Pathological changes during the development of the vestibular sensory and ganglion cells of the bronx waltzer mouse. Scanning and transmission electron microscopy

DevelopmentaIBrain Research, 18 (1985) 285-295 Elsevier

285

BRD 50169

Pathological Changes During the Development of the Vestibular Sensory and Ganglion Cells of the Bronx Waltzer Mouse. Scanning and Transmission Electron Microscopy DANIELLE DEMI~MES and ALAIN SANS INSERM - U. 254, U.S. T.L., Laboratoire de Neurophysiologie Sensorielle, 34060 Montpellier, Cedex (France) (Accepted September 17th, 1984) Key words: Bronx waltzer mutant mouse - - vestibular receptors and ganglion - - development - - structural abnormalities

Vestibular receptors and ganglia of homozygous Bronx waltzer (by~by) mice were investigated by scanning and transmission electron microscopy at various stages between 3 days and 90 days after birth. Scanning electron microscopy revealed that there was already a considerable lack of hair bundles in the maculae utriculi, as well as in the cristae ampullares by the 3rd day after birth. During development, the growth of the remaining hair bundles was observed but the most of them exhibited morphological abnormalities. Transmission electron microscopy revealed early degeneration of sensory cells followed by delayed maturation of the remaining sensory cells. The sensory cells which seem unaffected displayed immature features in adult animals. In type I hair cells, the calyces were incomplete, contacts between the cell and the afferent calyces were immature and synaptic bodies persisted. In some type II hair cells, there was an abnormal overabundance of afferent nerve endings, which implies that these type II cells could be immature type I cells. Immature features were also observed in the vestibular ganglia, particularly the absence of the myelin sheath around the perikarya. We discuss the relationship between these vestibular morphogenetic abnormalities and those described in the cochlear system.

INTRODUCTION H e r e d i t a r y inner disorders exist in several different forms in the mouse, and the anatomical and electrophysiological studies d e v o t e d to t h e m are of great interest, since different categories of mouse mutants can be correlated to certain equivalent abnormalities in the human inner ear and can be suitably used as models for a b e t t e r understanding and interpretation of the associated hereditary h u m a n diseasesS,6,z9. One of the m u t a n t genes affecting the inner ear is n a m e d Bronx waltzer (symbol by). This new mutation, described by D e o l and G l u e c k s o h n - W a e l s c h 8 in the mouse, is autosomal recessive. H o m o z y g o t e s (by~by) have a circling or waltzing-type behavioral disorder and m o r p h o g e n e t i c abnormalities in the cochlea, as well as in the vestibular sensory epithelia. In the vestibular system, degenerative changes and malformations have, until now, only been described with light microscopy in the maculae and the cristae ampullares 7. D e o l supposes that this d e g e n e r a t i o n

appears so early in the d e v e l o p m e n t that these organs p r o b a b l y never function. Thus, this m u t a t i o n is very interesting from an electrophysiological point of view1,2 since it offers a unique o p p o r t u n i t y to examine electrophysiological responses from cochleas containing almost exclusively one type of sensory cell, the outer hair cell. Morphological analysis, which defines the onset and decline of these hereditary malformations in the vestibular system, can also provide a better understanding of the physiology and development of the vestibular receptors. The present r e p o r t describes m o r e extensively these abnormalities in the bv/bv mouse, using transmission and scanning electron microscopy in the sensory epithelia of the vestibular receptors and vestibular ganglion cells. O u r objective was to d e t e r m i n e whether degeneration, as in the cochlear system, is selective for one of the two types of hair cells (type I and II) and w h e t h e r m u t a t i o n preferentially affects either the afferent or the efferent system, p r o b l e m s which at present have not been resolved.

