Innervation densities of inner and outer hair cells of the human organ of Corti. Evidence for auditory neural degeneration in a case of Usher's syndrome.

The innervation densities, i.e. the number of afferent endings per inner and outer hair cell, number of synaptic specializations per inner hair cell and the number of efferent endings per outer hair cell, were evaluated by serial section electron microscopy in the normal human inner ear and a case of Usher's syndrome (retinitis pigmentosa and sensorineural hearing loss). There was a significant decrease in the innervation density in Usher's syndrome as compared to normative controls, which could not be explained by the decrease in density of spiral ganglion cells. This was interpreted as suggesting a decrease in the number of neural processes and endings within the dendritic arborization of remaining spiral ganglion cells in this disorder.

Developmental stability of the tonotopic organization of the chick's basilar papilla.

In the past, the damage patterns produced on the papilla of the chick by loud pure tones of different frequencies have been used to study the development of sound analysis by the hair cells of the basilar papilla. One conclusion from these data was that the best response frequency of individual hair cells changes substantially with age. However, this method has been criticized as unreliable. Now experiments have been performed in which single characterized nerve fibers were stained with horseradish peroxidase to permit the unequivocal localization of specific frequency responses in the papillae of chicks of different ages (2 and 21 days after hatching). There was no statistically significant change in the tonotopic organization of the papilla between the two groups of animals.

Incidence of reciprocal synapses on outer hair cells of the human organ of Corti.

The incidence of reciprocal synapses at the base of outer hair cells of the human organ of Corti was studied by electron-microscopic serial section reconstruction. Reciprocal synapses were found in each of six ears from five individuals. A total of 1,386 serial sections and the neural poles of 25 outer hair cells in two individuals were examined. Fourteen of the 25 outer hair cells (56%) possessed reciprocal synapses. The incidence of reciprocal synapses increased from the first to the third rows of outer hair cells, as determined by the percentage of outer hair cells and afferent nerve terminals of each row demonstrating reciprocal synapses and the average number of reciprocal synapses per hair cell in each row. Hair cells with reciprocal synapses had more afferent and fewer efferent terminals per hair cell than hair cells without reciprocal synapses.

Serial section reconstruction of the neural poles of hair cells in the human organ of Corti. I. Inner hair cells.

Study of the anatomy of the cochlea, and in particular the morphology of synaptic relationships between hair cells and cochlear neurons, is essential for elucidation of the mechanisms of transduction of mechanical acoustic signals into electrical neural events. Because considerable gaps remain in our understanding of the microscopic anatomy of these synapses, particularly in the human, a reconstruction of neural pole of inner hair cells of the human organ of Corti was performed. The data are based on 526 serial sections from the basal turn (10 mm region) and 356 serial sections from the middle turn (26 mm region). This provided complete data on 3 and partial data on 5 inner hair cells. Afferent terminals on inner hair cells were variable in size, ranging 1 to 20 micrometers in diameter. Branching of large fibers to produce multiple terminals innervating from 1 to 3 inner hair cells was common. Each inner hair cell received approximately 6 to 8 different nerve terminals. In addition, each terminal possessed a variable number of synaptic contacts. Junctional membrane specialization consisted of synapses, desmosomes, coated vesicles and arrays of microtubules and membrane cisternae. Specialization at synapses consisted of asymmetrical membrane thickening. At inner hair cells the postsynaptic membrane was thicker than the presynaptic membrane. Eighty-three percent of synapses had presynaptic bodies. Vesiculated efferent terminals synapsed on afferent fibers at the base of inner hair cells, but never directly on the inner hair cell. These anatomical data demonstrate distinct differences between the human and animal inner ear, which are important in the interpretation of neurophysiological data in animals and the formulation of hypotheses that involve assumptions crossing species.

Auditory stimulation alters the pattern of 2-deoxyglucose uptake in the inner ear.

