha) and beta (TGFbeta) on the neonatal cochlea in vitro after neomycin intoxication. Our studies show that cochlear sensory epithelium behaves differently while maintained in various culture conditions, although we did not observe regeneration whatever the molecules or growth factors tested. The ototoxic action of neomycin in vitro produced a specific death of hair cells, except in the apical region. Organ of Corti of rats 3 days after birth always presented two regions that responded differently to the antibiotic: a widespread scar region extending from the basal cochlea up to the beginning of the apical turn, where most hair cells had disappeared, and a second region called the resistance region localized in the apex, and which was more or less developed depending on culture conditions. The length of the resistance region was modulated by molecules or growth factors added to the feeding solution suggesting that some of them could produce a protective action on hair cells against neomycin. Slight protection effects may be found with RA and insulin, however, the most definite protection results from the combination of insulin with TGFalpha as shown by the large increase in the length of the resistance region compared to organ of Corti treated with antibiotic alone. The tested molecules and growth factors did not promote cochlear hair cell regeneration in vitro after neomycin treatment, however some of them may offer a protective action against ototoxicity.

CD1 hearing-impaired mice. I: Distortion product otoacoustic emission levels, cochlear function and morphology.

The levels of distortion product otoacoustic emissions (DPOAEs) were measured in a strain of hearing-impaired mutant mice (CD1) at various stages of outer hair cell impairment and compared to those of a control inbred strain (CBA/J). Parallel measurements of cochlear potentials and auditory brainstem evoked responses (ABRs) were performed and surface preparations of organs of Corti were observed using phalloidin staining of filamentous actin. Comparison of DPOAEs (elicited by stimulus levels of 60 and 70 dB SPL) with standard functional tests allowed the categorization of CD1 ears into two groups on the basis of the presence or absence of DPOAE, which corresponded to mean ABR thresholds greater or less than 40 dB nHL respectively. When adopting ABR threshold as the gold standard, this procedure yielded rates of false-positives and -negatives ranging from 5 to 16%. However, individual predictions of electrophysiological function from DPOAE levels were not accurate, owing to their large variance, and attempts to optimize stimulus levels did not reduce this variance. In contrast, the profiles of DPOAE level vs. f2 exhibited large correlations with ABR threshold profiles as a function of f2. It was also noteworthy that the mean levels of DPOAEs in CD1 mice recorded in frequency intervals with normal ABR thresholds were significantly smaller than those of CBA/J mice. Although hearing loss was revealed early both by DPOAEs and by other functional tests, surface preparations often remained normal until about 3-4 months of age.

CD1 hearing-impaired mice. II. Group latencies and optimal f2/f1 ratios of distortion product otoacoustic emissions, and scanning electron microscopy.

In our companion paper (Le Calvez et al., 1998), the levels of distortion product otoacoustic emissions (DPOAE) were collected in the ears of CD1 mice with progressive degeneration of cochlear outer hair cells (OHC). Their comparison to standard functional measurements such as auditory-evoked brainstem responses (ABR) showed that CD1 ears could be classified as normal or impaired in a frequency-specific manner using DPOAE levels. The present work reports how DPOAE phases and levels of young CD1 mice were affected by varying the frequency ratio of eliciting stimuli at frequencies f1 and f2. Normally hearing CBA/J mice served as controls. The rate of phase change of DPOAE when f1 was varied and f2 was fixed allowed the group delay of DPOAE to be derived. The changes of DPOAE levels during this procedure disclosed bandpass characteristics that several reports (Fahey and Allen, 1986; Brown and Gaskill, 1990) assumed to be the reflection of important features of cochlear micromechanics, possibly in relation to the coupling of OHCs to the tectorial membrane. Group delays became significantly shorter when ABR thresholds exceeded 40 dB elevation. The bandpass filter characteristics strikingly depended on auditory function so that the optimal ratio f2/f1 progressively shifted from 1.24 to 1.50 or more when hearing loss increased. A difference was also noted between CD1 ears whose ABR thresholds were not yet increased and control CBA/J (optimal ratio 1.20). Scanning electron microscopy disclosed a variety of often minor OHC lesions that were only roughly correlated with cochlear function. However, the presence of abnormalities in the reticular lamina associated with early changes of DPOAE fine structure as a function of f2/f1 supported the hypothesis of some involvement of micromechanical features in the bandpass filter characteristics of DPOAE. The sensitivity of their measurement in pathological situations is potentially interesting.