Correspondence: D. Demfmes, U.S.T.L., Laboratoire de Neurophysiologie Sensorielle, Place E. Bataillon, 34060 Montpellier, Cedex, France. 0165-3806/85/$03.30 © 1985 Elsevier Science Publishers B.V. (Biomedical Division)

286

Fig. 1. Scanning electron micrographs. A: lateral crista from a 3-day-old control mouse (+~by). B: lateral crista from a 3-day-old bv/bv mouse showing the lack of hair bundles. C: close-up view of the surface of a crista ampullaris from a 3-day-old by~by m o u s e with short hair bundles and cuticular plates devoid of sensory hairs bulging upward into the endolymph. D: crista ampullaris from a 5-day-old by~by mouse, t, taurus. E: surface view of a macula utriculi from a 12-day-old bv/bv mouse showing co-existence of short and longer hair bundles. F: crista ampullaris from a 12-day-old by~by m o u s e showing the growth of the remaining hair bundles in comparison with a 3-day-old stage. Figs. A - F : bar = 10 ,um.

287

Fig. 2. Scanning electron micrographs of vestibular receptors from bv/bv mice at adult stage. A: surface topography of a macula utriculi. The remaining hair bundles surround the region of the striola. B: crista ampullaris with an evident loss of hair bundles. C: macula utriculi showing the cuticular plates with abnormal hair bundles. D: abnormal hair bundles of crista ampullaris. E: hair bundle of a macula utriculi showing scarcity and disorganization of the hairs. Figs. A, B: bar = 10 #m; C, D, E: bar = 5 am.

288 MATERIALS A N D M E T H O D S

RESULTS

This study is based on a total of 17 mice, 11 of them homozygous for the bv gene (bv/bv) and 6 of them heterozygous. 5, 10, 35 and 90 day-old mice were studied with transmission electron microscopy (TEM) and 3, 5, 12, 35 and 90 day-old mice with scanning electron microscopy (SEM). Since the waltzing behavior of the bv/bv mice was not always distinguishable, the distinction between homozygous and heterozygous mice was based on criteria which are clearly visible with SEM. In the same animal, the receptors of one labyrinth were observed with SEM, and those of the other labyrinth were observed with TEM. The animals were bred in our laboratory from a stock maintained by Deol at the University College of London. The adult mice were anaesthetized with intraperitoneal pentobarbital and intracardially perfused with a solution of 3.5% glutaraldehyde in 0.1 M phosphate buffer; this vascular perfusion was simultaneously accompanied by in situ fixation performed directly through the oval window. The young mice were sacrificed by decapitation under anaesthesia and dissection was immediately performed in the same fixative. For TEM, the vestibular receptors and the vestibular ganglia were dissected out and postfixed for about one hour in a 1% osmium tetroxide buffered solution followed by rapid dehydration, and then embedded in Araldite. Semithin sections were stained with 1% toluidine blue to visualize all the sensory epithelia, and ultrathin sections were cut, stained with uranyl acetate and lead citrate, and then examined. For SEM, the specimens were perfused according to the same fixative procedures, dehydrated, dried in a critical-point dryer using liquid CO2, and then coated with a very fine film of gold at a pressure of 5 x 10-2 Torr.

SEM observations Three days after birth, the cristae ampullares and the maculae utriculi exhibited significant pathological features. Beginning at this stage, the sensory epithelia of the by~by mice displayed an obvious scarcity of hair bundles compared to the epithelia in the vestibular receptors of the +/bv mice (Figs. 1A, B). This lack of hair bundles did not preferentially affect any particular area on each receptor. On all the epithelial surfaces, very numerous randomly scattered cuticular plates were observed. They totally lacked hairs, (kinocilium and stereocilia), and were bulging upward into the endolymphatic space (Fig. 1C). Several cuticular plates covered with hair bundles in different stages of growth were also observed. It should be noted that on the otolithic membrane of the maculae utriculi, the otoconia had a normal configuration. Five days after birth, the pathological features were similar. However, the growth of hair bundles on the remaining cells was observed (Fig. 1D). Twelve days after birth, a significant lack of hair bundles was still observed in the cristae ampullares as well as in the maculae utriculi. In the cristae, this lack of hair bundles did not preferentially involve any specific area of the epithelia, but in the utricle, a greater number of cells persisted in a peripheral crown all around the offset zone of the striola. Some of the hair bundles on the remaining sensory cells were of normal size, others had numerous abnormalities. A coexistence of short and longer bundles was also seen, the continuing presence of short bundles at this stage being a feature of immaturity (Figs. 1E, F). Among the abnormal forms, some cuticular plates showed no signs of stereocilia. There were fewer of these ob-

