Variations in shape-sensitive restriction points mirror differences in the regeneration capacities of avian and mammalian ears.

When inner ear hair cells die, humans and other mammals experience permanent hearing and balance deficits, but non-mammalian vertebrates quickly recover these senses after epithelial supporting cells give rise to replacement hair cells. A postnatal decline in cellular plasticity appears to limit regeneration in mammalian balance organs, where declining proliferation responses are correlated with decreased spreading of supporting cells on artificial and native substrates. By culturing balance epithelia on substrates that differed in flexibility, we assessed spreading effects independent of age, showing a strong correlation between shape change and supporting cell proliferation. Then we made excision wounds in utricles cultured from young and old chickens and mice and compared quantified levels of spreading and proliferation. In utricles from young mice, and both young and old chickens, wounds re-epithelialized in <24 hours, while those in utricles from mature mice took three times longer. More cells changed shape in the fastest healing wounds, which accounted for some differences in the levels of proliferation, but inter-species and age-related differences in shape-sensitive restriction points, i.e., the cellular thresholds for shape changes that promote S-phase, were evident and may be particularly influential in the responses to hair cell losses in vivo.

Mefloquine damage vestibular hair cells in organotypic cultures.

Mefloquine is an effective and widely used anti-malarial drug; however, some clinical reports suggest that it can cause dizziness, balance, and vestibular disturbances. To determine if mefloquine might be toxic to the vestibular system, we applied mefloquine to organotypic cultures of the macula of the utricle from postnatal day 3 rats. The macula of the utricle was micro-dissected out as a flat surface preparation and cultured with 10, 50, 100, or 200 µM mefloquine for 24 h. Specimens were stained with TRITC-conjugated phalloidin to label the actin in hair cell stereocilia and TO-PRO-3 to visualize cell nuclei. Some utricles were also labeled with fluorogenic caspase-3, -8, or -9 indicators to evaluate the mechanism of programmed cell death. Mefloquine treatment caused a dose-dependent loss of utricular hair cells. Treatment with 10 µM caused a slight reduction, 50 µM caused a significant reduction, and 200 µM destroyed nearly all the hair cells. Hair cell nuclei in mefloquine-treated utricles were condensed and fragmented, morphological features of apoptosis. Mefloquine-treated utricles were positive for the extrinsic initiator caspase-8 and intrinsic initiator caspase-9 and downstream executioner caspase-3. These results indicate that mefloquine can induce significant hair cell degeneration in the postnatal rat utricle and that mefloquine-induced hair cell death is initiated by both caspase-8 and caspase-9.

Damage and functional recovery of the Atlantic cod (Gadus morhua) inner ear hair cells following local injection of gentamicin.

This study addresses the ultrastructural and functional damage and subsequent recovery of the inner ear in the Atlantic cod following intrasaccular gentamicin injection. Inner ear damage was assessed using SEM and measurements of AEP following 250-Hz pure-tone stimuli. Data from gentamicin-treated fish were compared with control (no injection) and sham (injection of saline) fish. Control fish had normal response thresholds associated with well-developed hair cell bundles in their macula sacculi. Sham fish had higher response thresholds compared with control fish during the first week post-intervention, but response thresholds were subsequently normal. Treated fish displayed significant inner ear damage associated with an increased average AEP threshold on the third day following treatment. Thereafter, inner ear tissue displayed signs of progressive regeneration until it was comparable to controls from the 14th day. Response thresholds were similar to those of control fish from the 17th day following treatment. These observations suggest that the macula sacculi of Atlantic cod can regenerate towards a near-complete functional and ultrastructural recovery within 17-21 days following ototoxic gentamicin treatment.

Spontaneous hair-cell renewal following gentamicin exposure in postnatal rat utricular explants.

We have established an in vitro model of long-time culture of 4-day-old rat utricular maculae to study aminoglycoside-induced vestibular hair-cell renewal in the mammalian inner ear. The explanted maculae were cultured for up to 28 days on the surface of a membrane insert system. In an initial series of experiments utricles were exposed to 1 mM of gentamicin for 48 h and then allowed to recover in unsupplemented medium or in medium supplemented with the anti-mitotic drug aphidicolin. In a parallel control series, explants were not exposed to gentamicin. Utricles were harvested at specified time points from the second through the 28th day in vitro. Whole-mount utricles were stained with phalloidin-fluorescein isothiocyanate and their stereociliary bundles visualized and counted. In a second experimental series 2'-bromo-5'deoxyuridine labeling was used to confirm the antimitotic efficacy of aphidicolin. Loss of hair-cell stereociliary bundles was nearly complete 3 days after exposure to gentamicin, with the density of stereociliary bundles only 3-4% of their original density. Renewal of hair-cell bundles was abundant (i.e. 15x increase) in cultures in unsupplemented medium, with a peak of stereociliary bundle renewal reached after 21 days in vitro. A limited amount of hair-cell renewal also occurred in the presence of the anti-mitotic drug, aphidicolin. These results suggest that spontaneous renewal of hair-cell stereociliary bundles following gentamicin damage in utricular explants predominantly follows a pathway that includes mitotic events, but that a small portion of the hair-cell stereociliary bundle renewal does not require mitotic activity.

Gentamicin tympanoclysis: effects on the labyrinthine sensory cells.