The 2-deoxy-D-glucose (2-DG) autoradiographic technique was adapted for application to the inner ear. The uptake of [14C]-DG during silence was compared with that observed during exposure to wide band noise (WBN) or pure tones at an intensity level of 85 db SPL. In silence, the highest levels of 2-DG uptake were observed in the spiral ligament, spiral prominence and stria vascularis, with approximately equal levels of uptake in each structure. The high levels of 2-DG uptake observed in the ligament and prominence are surprising, and suggest a more active role for these structures in cochlear function than has previously been suspected. Levels of uptake in the organ of Corti, spiral ganglion and VIIIth nerve were much lower, although well above background. During exposure to WBN, 2-DG uptake increased markedly in the VIIIth nerve, and spiral ganglion throughout the cochlea, and in the organ of Corti in the lower basal turn. 2-DG uptake did not change significantly in the spiral ligament or stria vascularis. During pure tone exposure, increased 2-DG uptake was noted in localized regions of the VIIIth nerve and spiral ganglion.

Age-related changes in the C57BL/6J mouse cochlea. II. Ultrastructural findings.

The C57BL/6J mouse organ of Corti was studied using transmission electron microscopy. The basal coil of the cochlea was examined in mice 1-50 days of age. At birth the cochlea was very immature but both types of hair cells were innervated. Inner hair cells (IHC) were connected to afferent and efferent processes, and efferent endings synapsed on IHC afferents. Outer hair cells (OHC) were innervated exclusively by afferents, which made well-defined synapses with the cell. Maturation at IHC took place rapidly and was essentially over by about 12 days of age. Changes were largely restricted to an increase in presynaptic specializations opposite afferents and a decrease in postsynaptic specializations (cisterns) opposite efferents. Ontogeny at OHC took place over a longer, 2.5 week, period. Afferent synapses lost their presynaptic specializations (synaptic bodies), and then efferent fibers arrived below OHC. The efferents subsequently made temporary axo-dendritic synapses with the afferents before replacing most of them at OHC. The first synapses between efferent endings and OHC were seen at 9 days of age, but it was not until about 20 days of age that mature synapses were seen. Some evidence of hair cell degeneration was seen in 30- and 50-day-old mice. The results are discussed in terms of sensory hair cell differentiation, the disappearance of OHC synaptic bodies, and age-related changes in auditory system function.

Efferent components of the auditory system.

The origins and terminations of the olivocochlear bundle, which provides an efferent innervation to the organ of Corti, are described on the basis of experiments using axonal transport of tracer substances and light microscopy in the cat. The cells of origin were labeled by the retrograde tracer horseradish peroxidase which was injected unilaterally into the cochlea. Labeled cells in the superior olivary complex could be dichotomized according to their location (lateral or medial), their size (small or large), and their preferred side of projection (uncrossed or crossed). All labeled olivocochlear neurons exhibited a positive reaction for acetylcholinesterase. To determine the cochlear projections of the neurons, injections of a radioactive amino acid were made into either the lateral or medial olivocochlear cell group. After allowing time for synthesis and axonal transport of radio-labeled protein to reach synaptic endings in the cochleas, the tissue sections of these specimens were processed for autoradiography. The results indicate that lateral olivocochlear neurons project to the regional beneath the inner hair cells of both sides, whereas medial olivocochlear neurons project to the region beneath the outer hair cells of both sides. These findings are in substantial accord with previous experimental work but suggest that the organ of Corti receives a dual efferent innervation which is organized according to the location and morphology of its cells of origin. Accordingly, it is proposed that the two efferent components of the cochlear innervation described here be referred to as the lateral and medial olivocochlear systems, replacing the current designations of crossed and uncrossed olivocochlear bundles, the latter which are demonstrably heterogeneous in their origins and terminations and, probably, also in their functions.

Method for tracing single fibers in the organ of Corti with horseradish peroxidase.

A method based on the axonal transport of horseradish peroxidase (HRP) has been developed to trace single fibers in the organ of Corti. HRP was injected into the auditory nerve or cochlear nucleus of the adult chinchilla, and the entire cochlea was subsequently incubated for reaction product. Of most significance was the Golgi-like labelling of single radial afferents innervating the inner hair cells. Heretofore, radial afferents have been difficult to trace with more conventional staining techniques, especially in adult animals. Outer spiral fibers, presumably afferent, were also labelled in one of the cochleas.

Degeneration patterns in human ears exposed to noise.