Developmental distribution of astrocytic proteins in the rat cochlear nucleus.

To investigate the developmental distribution of cochlear nucleus (CN) astrocytes, we used immunocytochemical localization of glial fibrillary acidic protein (GFAP) and S100beta in rats at 0, 5, 10, 15, 21, 30 postnatal days plus the adult. Differential developmental trends were observed for both proteins. The spatial distribution showed a progressive increase of the number of GFAP-immunoreactive (GFAP-IR) astrocytes during development. GFAP positive cells occurred first in the granule cell domain of the ventral CN and in the molecular cell layer of the dorsal CN, then followed an outside to inside pattern of progression. The GFAP-IR reached an adult distribution 1 month after birth. By contrast with GFAP, the apparition of S100beta-immunoreactivity (S100beta-IR) was abrupt (between 0 and 5 days) followed by a rapid stabilization of density and distribution of IR cells (between 15 and 21 days). The developmental distribution of S100beta-IR cells occurred from the posterodorsal region and progressed toward a rostroventral direction. With contrast to GFAP-IR astrocytes, S100beta-positive cells were mainly restricted to the central part of the CN, while only few IR astrocytes were observed in the granule cell domain of the ventral CN or in the molecular cell layer of the dorsal CN. This differential distribution suggests that both antigens were expressed by two different cell populations at least, it is obvious during the first postnatal week. The gradual expression of GFAP and S100beta is interpreted as reflecting the time course of astrocytic maturation. These data suggest that the maturation of CN astrocytes may be linked to the final maturation of CN neurons.

Localization of neurotrophin-3-like immunoreactivity in the rat cochlear nucleus.

Immunohistochemistry as well as immunohistofluorescence were used to investigate the distribution of the neurotrophin-3 (NT3) in the adult rat cochlear nucleus. We found a widespread distribution of NT3 immunolabeled neurons throughout the three divisions of this nucleus. NT3-like immunoreactivity was clearly population-specific, with some cell groups heavily (various small neurons and granule cells) or moderately (large neurons of the ventral cochlear nucleus) stained, while others remained negative (a major fraction of medium and large neurons of the dorsal cochlear nucleus). Double-labeling experiments were performed using antibody against the glial fibrillary acid protein, a classic marker for mature astrocytes. This colocalization study revealed that NT3 immunoreactivity was also present in a subpopulation of astrocytes, particularly in the glia limitans and their projections. Numerous small cells also colocalized NT3 together with the glial marker in the granule cell domain and in the molecular cell layer of the dorsal cochlear nucleus. These results suggest that NT3 may exist in widespread populations of adult cochlear nucleus neurons as well as in glial cells. This abundant distribution of NT3-like immunoreactivity implies that this neurotrophin may have an important role in the continued maintenance of mature cochlear nucleus and makes it an attractive candidate for playing a role in regulation or stabilization of neuronal circuits in this nucleus.

Spatial distributions of retinoic acid receptor gene transcripts in the prenatal mouse inner ear.

The expression patterns of the three mouse retinoic acid (RA) receptor gene isotypes (RARalpha, RARbeta, and RARgamma) and retinoid X receptor gene isotypes (RXRalpha, RXRbeta, and RXRgamma) have been investigated by in situ hybridization analysis of their RNA transcripts in the inner ear of mouse fetuses at 18.5 days of gestation. Two RARs (RARalpha and RARgamma) and two RXRs (RXRalpha and RXRbeta) presented an almost ubiquitous transcript distribution with overlapping expression in several regions of the cochlea, such as Kölliker's organ, the organ of Corti, the spiral limbus, and nervous structures. The organ of Corti showed an enhanced in situ labeling with RARalpha and RXRbeta. By contrast, RARbeta and RXRgamma displayed more restricted expression patterns. RXRgamma in particular was strongly expressed in Kölliker's organ and in the spiral ganglion. This expression pattern suggests that RA may be involved in the differentiation of several cochlear cell types. Moreover, the colocalization of several RAR and RXR gene transcripts suggests possible heterodimerization between these receptors in several regions of the cochlea.

Factors modulating supernumerary hair cell production in the postnatal rat cochlea in vitro.

