The expression of fibroblast growth factors and their receptors in the embryonic and neonatal mouse inner ear.

Four different fibroblast growth factor receptors (FGFR) are known, three of which have splice variants (known as the b and c variants) in the FGF-binding domain, to give different patterns of sensitivity to the different FGFs. The expression of the b and c variants of the FGF receptors, together with the expression of the ligands FGF1, FGF2, FGF3, FGF7, FGF8b and FGF8c, was determined by quantitative reverse transcription-polymerase chain reaction in developing whole mouse inner ears, and in dissected components of the postnatal mouse inner ear. At embryonic age (E)10.5 days, when the otocyst is a simple closed sac, the receptor most heavily expressed was FGFR2b, relative to the postnatal day 0 level. Over the period E10.5-E12.5, during which the structures of the inner ear start to form, the expression of the different FGF receptors increased 10(2)-10(4) fold per unit of tissue, and there was a gradual switch towards expression of the 'c' splice variants of FGFR2 and FGFR3 rather than the 'b' variants. At E10.5, the ligands most heavily expressed, relative to the postnatal day 0 level, were FGF3, FGF8b and FGF8c. In the postnatal inner ear, the patterns of expression of receptors and ligands tended to be correlated, such that receptor variants were expressed in the same regions as the ligands that are known to activate them effectively. The neural/sensory region expressed high levels of FGFR3c, and high levels of the ligand FGF8b. The same area also expressed high levels of FGFR1b and FGFR2b, and high levels of FGF3. The lateral wall of the cochlea (including the stria vascularis and the spiral ligament) expressed high levels of FGFR1c and FGF2. It is suggested that the different FGF receptors and ligands are expressed in a spatially coordinated pattern, to selectively program cochlear development.

Basic fibroblast growth factor and fibroblast growth factor receptors in adult olfactory epithelium.

Basic fibroblast growth factor (FGF2) stimulates proliferation of the globose basal cells, the neuronal precursor in the olfactory epithelium. The present study investigates the expression of basic fibroblast growth factor and fibroblast growth factor receptors in the adult olfactory epithelium. FGF2 immunoreactivity was expressed widely in the olfactory epithelium, with the highest density of immunoreactivity in the supporting cells. In contrast, most cells in the epithelium expressed FGF2 mRNA. Fibroblast growth factor receptor-1 (FGFr1) immunoreactivity was densest in the basal cell and neuronal layers of the olfactory epithelium and on the apical surface of supporting cells. In the lamina propria FGF2 immunoreactivity and mRNA were densest in cells close to the olfactory nerve bundles. FGFr1 immunoreactivity was heaviest on the olfactory ensheathing cells. Using reverse transcriptase-polymerase chain reaction analysis, the olfactory epithelium was shown to express only three receptor splice variants, including one (FGFr1c) with which basic fibroblast growth factor has high affinity. Other receptor splice variants were present in the lamina propria. Taken together, these observations indicate endogenous sources of FGF2 within the olfactory epithelium and lamina propria and suggest autocrine and paracrine pathways via which FGF2 might regulate olfactory neurogenesis. The observation of only three receptor splice variants in the olfactory epithelium limits the members of the fibroblast growth factor family which could act in the olfactory epithelium. The widespread distribution of receptors suggests that fibroblast growth factors may have roles other than proliferation of globose basal cells.

Lateral interactions account for the pattern of the hair cell array in the chick basilar papilla.