Fig. 3. A. Light micrograph of a toluidine-blue-stained semithin section from the vestibular ganglion of an adult bv/bv mouse. Note the densely packed neurons. B - D . Electron micrographs from vestibular ganglion in an adult by~by mouse. B: two closely apposed neurons showing the single layer of glial sheath typical of an immature stage (arrow). C: long glial digitations (arrows) intermingled with evaginations of the neuronal cytoplasm, n, neuron; f, myelinated fiber. D: an unusual example of the apposition of two neurons with lamellae of loose myelin. E, F. Electron micrographs from the vestibular epithelium of a 5-day-old bv/bv mouse. E: utricular epitheli-

289

um with a nerve calyce (nc) surrounding the basal part of a hair cell (hc), containing dense-core vesicles, a synaptic body (sb), coated vesicle (cv) and membrane thickenings (arrows). F: afferent nerve ending (a) full of dense-core vesicles making contact with a degenerated hair cell (hc). A cytoplasmic extension of a supporting cell (sc) partially separates the afferent ending from the cell.

290 served than on day 3. Ciliary tufts rarely exhibited the classic pipe organ configuration of the stereocilia that occurs in normal animals. These tufts displayed certain irregularities such as the gradual shortening of the hairs in the same bundle. However, this sort of hair bundle was also observed in the vestibular epithelia of +~by mice. Moreover, the density of the microvilli in these mutants was higher at this stage, compared to that of the epithelial microvilli in +/bv mice. This again, is a sign of immaturity, since it has been observed that these microvilli regress during maturation TM. In the adult stage, the lack of hair bundles was severe both in the cristae and in the maculae utriculi (Figs. 2A, B). Some of the remaining hair bundles had stereocilia of normal length compared to the length of the stereocilia in the hair bundles of +~by mice. Other bv/bv sensory cells displayed abnormal patterns of stereocilia arrangement on their cuticular plates (Figs. 2C, D). The maculae exhibited the widest variation in hair bundle abnormalities, such as cuticular plates with reduced numbers of stereocilia (Fig. 2E), cuticular plates with disordered arrangements of stereocilia, or with disorganized distribution of stereocilia lengths. Between these areas of abnormal hair cell forms, there were areas lacking sensory cells. TEM observations Vestibular ganglion Five days after birth, the vestibular ganglion cells in by~by mice appeared morphologically normal and qualitatively similar to the neurons of +~by vestibular ganglia. At this immature stage, all of the neurons and their processes were still unmyelinated and surrounded by sheaths of Schwann cell cytoplasm. Myelination was beginning in some portions of the intraganglionic fibers. During dissection, the vestibular ganglion appeared visibly smaller in by~by mice than

in +/bv mice, but it was not possible to estimate the lack of neurons quantitatively. In the adult stage, the most striking observation in this mutant was that most of the neurons of the vestibular ganglion were unmyelinated (Figs. 3A, B). In contrast, the intraganglionic vestibular fibers arising from the unmyelinated neurons were normally myelinated. No pathological signs or abnormal organelles were observed, such as the lipofuscin granules frequently observed in Meniere's disease13. The lack of vestibular neurons was evident in adult by~by labyrinths. The unmyelinated neurons of by~by vestibular ganglia contained the usual cytoplasmic organelles with occasional dense-core vesicles (enlarged cisterns of rough endoplasmic reticulum). They were surrounded by a considerable amount of Schwann cell cytoplasm with abundant organelles, often exhibiting many filopodial extensions which intermingled with many evaginations on the cell membrane of the neuronal cytoplasm (Fig. 3C). There were some islands of vestibular ganglion neurons with polygonal shape which were densely packed, as observed in immature stages. In cases where two unmyelinated neurons were densely packed at the junctional region of the two adjacent glial sheaths, the Schwann cytoplasm exhibited discontinuous lamellae of loose myelin with desmosome-like junctions (Fig. 3D). The extremely rare myelinated neurons exhibited an ensheathment composed of multiple layers of both compact and loose myelin. Vestibular receptors Five days after birth, the sensory epithelium of the cristae ampullares and utricle in the by~by mice was lacking sensory cells compared to the receptors of +~by heterozygotes (controls). The remaining sensory cells in by~by mice were poorly differentiated but their morphology was identifiable and two types of hair cells (type I and II) could be distinguished according to known classification criteria. The type II