It has been shown in the past that extra hair cells or supernumerary cells can be produced when neonatal cochleae are maintained in vitro. In this report, we investigated the effects of the culture methods, molecules and growth factors that are thought to be involved in cell proliferation. Quantitative studies of supernumerary hair cells were made by measuring the cell density over the entire spiral lamina at two postnatal stages: birth and 3 days after birth. With a standard feeding solution without serum, a difference in cell density was observed between the two methods of culture. Cochlear explants in a standard feeding solution supplemented with serum showed an increase of cell density only when the explantation is made at birth. Retinoic acid added to the standard feeding solution did not increase the hair cell density, while insulin induced an increase, especially at 5 micrograms/ml. Several growth factors were tested. Epidermal growth factor (EGF) presented a dose dependent effect with an increase of up to 30% of hair cell density that was observed in the basal region when the explantation was made at birth. Transforming growth factor-alpha did not induce an increase of cell density, whereas transforming growth factor-beta presented an effect on hair cell density, with a dose dependent effect reaching 37.4% for the basal inner hair cells. Interpretation of these results is limited because of the lack of data concerning the presence of specific membrane receptors. One possibility is that insulin stimulates hair cell differentiation from existing undifferentiated cells. Another hypothesis may be related to the EGF and transforming growth factor-beta, where these molecules might induce transdifferentiation of cells by acting on the transmembrane molecules and the extracellular matrix.

Age-related changes in GFAP-immunoreactive astrocytes in the rat ventral cochlear nucleus.

The age-related changes in the ventral cochlear nucleus (VCN) as revealed by glial fibrillary acid protein (GFAP) immunoreactivity were analyzed in the following age groups: 3-, 6-, 12-, 18-, and 24-month-old Sprague-Dawley rats. A cartographic and a quantitative analysis showed a significant increase in the number of GFAP positive astrocytes during the first year of life and a significant decrease in older rats. We also observed an age-induced modification in the spatial distribution of GFAP positive astrocyte. In the anterior part of the VCN of the 3- and 6-month-old rats, we observed a significant decrease in the rostro-caudal as well in the dorso-ventral axes. In the posterior part of the VCN, a significant decrease in the dorso-ventral axis could be also observed, but no significant difference in the spatial distribution was obtained in the rostro-caudal axis. In older rats, the distribution appeared homogeneous throughout the nucleus. Additionally, aging was associated with a significant increase in GFAP positive astrocyte sizes, except for immunolabelled astrocytes in the granule cell layer. The different levels of GFAP expression occurring in the VCN during normal aging could reflect a progressive decline of cellular activity in the VCN, without severe cell degeneration or synaptic loss.

Monitoring of functional changes after transient ischemia in gerbil cochlea.

Ischemia and reperfusion are involved in numerous sensorineural pathologies. A model of reversible cochlear ischemia has been designed in Mongolian gerbil. Selective labyrinthine ischemia of variable duration (4-10 min) was achieved through a posterior transcranial approach. Ischemia and reperfusion were controlled with the help of laser Doppler velocimetry. Functional changes were monitored every 1-10 s throughout experiments, using cochlear potentials and otoacoustic emissions. After interruption of blood flow, all signals rapidly began to decay. In contrast to cochlear potentials, otoacoustic emissions always exhibited a plateau before reaching noise floor only after approximately 4-5 min. Upon ischemia release, cochlear blood flow recovered instantly and completely and cochlear potentials rapidly improved in most cases, in contrast to otoacoustic emissions that underwent a delayed decay after immediate partial recovery. The phase and group latency of otoacoustic emissions exhibited only small changes throughout ischemia and reperfusion, suggesting adaptive rather than damaging mechanisms. Cochlear function returned to normal after 5 min 30 s ischemia but longer complete ischemia sometimes led to irreversible damage despite the systematic presence of some recovery just after ischemia release. This behavior suggests that reperfusion in itself can be deleterious to a sensorineural organ and this model can be useful for identifying the noxious mechanisms of ischemia and reperfusion.

Ontogenesis of rat cochlea. A quantitative study of the organ of Corti.

A systematic quantitative set of data concerning the organ of Corti in developing Sprague-Dawley rats at intervals from 18 days of gestation to 10 days after birth (DAB) is provided in this study. Using phalloidin staining, the total number of inner and outer hair cells, the whole length of cochlea, as well as the diameter of inner and outer hair cells and the intercellular space between inner hair cells were determined in order to analyze the quantitative change of inner and outer hair cells during development and to explore some roles of the factors regulating the growth of cochlea. The results show that: (1) The length of cochlea approached its adult size by 7DAB. (2) The growth of the extreme part of the apex was responsible for the delayed elongation of the cochlea. (3) Growth in the cochlear length mainly results from an increase of cell diameter tempered by a decrease of intercellular space. (4) The adult size of inner and outer hair cells was obtained by 7-14DAB. (5) The final number of inner and outer hair cells was reached at 3DAB and remained constant through adulthood. No significant hair cell overproduction and cell death were observed during ontogenesis of the cochlea. The negligible importance of overproduction and missing hair cells during hair cell differentiation suggest that there is a precise regulation phenomenon for producing the right spatial organization of the organ of Corti.