It has been suggested that lateral interactions set up the array of hair cells and supporting cells in the chick basilar papilla. The presence of a hair cell would inhibit adjacent cells from becoming hair cells, and promote the formation of supporting cells. Models of cell specification were tested, starting with a closely packed array of multipotent progenitor cells. Lateral interactions, in which emerging hair cells promoted a supporting cell phenotype in adjacent cells, and in which emerging supporting cells promoted a hair cell phenotype in adjacent cells, produced an array of cells similar to that observed experimentally in the distal and central parts of the basilar papilla. In these areas, the ratio of supporting cells to hair cells is very close to 2:1, each hair cell on average being surrounded by six supporting cells, and each supporting cell being surrounded by three hair cells and three supporting cells. Identical patterns of hair and supporting cells could be produced by models in which either of the lateral inhibitory factors was replaced by a diffusive factor, i.e. a factor which acts on all cells in the model irrespective of position. The agreement of the model with observed cell ratios supports the view that the fate of both hair cells and supporting cells in the chick basilar papilla is a product of cell interactions within the papilla. It is possible that one factor, that provides contact lateral inhibition and promotes the formation of supporting cells, is the Notch/Delta system. It is possible that the other factor is retinoic acid, a diffusive factor that promotes the formation of hair cells.

Expression of EphA4 in developing inner ears of the mouse and guinea pig.

The expression of EphA4, an Eph-class receptor tyrosine kinase, was determined by immunohistochemistry in developing inner ears of the mouse and the guinea pig. In the mouse, EphA4 expression was visible in the fibroblasts of the spiral ligament and in the structures that were to become the osseous spiral lamina. Cochlear nerve ganglion cells expressed ephrin-B2, and the modiolus expressed mRNA coding for ephrin-B3, both transmembrane ligands for EphA4. In contrast, in the guinea pig, cells of the cochlear nerve ganglion expressed EphA4, as did supporting cells of the organ of Corti (Hensen's cells and inner pillar cells). There was also some expression in fibroblasts of the spiral ligament but none in the structures that were to become the osseous spiral lamina. It is suggested that in the mouse, EphA4 may help direct the cochlear innervation towards the organ of Corti by a repulsive interaction, but that this is highly species dependent.

A tyrosine kinase screen of mouse vestibular maculae.

Receptor tyrosine kinases allow extracellular signals to influence intracellular events, while other tyrosine kinases are involved in intracellular signalling. They may therefore be involved in the development, maintenance and repair of the sensory epithelia of the inner ear, since these are believed to be affected by inter- and intracellular signalling. In order to analyse possible tyrosine kinases expressed in sensory areas of the inner ear, a reverse transcription polymerase chain reaction screen of microdissected sensory epithelia was undertaken, using primers targeted at conserved sequences in tyrosine kinase domains. Tissue was taken from the maculae of the mouse vestibular organs, and consisted mainly of hair cells and their supporting cells. Of 80 clones sequenced, 49 coded for tyrosine kinases, and 11 for other known molecules. Further analysis of one of the sequences, for FGF receptor 4, showed a novel variant, expressed in the inner ear and elsewhere, with a variation in the intracellular domain which suggests differential activation of known signalling pathways. Other clones coded for tyrosine kinases expected to be involved in cell surface and intracellular signalling. The technique forms a powerful tool for analysing a range of the tyrosine kinases expressed, and provides a starting point for the analysis of cell-cell signalling in the inner ear.

Fibroblast growth factor receptor-4 splice variants cause deletion of a critical tyrosine.

We have identified two novel isoforms of fibroblast growth factor receptor-4 (FGFR4). They result from alternative splicing of intron 17. Two transcripts, both slightly larger than the one coding for the known mouse FGFR4, are generated. The shortest (FGFR4-17a) includes the 31-most 3'-nucleotides of intron 17; the longest (FGFR4-17b) includes all 114 nucleotides of intron 17. Translation of the FGFR4-17a and FGFR4-17b splice variants predicts that both novel putative FGFR4 isoforms have a truncated C-terminal intracellular tail. The first amino acid residue affected by the insertions in both novel isoforms is Tyr-760, a residue that may play a crucial role in intracellular signaling through stimulation of the phosphatidylinositol-biphosphate pathway.

Fibroblast growth factor receptor expression in outer hair cells of rat cochlea.