Fig. 4. A, B. 5 day-oldbv/bv mice. A: rounded cell of the sensoryepithelium with its afferent nerve endings (arrows). B: detail of afferent nerve endings (a) making contact with a rounded cell. C-G. Adult bv/bv mice. C: this afferent nerve ending contacting a type II hair cell (hc) exhibits a post-synapticthickening and contains multiple dense-core vesicles. At the apposition of the afferent dendrites there is a synaptic body and a cistern (arrow). D: a complex of synaptic bodies present in a type I hair cell. E: synaptic contacts be-

291

tween a nerve calyce (nc) and a type I hair cell (hc). Membrane thickenings (arrows) localized opposite each synaptic body (sb). Coated vesicle (cv). F: sensory epithelium of a crista: the basal part of this hair cell exhibits an overabundance of vesiculated afferent nerve endings (arrows). G: multiple efferent nerve endings (e) in the basal region of the sensory epithelium of a crista.

292 hair cells exhibited vesiculated afferent dendrites containing clear and dense-core vesicles of varying size, with very rare synaptic bodies at the synaptic junction. Some extremely rare nerve calyces were seen growing around the type I hair cells exclusively in the utricular epithelium (Fig. 3E). At this stage, they were not yet found in the epithelium of the cristae. In contrast, utricular epithelium of +/bv mice contained several completely developed afferent calyces. The efferent system of by~by mice was present; the efferent nerve endings had established axo-dendritic synapses with the afferent dendrites, but the efferent nerve endings were rarely in contact with the sensory cells. Other cells were small and rounded with a scarcity of organelles in the cytoplasm and poorly developed with respect to the nucleus (Fig. 4A). Their innervation was poor (Fig. 4B) and afferents were mostly totally absent. In the latter case, they were surrounded by hypertrophied cytoplasm of the supporting cells. Slender processes of the supporting cells surrounded the afferent dendrites, separating them from the rounded cells (Fig. 3F), which protruded from the surface of the sensory epithelium and were probably eventually expelled into the endolymphatic space. These were probably the cells that were seen, with SEM, to be bulging toward the endolymph. Moreover, a striking feature was the presence of multiple degenerating cells whose cytoplasm contained clear vacuoles and/ or dense bodies. Numerous cytoplasmic remnants (broken pieces of membranes and mitochondria) were frequently observed in the endotymph. Phagocytosis of degenerating cells by the supporting cells never occurred. The surface of the sensory epithelium was covered only by the hypertrophied supporting cells whose cytoplasm was full of vacuoles which contained amorphous material. Ten days after birth, few morphological differences from the preceding stage were observed. Numerous round cells were still to be seen migrating toward the surface of the epithelium and there were fewer degenerated cells than in the preceding stage. Among the remaining cells, the two types (I and II) could be distinguished morphologically. Several rare type I cells with fully-developed calyces were present. Thirty-five and 90 days after birth, the abnormalities were even more accentuated. The lack of sensory