Various forms of sensorineural degeneration patterns related to noise exposure are illustrated in six pairs of temporal bones selected from a group of 33 male patients with histories of noise exposure. For the entire group the commonest form of lesion, associated with a 4-kHz dip in the audiogram, was a relatively diffuse degeneration in the second quadrant of the basal turn, in the 9-13 mm area. An advanced form of this lesion had a wide gap of more or less complete sensorineural degeneration affecting the entire second quadrant and displaying various degrees of extension toward the apex and base. The pattern associated with an "abrupt high-tone loss," with more or less complete hair cell and nerve degeneration in both the second and first quadrants and extending to the basal end of the cochlea, was rare. In one case this pattern appeared to have evolved from the first type of lesion as the remaining nerve fibers in the first quadrant had degenerated. The protective effect of the acoustic shadow of the head for the right ear, in shooting from the right shoulder, is demonstrated for the higher frequencies. Two almost identical cases of sharply-circumscribed single areas of degeneration in the first quadrant and one case with two such areas represent the third type of lesion. In one of these cases there was a history of firearm usage. It is postulated that this type of lesion is caused by impulse noise. In most of the material the degeneration pattern differed markedly from the diffuse degeneration seen with presbycusis. Degeneration patterns with knife-sharp transitions between completely degenerated and apparently undamaged areas appear to be characteristic of noise-induced injuries.

Some unresolved problems concerning the cochlear nerve.

Three types of afferent fibers innervate the hair cells of the organ of Corti: 1) specific radial fibers which establish contacts with a very few neighboring internal hair cells; 2) spiral fibers, each one of which establishes contact with a number of external hair cells distributed throughout long segments of the cochlea; and 3) unspecific radial fibers which are collaterals arising radially at irregular intervals from fibers of the ganglionic spiral bundles and which establish contact with internal hair cells. The existence of spiral ganglionic bundles of fibers oriented apicalward has long been described, and the fact that a number of ganglionic spiral fibers give off radial collaterals to innervate internal hair cells was illustrated by Cajal and by Lorente de Nó. However, those structural details are not mentioned in the modern literature. In the ventral nucleus there are neurons with efferent axons which join the trapezoid body and cells with short axons ramified within the ventral nucleus itself. Two types of cells with efferent axons are illustrated and described, the spherical or bushy cell and the basket cell; and it is shown that branches of division of the two types of efferent axons form association tracts which end in the tuberculum acusticum. Also, the fact is illustrated that fibers having their cells of origin in that tuberculum form association paths which end in the ventral nucleus by means of extensive ramifications which form numerous synaptic endings. The dendritic and fibrillar plexuses in the ventral nucleus are described, an analysis is made of the relationships between the two plexuses and of the synaptic junctions that mediate transmission of nerve impulses. The synaptic junctions belong to a considerable number of types and in all illustrations the important fact repeatedly appears that one and the same fiber may form synaptic endings of widely different sizes and shapes located either on the same neuron or, more frequently, on different neurons. The intimate structure of each type of synaptic ending cannot be revealed by light microscopy, but only light microscopy can reveal to which kind of fiber the synaptic endings do belong. The presentation is concluded with a brief and preliminary discussion of physiological corollaries of certain general features of the anatomy of the acoustic system.

Histopathological observations of presbycusis.

Temporal bone histopathology of 17 aged patients who had spontaneous and gradually progressive bilateral sensorineural hearing losses associated with aging was studied. Six cases in the present material showed the gradually sloping audiometric curve; nine cases, abrupt high tone hearing loss; and two cases, the flat audiometric curve. The most prominent histopathological change in the inner ear was a decrease in the population of the spiral ganglion cells. However, diffuse senile atrophy was also often seen in the organ of Corti and the stria vascularis. A positive correlation between the degree of artheriosclerosis and the degree of sensorineural degeneration in the cochlea was not obtained in the present cases. Also, the correlation was not found to be consistent between the type of the audiometric curve and the localization of lesions in the sensory, the neural or the vascular elements in the cochlea. Our observations show that a certain type of audiometric curve does not necessarily indicate a lesion in a specific cochlear element.