Fibroblast growth factors (FGFs) are critical for normal development of the organ of Corti, and may also protect hair cells from ototoxic damage. Four different fibroblast growth factors are known, three of which have different splice variants in the extracellular immunoglobin-like (Ig) III FGF-binding domain, giving different patterns of sensitivity to the different FGFs. Analysis of a cDNA library of rat outer hair cells by the polymerase chain reaction, using isoform specific primers, showed expression only of FGF receptor 3, splice variant IIIc. This allows us to predict the pattern of sensitivity to applied FGFs, which may be useful in targeting outer hair cells selectively during an FGF-based strategy for cochlear therapy.

The expression of messenger RNAs coding for growth factors, their receptors, and eph-class receptor tyrosine kinases in normal and ototoxically damaged chick cochleae.

Messenger RNAs coding for growth factors and receptor tyrosine kinases were measured by quantitative competitive and by semi-quantitative reverse-transcription polymerase chain reaction in whole and dissected chick inner ears. The fibroblast growth factor (FGF) receptor 1 chick embryonic kinase (CEK) 1 was expressed in all structures examined (otocyst, hatchling whole cochlea, cochlear nerve ganglion, and cochlear and vestibular sensory epithelia), although slightly more heavily in the otocyst. The related fibroblast growth factor receptors CEK 2 and 3 were preferentially expressed in the nerve ganglion and in the vestibular sensory epithelium, respectively. FGF1 mRNA was low in early development, increasing to mature levels at around embryonic age 11 days, while FGF2 mRNA was expressed at constant levels at all ages. In response to ototoxic damage, FGF1 mRNA levels were increased in the early damaged cochlear sensory epithelium. Immunohistochemistry for CEK1 showed that normal hair cells expressed the receptor heavily on the hair cell stereocilia, while with early damage, CEK1 came to be expressed heavily on the apical surfaces of the supporting cells. In normal chicks, the CEK4 and CEK8 eph-class receptor tyrosine kinases were expressed relatively heavily by the cochlear nerve ganglion, and CEK10 was expressed relatively heavily by the cochlear hair cell sensory epithelium. The results suggest that the FGF system may be involved in the response of the cochlear epithelium to ototoxic damage. The eph-class receptor tyrosine kinase CEK10 may be involved in cell interactions in the cochlear sensory epithelium, while CEK4 and CEK8 may play a role in the cochlear innervation.

Sequence and tissue distribution of chicken cellular nucleic acid binding protein cDNA.

We sequenced cDNAs coding for chicken cellular nucleic acid binding protein (CNBP). Two slightly different variations of the open reading frame were found, each of which translates into a protein with seven zinc finger domains. The longest transcript contains an in-frame insert of 3 bp. The sequence conservation between chick CNBP cDNAs with human, rat and mouse CNBP cDNAs is extreme, especially in the coding region, where the deduced amino acid sequence identity with human, rat and mouse CNBP is 99%. CNBP-like transcripts were also found in various tissues from insect, shrimp, fish and lizard. Regions with remarkable nucleotide conservation were also found in the 3' untranslated region, indicating important functions for these regions. Quantitative reverse transcription polymerase chain reaction (RT-PCR) indicated that in the chick, CNBP in present in all tissues examined in approximately equal ratios to total RNA. RT-PCR of total RNA isolated from different phyla indicate CNBP-like proteins are widespread throughout the animal kingdom. The extraordinary level of conservation suggests an important physiological role for CNBP.

The incorporation and turnover of radiolabelled amino acids in developing stereocilia of the chick cochlea.

Hair cell stereocilia are composed of packed actin filaments, oriented such that the preferred end for the addition of actin monomers is at the tips of the sterocilia. It has therefore been suggested that when stereocilia grow, they do so from their tips (Tilney and DeRosier, 1986, Dev. Biol. 116, 119-129). In order to test the hypothesis, radiolabelled amino acids were applied to the air-sac of chicken eggs at day 17 of incubation, i.e., at the beginning of a phase in which the stereocilia have achieved their mature width, but are growing rapidly in length. Incorporation of radiolabel was studied autoradiographically, followed by image analysis and averaging grain counts over many hair cells. In contrast to the position expected from the above hypothesis, there was no sign of preferential incorporation of label in the upper part of the stereociliary bundle. The greatest density of labelling was found in the lower part of the bundle, while the upper part of the bundle was under-represented in the autoradiographic averages. The turnover time (to fail to 1/e) was significantly greater in the bundle (16 days) than in the cuticular plate or in the rest of the cell (9 days). The results (i) give no support for the hypothesis that stereocilia grow from the tips, and (ii) suggest that during development at least some components of the stereocilia turn over with a relatively short time course.