cells was so severe that the epithelium was occasionally limited to a thin layer of supporting cells. No rounded cells and no degenerating cells were distinguished at these stages. In the utricles, a few scattered hair cells remained. Some of them were deformed, but in general they were morphologically differentiated into type I and type II. Cell-counts on semithin and ultrathin sections showed that type II cells were more numerous than the calyciform type I cells, whereas the latter are more numerous in the epithelia of normal mice 11.15. A few afferent and efferent nerve endings had contacted the two types of cells. In the type II hair cells, presynaptically-located synaptic bodies were present opposite to the vesiculated afferent endings (Fig. 4C), and in the type I hair cells, numerous synaptic bodies, which were often grouped together and immature in appearance (Fig. 4D), persisted adjacent to the hair cell membrane at sites of synaptic specialization. In bv/bv mice there was a striking difference between the utricle and the cristae: in the utricular epithelium, very few hair cells survived, whereas a substantial number of these cells, grouped in isolated islands, were present in the cristae. These surviving hair cells had abundant afferent nerve terminals. The type II cells displayed numerous vesiculated afferent nerve endings; in several cases, these latter nerve endings were extremely numerous, as many as 13 or 14, and they were clustered around the basal part of the cell, whose profile, because of its flask form, resembled a type I cell (Fig. 4F). This abnormal overabundance of afferent nerve endings contacting cells with a flask form and classified as type II because of the pattern of the innervation raises the question of the exact type (II or I) of these cells. Certain type I cells both in the cristae and in the utricles had fully developed calyces. On others, the calyces were incomplete. In cases where the calyces completely surrounded the cells, the apposed membranes of the hair cell and the calyx were not thickened as described in the normal adult by Favre and Sans10 and displayed only a few focal synaptic specializations with a thickening of the two apposed membranes (Fig. 4E). In the cristae, the efferent system persisted, essentially localized in the basal region of the epithelium (Fig. 4G) and showed no clear signs of degeneration. The efferent nerve endings contacted the cells and the afferent dendrites. Certain of these efferents were isolated

293 and surrounded by cytoplasmic digitations of supporting cells. The afferent system occasionally displayed signs of degeneration (an accumulation of organelles, whorl-like bodies,, dense particles) at one month, but never at 3 months. DISCUSSION

Vestibular ganglion During the first postnatal week, the vestibular ganglion in by~by mice displays no pathological or degenerative signs and is morphologically normal and identical to that of the +~by mice. No signs of degeneration were also observed in the spiral ganglion of these same mutants, whereas abnormal cytoplasmic organelles were observed in the ganglion cells, consisting of vacuoles and fibrillar material16. Abnormal organelles have also been reported in other mouse mutants such as shaker-126 and deafness 20. In the adult stage, the vestibular ganglion still displays features of the immature stage, i.e. the dense clustering of some cells and the presence of several filopodial extensions along the soma as well as the lack of a myelin sheath, which are morphological characteristics of very young stages described in the normal postnatal development of the spiral ganglion cells 25. But the most striking finding in this bv/bv mutant is the lack of a myelin sheath around the perikarya. In most of the mammalian species, except man, most normal vestibular ganglion cells are surrounded by sheaths of compact or loose myelin3,12,22,28. This lack of perikaryal myelin sheath probably has important functional consequences for the electrical properties of the cell and the conduction velocity of nerve impulses 30 but until now, the vestibular system of these by~by mutants has never been studied from an electrophysiological point of view, unlike the cochlear system 1,2. There were still no signs of degeneration in bv/bv vestibular ganglion neurons in the adult stage. The cytoplasm and its organelles had a normal appearance, similar to descriptions in the literature12,22, 28. The observed loss of neurons was difficult to estimate in our case but it was visibly less significant than that of the spiral ganglion in the same mutant, in which a severe loss was observed at birth and was estimated to be about 20% on the 10th postnatal day 16. According to Deol 7, the ganglion grows progressively thinner with age as a result of a cell loss starting from an

almost normal population at birth. In our case, since no signs of degeneration were observed, we consider that the mutation affects the proliferation of ganglion cells, probably at an early stage of embryogenesis.