A study of cochlear innervation patterns in cats and rats with the Golgi method and Nomarkski Optics.

Cochlear innervation patterns were studied in infant cats and rats with the rapid Golgi method. Examination of thick serial sections and surface preparations with the differential interference contrast microscope (Nomarski optics) allowed direct visualization of individually impregnated spiral ganglion cells, complete with their peripheral processes and endings in the organ of Corti. Individually impregnated efferent fibers could be recognized as heavily varicose axons that project radially to endings beneath inner and outer hair cells after taking a tangential course in the intraganglionic spiral bundle. It was often possible to visualize unimpregnated hair cells in contact with the impregnated endings of both types of fibers. There are at least two types of spiral ganglion cells in the cochlea of the infant cat and rat. One type innervates only inner hair cells by means of radial fibers. These ganglion cells constitute the overwhelming majority of ganglion cells impregnated in our preparations, and each cell typically innervates two inner hair cells. Hence, these ganglion cells establish nearly "point-to-point" connections between the auditory nerve and the organ of Corti. The other type of ganglion cell innervates outer hair cells by means of long spiral fibers; each cell typically innervates many outer hair cells through the numerous angular enlargements and short end branches of its spiral fiber. In addition, a few of these spiral fibers also send branches to inner hair cells by means of short collaterals; it remains to be seen if such fibers also occur in mature cochleas. Efferent fibers have been traced to inner and outer hair cell regions. The simplest pattern is formed by fine beaded axons with only a few branches ending mainly beneath inner hair cells. More complex patterns are formed by larger axons with many branches ending beneath inner or outer hair cells. Many efferent fibers send branches to both inner and outer hair cells. Electrophysiological studies so far have not demonstrated different populations of units that clearly correspond to the spiral and radial fibers. Therefore, the physilogical differences between inner and outer hair cell innervation remain undefined.

A study of neurone activity in the spiral ganglion of the cat's basal turn.

A small region of the spiral ganglion in the cat was surgically exposed through the round window. Metal microelectrodes were used to record extracellularly the electrical activity of single spiral ganglion cells. The response characteristics of the cells seemed to be, in general, similar to those seen for auditory-nerve fibres recorded with micropipets in the internal auditory meatus. Data are presented on spontaneous activity, tuning curves, responses to clicks, continuous tones, tone bursts and noise bursts. The relation between frequency selectivity of units and location along the basilar membrane is discussed. -Some units differed in behaviour from auditory-nerve fibres with respect to dead times in interspike-interval histograms and shapes of poststimulus-time histograms of responses to tone bursts and noise bursts. The significance of these deviations is unknown.

Effect of intracochlear aminooxyacetic acid on cochlear potentials and endolymph composition.

Summary--Aminooxyacetic acid (AOAA; 1 MM) when perfused through the scala tympani of guinea pigs produced a biphasic reduction in the endocochlear potential, together with a shift in the input-output functions of the cochlear microphonics and the compound action potential of the cochlear nerve. Since the ionic (Na+, K+) content of the scala media was not alerted, it appears that such changes did not underlie the reduction recorded in the endocochlear potential. The negative direct current potential recorded in the organ of Corti did not appear to be affected by the drug. Attempts to antagonize the effects of AOAA with pyridoxine failed.

A quantitative analysis of the afferent innervation of the organ of corti in guinea pig.

A quantitative analysis of the afferent innervation of the organ of Corti was made on normal and vestibular nerve-sectioned guinea pigs. Section of the vestibular nerve at the internal auditory meatus provided an efficient means of eliminating the efferent innervation to the cochlea without significant loss of afferent fibres. Nerve counts on normal and de-efferented animals revealed that about 10-15 % of the cochlear afferent innervation supplies the outer hair cells. The remaining 85-90% of afferent fibres innervate the inner hair cells. As in cats, all tunnel spiral bundle fibres and upper tunnel crossing fibres were efferent to outer hair cells. Since unmyelinated fibres in the osseous spiral bundle were not counted, quantitative analysis of the efferent innervation to inner hair cells could not be made. However, a significant loss of myelinated fibres in the osseous spiral lamina after vestibular nerve section confirms that many myelinated efferent fibres are present in this region.