Postnatal changes in the reticular lamina of the guinea pig organ of Corti.

The dimensions of the apical surfaces of hair cells were measured in guinea pigs, aged from 3 weeks before term to 25 weeks after birth. In the basal two-thirds of the cochlea, the apical surfaces of the outer hair cells and their supporting cells changed with age, shrinking in a direction radial across the cochlear duct. There was an associated widening of the angle of the 'V' of the rows of stereocilia. Further apically, between 12 and 16 mm from the base of the cochlea, the outer hair cells and their supporting cells underwent the opposite change, becoming wider in a radial direction with age. The changes were seen before birth and continued for more than 3 weeks after birth. The results suggest that the guinea pig cochlea continues certain developmental processes for a considerable time after birth.

The effects of moderate and low levels of acoustic overstimulation on stereocilia and their tip links in the guinea pig.

Guinea pigs were exposed to pure tones of 10 kHz at intensities between 98 and 115 dB SPL for 5-30 min, to produce varying degrees of acoustic trauma. Changes in auditory thresholds were measured electrophysiologically, and the animals were immediately fixed for scanning electron microscopy. Correlation between morphological changes to the hair bundle and losses in threshold, showed that with the smallest degrees of trauma (98 dB SPL for 15 min, mean maximum threshold loss of 22 dB), damage was confined to a small stretch of inner hair cells (IHC), with only subtle changes to the stereocilia of the outer hair cells (OHC). At exposure intensities greater than 102 dB SPL (duration: 15 min) the IHC stereocilia in the centre of the lesion were always substantially disarrayed. Substantial damage to the OHC bundles was seen only with exposures above 110 dB SPL (duration: > or = 5 min), producing threshold losses of 50 dB or more. Tip links were lost only where the stereocilia were disarrayed. It is concluded that the tip links are not the most vulnerable components of the cochlear hair cell, but that relatively low levels of acoustic stimulation can cause significant damage to the stereociliary bundle of the IHCs.

Morphology of the monotreme organ of Corti and macula lagena.

The organ of Corti and macula lagena were studied by scanning and transmission electron microscopy in two species of monotreme, the platypus and echidna. In both species, the organ of Corti had a fundamentally mammalian conformation, with distinct outer and inner hair cells, separated by a tunnel of Corti. However, unlike eutherian mammals, the monotremes had three or four rows of pillar cells, and four to five rows of inner hair cells. The organ of Corti was much shorter than in eutherian mammals, at 4.4 mm (platypus), and 7.6 mm (echidna). While the total number of outer hair cells (3,350 platypus, 5,050 echidna) was many fewer than in most eutherian mammals, the total number of inner hair cells (1,600 platypus, 2,700 echidna) was comparable with that in eutherian mammals. The stereocilia on both inner and outer hair cells underwent a systematic change in orientation across the cochlear duct, with those nearest the tunnel of Corti having their axis of symmetry oriented transversely across the duct, and those on the outer edge of the organ having the axis oriented nearly longitudinally along the duct. The macula lagena had signs of a vestibular epithelium, with tall bundles of stereocilia, a division into areas with bundles of opposing orientation and type I and type II hair cells.

The changing microtubule arrangements in developing hair cells of the chick cochlea.