Vestibular receptors The observations with scanning and transmission electron microscopy demonstrate that the morphological abnormalities of the sensory epithelia in the bv/bv mice are very early, since they are visible in the first postnatal stages, which implies that the pathological changes start prenatally, while in other degenerative-type mutants, the inner ear appears to be morphologically normal before undergoing degenerative processesS, 7. In the bv/bv mutant, hair cells are already missing by the 3rd postnatal day, whereas in normal mice all the sensory cells are present at birth, since the terminal mitoses of most of the hair cells occur between the 14th and the 18th day of gestation 23. Moreover, the degeneration affects all the vestibular receptors in this mutant. However, in the adult stage, a greater number of cells persists in the cristae, compared to the utricles. Since the maturation of the utricles is earlier than that of the cristae 24, one can assume that there is a certain relationship between the impact of the mutation and the staggered timing of maturation in these receptors. In other mutants, the degeneration is confined to one of the two types of vestibular receptors. In varitint-waddler mice, the utricular maculae are not affected, whereas in jerker mice the cristae escape degeneration 4. Some type I and type 119,31 sensory cells, seem to be unaffected by degeneration and persist until the adult stage. During postnatal development, the calyces and the hair bundles continue to grow, but the maturation of the receptors is delayed in comparison to that of normal mouse strains 19. Similar observations have been reported in the deaf shaker-1 mouse 26 which express a combination of retarded development and a premature degeneration. Examination by SEM shows that almost all the hair bundles display morphological abnormalities and stereociliary disorganization, whereas under TEM the rare remaining sensory cells have a morphologically normal appearance, and are contacted by some nerve endings. In the adult stage, the two systems of afferent and efferent innervation are present in the vestibular receptors. This innervation seems more abun-

294 dant in the cristae, where there are numerous afferent dendrites and where numerous efferent nerve endings are located in the basal part of the epithelium and often contact nothing in areas where the cells are missing. This mutation does not appear to affect the efferent system, nor the cochlear efferent system whose outer hair cells escape degeneration, i.e. those which are abundantly innervated by the efferent system16. It is still not known whether these synapses are functional. In the adult stage, the two types of sensory cells coexist in the cristae as well as in the maculae utriculi. The occurence of the type II sensory cells is numerically greater than the type I cells in homozygotes. However, it is known that in other normal animals calyciform cells (type I) predominate and constitute about 60% of the total population of the sensory cells~l, 15. In this mutant, certain sensory cells classified as type II whose bases are surrounded by a multitude of afferent nerve endings and whose morphological profile resembles a classic type I profile (a flask with a thin neck), are probably immature type I cells whose calyces never formed. Favre and Sans11 have hypothesized that during ontogenesis the innervation of type I cells passes through an intermediate stage which corresponds to type II cell innervation. Hence, the immature innervation observed in this mutant leads us to suggest, in accordance with other data 27, that the two types of hair cells are genetically programmed and not a result of the influence of the nerve terminals on the differentiating sensory cells. Other obvious signs of immaturity should be noted in the adult stage, i.e. the presence of incomplete calyces, the presence of calyces contacting type I cells in an immature fashion with only a limited number of focal membrane thickenings~0, the frequent presence of synaptic bodies in type I cells 10, and finally, the presence of vesicles in the afferents contacting the type II cells TM. Because abnormalities are observed very early in the vestibular receptors as well as in the cochlea, it can be assumed that the degenerative processes start prenatally. Therefore, the Bronx waltzer mutation should be classified as 'morphogenetic' rather than 'degenerative'. There is an abnormal cytodifferentiation from the very beginning in both the sensory epithelium and the vestibular ganglion neurons and then a delayed or absent maturation. Some degenerative