Modification of tonotopic representation in the auditory system during development.

During the early development of the bird and the mammalian peripheral auditory system, a restricted range of low--mid frequencies is recorded in immature animals. These early recordings are correlated to the base or mid-basal region of the cochlea which codes high frequencies in the adult. In order to reconcile the functional observations with anatomical ones, two main hypotheses have been put forward: one called the development of the place principle derived from observations of acoustic trauma in chick cochlea and a second derived from auditory nerve fiber recordings in kittens. Whatever the theories, the tonotopic shift during development is a well-established phenomenon in both birds and mammals that could be explained by a synthetic theory including active and passive cochlear processes. The tonotopic shift observed in the central auditory system mimics quite closely the frequency representation of the peripheral auditory system. The same trend is observed in all auditory nuclei including the cortex, except that the frequency representation is more complex because it shows tonotopic maps that can be twisted in three dimensions. From current observations, there is a simultaneous onset of tonotopic maps across auditory nuclei up to the cortex. A hypothesis is presented related to the frequency changes observed in the cochlea that affect the central auditory pathway, along with possible consequences on auditory behavior.

Neurotrophin receptor immunostaining in the rat ventral cochlear nucleus.

By virtue of its known segregated distribution of cell types, their known neurotransmitters and neurophysiologic properties, the cochlear nucleus is an excellent model and provides the opportunity to study the relation between neurotrophins and their receptors along with the functional properties of the adult cochlear nucleus. To investigate the potential role of neurotrophins in the mature cochlear nucleus, we determined the expression of the three major neurotrophin tyrosine kinase receptors (Trk) in the adult rat ventral cochlear nucleus, as revealed by antibodies against the full-Trk proteins. A qualitative and a cartographic analysis showed a widespread distribution of the three Trk receptors throughout the nucleus. Immunostaining was mainly restricted to neurons as shown by the lack of double immunostaining with specific markers for glial cells. However, we observed variability in immunostaining for given receptors. Three classes of cells were distinguished by their specificity for Trk receptors. The first one was a cell population that stained for TrkA or TrkB. This population characterizes the majority of small and small round neurons and fusiform cells. The second group consists of TrkC-immunolabeled cells and comprises the majority of spherical, globular, granule and small multipolar cells. The third group consists of cells that seem to be immunopositive for all three Trk receptors. This group includes in part multipolar, giant and octopus cells. A possible correlation between Trk expression and cell functional properties is suggested: TrkA- and TrkB-immunoreactive cells could include inhibitory neurons while cells stained for TrkC could include excitatory neurons. The abundant and widespread neuronal distribution of signal-transducing forms of TrkA, TrkB and TrkC predicts that their cognate ligands may exert significant effects on a large proportion of neurons within the mature ventral cochlear nucleus.

The spontaneous appearance of hair cell-like cells in the mammalian cochlea following aminoglycoside ototoxicity.

Young rats (in vivo) and cochleas from neonatal rats (in vitro) were treated with ototoxic antibiotics. Scanning electron microscope observations of the cicatricial epithelium of the former outer hair cell region revealed cells with a tuft of microvilli at their apical surface that could contain actin filaments, as observed by phalloidin staining. The apical organization of these hair cell-like cells was reminiscent of fetal hair cells topped with a bundle of microvilli. During both in vivo and in vitro observations, and despite the use of several growth factors in vitro, these hair cell-like cells did not differentiate into mature sensory cells. These hair cell-like cells might represent an attempt by the former sensory epithelium to regenerate.

Development of the auditory receptors of the rat: a SEM study.