It has been suggested that microtubules in auditory hair cells might be involved in directing the morphological and hence functional polarisation of the sensory hair bundles. The distribution of microtubules was studied in hair cells of the chick cochlea, during the developmental stages when the stereocilia and cuticular plate were being formed. Cochleae were immunofluorescently labelled with antibodies to tubulin at specific stages in development, and hair cell ultrastructure was observed by electron microscopy. We found that the microtubule array changed from a simple symmetrical apical plate with a central kinocilium before the cuticular plate forms, to a ring with the kinocilium to one side when the cuticular plate begins to form, through to a cup-like arrangement below the cuticular plate once the plate has formed. In the earliest stages, no asymmetries were observed in the distribution of the microtubules, suggesting that structures other than the microtubules set up the functional polarisation of the stereociliary bundle.

Early hair-cell degeneration in the extreme apex of the guinea pig cochlea.

Guinea pigs, aged from 3 weeks before term to 31 weeks after birth, were prepared for scanning electron microscopy. Examination of the extreme apex of the cochlea showed apparently pathological hair cells, even 3 weeks before term. The pathologies included loss and fusion of stereocilia, and the formation of giant stereocilia. The pathologies were most prevalent on row 3 of outer hair cells, declining to outer hair cell rows 2 and 1, with the inner hair cells being least affected. The abnormalities increased with time, increasing rapidly over the first few weeks of life, and more slowly thereafter. It is suggested that early degeneration in the extreme cochlear apex forms a novel model for spontaneous hair cell degeneration, with applicability to other types of spontaneous hair cell degeneration.

An analysis of actin isoforms expressed in hair-cell enriched fractions of the chick basilar papilla by the polymerase chain reaction technique.

Actin mRNA was characterised in hair-cell enriched fractions of the chick basilar papilla, by means of the reverse-transcription polymerase chain reaction technique. Primers were directed against the 3' untranslated portions of the actin mRNAs. Evidence for beta-cytoplasmic and gamma-cytoplasmic actin mRNA was found; no evidence was found for alpha-skeletal, alpha-cardiac or type 5 cytoplasmic actin mRNAs. Since beta-actin is known to form bundles of filaments whereas gamma-actin does not, this suggests that the hair-cell stereocilia are composed of beta-actin.

Early events in auditory processing.

During the last year, further evidence has appeared concerning the basis of frequency selectivity in the cochlea, which may ultimately depend on a motile mechanism residing within the walls of the outer hair cells. Evidence has also appeared on hair-cell mechanotransduction, and on the way that the stimulus is coupled to the mechanotransducer channels.

A model for the mechanics of the stereociliar bundle on acousticolateral hair cells.

The stereociliar bundle on acousticolateral hair cells was modelled as a series of stiff rods (stereocilia), and springs (stereociliary links and rootlets). Predictions were made for the coupling of stimulus-induced deflections between the stereocilia on the hair bundle, and for the stretches of the different classes of link. Comparison of the results with the measured mechanical properties of hair bundles suggests that in the bullfrog sacculus the stiffness of a side link and a tip link are related to the rootlet's contribution to the stiffness of a stereocilium to deflection in approximately the ratio > or = 400:100:1. The results show that stretch of the tip links is closely related to the deflection of the hair bundle over a wide range of model parameters, while the stretch of the side links is more variable, and in some types of bundle the mean stretch of the side links may be zero or negative. The results are in accordance with the view that the tip links are in an appropriate position to detect the deflections, while the main role for the side links may be to couple the deflections between the stereocilia. The mechanical consequences of bundles of different configurations, as seen in different hair cell types, are investigated.

Mechanoelectrical transduction by hair cells.

Hair cells of the inner ear are one of nature's great success stories, appearing early in vertebrate evolution and having a similar form in all vertebrate classes. They are specialized columnar epithelial cells, with an array of modified microvilli or stereocilia on their apical surface, interconnected by a series of linkages. The mechanical stimulus causes deflection of the stereocilia, stretching linkages between them, and opening the mechanotransducer channels. On a slower timescale, hair cells adapt in order to maintain optimum sensitivity, with an adaptation motor within the stereocilia acting to keep the resting tension on channels constant.