processes are also involved after some stages of abnormal development. In addition, there are significant signs of cell degeneration in the vestibular receptors for at least one to two weeks after birth. Thus, there is a progressive loss of cells during development. The expulsion of these degenerating cells into the endolymph is a process that has also been reported in relation to the effects of ototoxic drugs ~7 and a subsequent separation of the nerve endings from the degenerating sensory cells by processes of the supporting cells has also been reported in the inner ear following irradiation with X-rays3L The description and interpretation of the numerous morphological anomalies in the two types of sensory cells should not lead us to neglect the fact that the overall loss of sensory cells is substantial in these vestibular epithelia. To date, no morphophysiological relationships have been demonstrated in the vestibular system of these by~by mice which would indicate whether the remaining cells that escape degeneration are functional. In the cochlear system, the +~by mutation affects almost all the inner hair cells, while the outer hair cells, predominantly innervated by efferents, escape pathological modifications 16. In the vestibular system, the mutation equally affects type I and type II hair cells, which display abundant afferent innervation under normal conditions and the remaining cells are always contacted by efferents. These visualizations with SEM and the ultrastructural findings demonstrate that in addition to an early degeneration, there is also a delayed development of the remaining sensory cells. It appears that the mutation blocks cellular development in the ganglion as well as in the receptors, since the cells are unable to acquire adult features. There is certain analogy here between this delay in maturation and the phenomena observed during experimental hypothyroidism in the rat (results to be published). The pathological anomalies of these mutants occur simultaneously in the vestibular receptors and in the ganglia, though less severely in the ganglia. Additional investigations of this +~by mutation affecting the inner ear during embryogenesis in particular an in vitro study of foetal inner ear and stato-acoustic ganglion 2~ should help to explain the role of neurosensory interactions between developmental processes in the vestibular receptors and the ganglion.

295 ACKNOWLEDGEMENTS This w o r k was s u p p o r t e d by G r a n t s f r o m I N S E R M ( C R L 826006 and R P C 135033).

(University College of London) for his courtesy in providing the mice. They also gratefully acknowledge J. Roudil and P. Sibleyras for photographic reproductions and A. Bara for typing the manuscript.

T h e a u t h o r s are i n d e b t e d to P r o f e s s o r M. S. D e o l

REFERENCES 1 Bock, G. R. and Yates, G. K., Cochlear electrophysiology in the Bronx Waltzer mutant mouse, J. Physiol. (Lond.), 332 (1982) 20-21. 2 Bock, G. R., Yates, G. K. and Deol, M. S., Cochlear potentials in the Bronx Waltzer mutant mouse, Neurosci. Len., 34 (1982) 19-25. 3 Chat, M. and Sans, A., Multipolar neurons in the cat vestibular ganglion, Neuroscience, 4 (1979) 651-657. 4 Deol, M. S., The anomalies of the labyrinth of the mutants varitint-waddler, shaker-2 and jerker in the mouse, J. Genet., 52 (1954) 562. 5 Deol, M. S., Inherited diseases of the inner ear in man in the light of studies on the mouse, J. Med. Genet., 5 (1968) 137-158. 6 Deol, M. S., Genetic malformation of the inner ear in the mouse and in man. In R. L. Gorlin (Ed.), Morphogenesis and Malformation of the Ear, A. R. Liss, New York, 1980, pp. 243-260. 7 Deol, M. S., The inner ear in Bronx Waltzer mice, Acta Otolaryngol. (Stockh.), 92 (1981) 331-336. 8 Deol, M. S. and Gluecksohn-Waelsh, S., The role of inner hair cells in hearing, Nature (Lond.), 278 (1979) 250-252. 9 Engstr6m, H., The innervation of the vestibular sensory cells, Acta Otolaryngol. (Stockh.), Suppl., 163 (1961) 30-41. 10 Favre, D. and Sans, A., Morphological changes in the afferent vestibular hair cells synapses during the postnatal development of the cat, J. Neurocytol., 8 (1979) 765-775. 11 Favre, D. and Sans, A., Embryonic and postnatal development of afferent innervation in cat vestibular receptors, A cta Otolaryngol. (Stockh. ) , 87 (1979) 97-107. 12 Fermin, C. D. and Igarashi, M., Vestibular ganglion of the Squirrel monkey, Ann. Otol. Rhinol. Laryngol. (St. Louis), 91 (1982) 44-52. 13 Galic, M. and Helms, J., Elektronenmikroskopische Untersuchungen fiber die H~iufigkeit, Verteilung und Struktur des Lipofuscins im Ganglion Vestibuli bei Morbus Meniere. In M. Portmann and J. M. Aran (Eds.), Inner Ear Biology, Vol. 68, INSERM, 1977, pp. 243-262. 14 Jones, D. G., Eslami, H., An ultrastructural study of the development of afferent and efferent synapses on outer hair cells of the guinea pig organ of Corti, Cell Tiss. Res., 231 (1983) 533-549. 15 Lindeman, H. H., Reith, A. and Winther, F. O., The distribution of type I and type II cells in the cristae ampullares of the guinea pig. A morphogenetic investigation, Acta Otolaryngol., 92 (1981) 315-321. 16 Lenoir, M. and Pujol, R., Age-related structural investigation of the Bronx Waltzer mutant mouse cochlea: scanning transmission electron microscopy, Hearing Res., 13 (1984) 123-134. 17 Lundquist, P. G. and Wers~ill, J., Scanning electron micro-