Fetal and postnatal ontogenesis of the rat cochlea, from the 16th gestational day (16DG) until 3 months post partum, were studied using scanning electron microscopy with emphasis on the stereocilia during the earliest stages of development. The epithelium of the cochlear duct in 16DG rat consisted of plygonal cells topped with numerous microvilli and one central kinocilium, which form the so-called Kölliker's organ. Inner hair cells (IHCs) appeared at 18DG in the basal cochlea. They were characterized by tufts of cilia of the same height and with a kinocilium. The first outer hair cells (OHCs) can be seen at 20DG. The earliest stages of ciliary differentiation, at 18DG for IHCs and 20DG for OHCs, were similar on both types of cells and were characterized by the presence of round bundles of cilia arising from the surrounding microvilli. A three-dimensional V-shaped organization for OHCs and the linear arrangement for IHCs appeared by the end of the first postnatal week, accompanied by the disappearance of transient cilia on the modiolar side of the hair cell and the kinocilium on the external side. The apical pole of OHCs reached adult-like morphology before that of IHCs. Various links between stereocilia were detected already at birth. Morphometric analysis showed that auditory cells from the base of the cochlea reached adult size by the end of the first postnatal week while those from the apex increased their size later. A review of the literature including comparative observations across species on the ontogenesis of the stereocilia shows that hair cells of the stato-acoustic system may present the same early ontogenesis.

Distribution of glial fibrillary acidic protein (GFAP) in the cochlear nucleus of adult and aged rats.

The age-related change in glial fibrillary acidic protein (GFAP) immunoreactivity was analyzed in young (3 months) and old (24 months) adult rat cochlear nuclei (CN). Quantitative analyses show a significant increase with age, in the number of GFAP positive astrocytes and processes in the old adult when compared with the young adult rat. There was also a differential distribution of GFAP immunoreactivity in the young adult CN where it predominates in the granular cell region, whereas in old rats, the GFAP immunoreactivity distribution was homogeneous in all parts of the nucleus. There was no change in the total number of neurons between these two stages in any part of the nucleus except for the antero-ventral CN, where a decrease in neuronal number was observed in the aged rats. The increase in GFAP immunoreactivity was related to an increase of both GFAP positive astrocyte number and processes. The increase of GFAP positive astrocytes may be due either to an alteration of auditory nerve fibers, changing the trophic interactions with post-synaptic cells, or to intrinsic alterations of CN neurons and local circuits reflecting aging of the CN.

Fimbrin expression in the developing rat cochlea.

The expression of fimbrin in the developing rat cochlea was analyzed using an immunohistochemical technique with fimbrin antibody. The cochlea displayed temporal and lateral-longitudinal gradients for fimbrin expression during development. Fimbrin immunoreactivity first appeared in the inner hair cell stereocilia of the basal turn on the first gestational day studied (day 18). At birth, both inner (IHC) and outer hair cell (OHC) stereocilia of the basal turn showed positive labeling with fimbrin antibody. The progression of appearance was always from IHCs to OHCs and fimbrin immunostaining appeared in the apical hair cells by postnatal day 6. Immunostaining was restricted to stereocilia and the cuticular plate, and no immunoreactivity was observed in neighboring structures of the epithelium. Double labeling using both fimbrin antibody and phalloidin binding revealed similar chronological expression from the earliest stage studied. Increasing fimbrin immunoreactivity was observed in hair cells until late postnatal and adult stages. This study suggests that fimbrin is expressed with F-actin during development and fimbrin together with actin may constitute the two basic molecules that participate in stereocilia formation. We speculate that fimbrin may help maintain the parallel growth of actin filaments within the stereocilia. These data additionally support previous findings that hair cell maturation occurs from the base to the apex and from IHCs to OHCs.

N-Methyl-D-aspartate receptor subunits NR1 and NR2C are overexpressed in the inferior colliculus of audiogenic mice.

Some non-DBA2 Albino Swiss mice exhibit noise induced epileptic seizures during a short period of postnatal development. Because N-methyl-D-aspartate (NMDA) glutamate ionotropic receptors are involved in the occurrence of audiogenic seizures, we investigated by in situ hybridization methods, the expression of the different subunits (NR1, NR2A, NR2B, NR2C) of this receptor in the central nucleus of the inferior colliculus (IC), a main relay of the auditory pathways. At postnatal day 20, the NR2C subunit is highly expressed in the IC of convulsive mice, while in non-convulsive mice a slight signal is only found for NR1, NR2A, and NR2B. In adult mice, the NR1 and NR2A signals are observed while the NR2B signal is almost undetectable. The audiogenic susceptibility may be related to the transient expression of the NR2C subunit during a brief neonatal period during which synaptic reorganization happens.