scope studies on the inner ear: a discussion on labyrinthine pathology, Biomed. Res., 2 (Suppl.) (1981) 379-389. 18 Mbi~ne, J. P., Favre, D. and Sans, A., The pattern of ciliary development in fetal mouse vestibular receptors: a qualitative and quantitative SEM study, Anat. Embryol., in press. 19 Nordemar, H., Postnatal development of the vestibular sensory epithelium in the mouse, Acta Oto-laryngol., 96 (1983) 447-456. 20 Pujol, R., Shnerson, A., Lenoir, M. and Deol, M. S., Early degeneration of sensory and ganglion cells in the inner ear of mice with uncomplicated genetic deafness (dn), Hearing Res., 12 (1983) 57-63. 21 Raymond, J., Desmadryl, G. and Sans, A., D6veloppement en culture organotypique de l'6pith61ium sensoriel vestibulaire chez l'embryon de Souris. Etude morphologique et radioautographique, C.R. Acad. Sci. (Paris), 294, S6rie III (1982) 639-644. 22 Rosenbluth, J. and Palay, S. L., The fine structure of nerve cell bodies and their myelin sheaths in the eighth nerve ganglion of the goldfish, J. Biophys. Biochem. Cytol., 9 (1961) 853-877. 23 Ruben, R. J., Development of the inner ear of the mouse: a radioautographic study of terminal mitoses, Acta Oto-laryngol., Suppl. 220 (1967) 1-44. 24 Sans, A. and Chat, M., Analysis of temporal and spatial patterns of rat vestibular hair cell differentiation by tritiated thymidine radioautography, J. comp. Neurol., 206 (1982) 1-8. 25 Schwartz, A. M., Parakkal, M. and Gulley, R. L., Postnatal development of spiral ganglion cells in the rat, Amer. J. Anat., 167 (1983) 33-41. 26 Shnerson, A., Lenoir, M., Van de Water, T. R. and Pujol, R., The pattern of sensory-neural degeneration in the cochlea of deaf shaker-1 mouse: ultrastructural observations, Develop. Brain Res., 9 (1983) 305-315. 27 Sobin, A., Wers~.ll, J., A morphological study on vestibular sensory epithelia in a strain of the waltzing guinea pig, Acta Oto-laryngol., Suppl. 396 (1983) 1-32. 28 Spassova, I., Fine structure of the neurons and synapses of the vestibular ganglion of the cat, J. Hirnforsch., 23 (1982) 657-669. 29 Steel, K. P. and Bock, G. R., Hereditary inner-ear abnormalities in animals. Relationships with human abnormalities, Arch. Otolaryngol. (Chic.), 109 (1983) 22-29. 30 Ylikoski, J., The fine structure of the sheaths of vestibular ganglion cells in the rat, monkey and man, Acta Oto-laryngol., 95 (1983) 486-493. 31 Wers~ill, J., Studies on the structures and innervation of the sensory epithelium of the cristae ampullaris of the guinea pig. A light and electron microscopic investigation, Acta Oto-laryngol., 126 (1956) 1-85. 32 Winther, F. O., X-ray irradiation of the inner ear of the guinea pig, Acta Oto-laryngol., 69 (1970) 307-319.