OBJECTIVES: The objective of the study was to determine whether a selective vestibular hair cell toxicity with sparing of the cochlear hair cells could be achieved by infusing different concentrations of gentamicin into the middle ears of adult cats. STUDY DESIGN: Prospective experimental animal study treating only the left ear of each cat, the right ear serving as individual control. METHODS: Gentamicin solution at concentrations of either 30 or 3 mg/mL was infused daily into the left middle ear of adult cats until overt ataxia occurred. After 1 month or 6 months, each cat was killed and its temporal bones prepared for optical microscopy. RESULTS: Animals treated with 30 mg/mL gentamicin until ataxic required a median of five daily doses. These animals had clear-cut cochlear basal turn hair cell losses accompanying toxic lesions in the utricle and cristae. In contrast, animals treated with 3 mg/mL gentamicin until ataxic required an average of 19 daily doses. These animals had lesions restricted to the utricle and cristae with sparing of the cochlea hair cells. Animals that failed to develop ataxia manifested neither lesions of the cochlear nor vestibular hair cells. CONCLUSION: Gentamicin tympanoclysis in the cat animal model, using a dilute solution and continued once daily until clinical ataxia occurs, is capable of producing selective vestibular hair cell toxicity while sparing cochlea hair cells.

Study of the gerbil utricular macula following treatment with gentamicin, by use of bromodeoxyuridine and calmodulin immunohistochemical labelling.

Effects of ototoxic drugs on the gerbil vestibular sensory epithelium were probed by use of immunocytochemical labelling with antibodies to both a mitogenic marker (bromodeoxyuridine) and a hair cell specific protein (calmodulin). Nine animals had gentamicin administered once daily for 5 days, as a transtympanic injection into the right middle ear. They additionally were given a daily intraperitoneal injection of bromodeoxyuridine, starting on the same day as the gentamicin injection and continuing until the day of sacrifice. Nine other animals, serving as controls for bromodeoxyuridine incorporation, received only the intraperitoneal injections of bromodeoxyuridine. The inner ears from three gerbils were obtained at 1, 2 or 4 weeks following the last gentamicin injection and utricles from the injected ears were processed for immunohistochemical analysis. In specimens where gentamicin was administered, we found evidence of bromodeoxyuridine incorporation in 17 cells (10 single cells and 7 pairs of cells) in a total of 216 sections taken from the central regions of the 9 utricles. However, in control specimens, no bromodeoxyuridine labelling was found in any cells of the 216 sections examined. Of 10 single cells labelled with bromodeoxyuridine, two cells in the hair cell layer were labelled with antibodies against calmodulin. One had a faint labelling in the nucleus and the other in the stereocilia, but not in the cell bodies. Of 7 pairs of cells, two pairs with nuclei localized in the hair cell layer had faint labelling for calmodulin in the nuclei, but no labelling in any other part of the cell. The other 13 cells labelled with antibodies to bromodeoxyuridine were not labelled with antibodies to calmodulin. Our results suggest that the bromodeoxyuridine-labelled cells could not be positively identified as hair cells based on immunohistochemical labelling for calmodulin.

Effect of cisplatin on the basement membrane anionic sites in the ampulla, macula, and stria vascularis of guinea pigs.

Our objective was to compare changes in the basement membrane anionic sites (BMAs) in the ampulla, macula, and stria vascularis following the infusion of cisplatin (CDDP). After CDDP was administered to anesthetized Hartley guinea pigs, the bony labyrinth was immersed in a solution of polyethyleneimine (PEI). The size and distribution of PEI particles associated with BMAs in the stria vascularis and in the dark cell and sensory cell areas of the vestibular labyrinth were determined by electron microscopy. A significant reduction in the number and size of PEI particles was observed on CDDP-treated strial vessels. The number and size of PEI particles on the basement membranes of the vestibular labyrinth did not differ from those in the control. Our findings suggest that the BMAs of the vestibular labyrinth were not significantly affected by the administration of a single dose of CDDP.

Effects of extracellular Ca2+ concentration on hair-bundle stiffness and gating-spring integrity in hair cells.

When a hair cell is stimulated by positive deflection of its hair bundle, increased tension in gating springs opens transduction channels, permitting cations to enter stereocilia and depolarize the cell. Ca2+ is thought to be required in mechanoelectrical transduction, for exposure of hair bundles to Ca2+ chelators eliminates responsiveness by disrupting tip links, filamentous interstereociliary connections that probably are the gating springs. Ca2+ also participates in adaptation to stimuli by controlling the activity of a molecular motor that sets gating-spring tension. Using a flexible glass fiber to measure hair-bundle stiffness, we investigated the effect of Ca2+ concentration on stiffness before and after the disruption of gating springs. The stiffness of intact hair bundles depended nonmonotonically on the extracellular Ca2+ concentration; the maximal stiffness of approximately 1200 microN.m-1 occurred when bundles were bathed in solutions containing 250 microM Ca2+, approximately the concentration found in frog endolymph. For cells exposed to solutions with sufficient chelator capacity to reduce the Ca2+ concentration below approximately 100 nM, hair-bundle stiffness fell to approximately 200 microN.m-1 and no longer exhibited Ca2+-dependent changes. Because cells so treated lost mechanoelectrical transduction, we attribute the reduction in bundle stiffness to tip-link disruption. The results indicate that gating springs are not linearly elastic; instead, they stiffen with increased strain, which rises with adaptation-motor activity at the physiological extracellular Ca2+ concentration.

Flavin adenine dinucleotide is a major endogenous fluorophore in the inner ear.

When illuminated with visible light, hair cells can exhibit autofluorescence (Lewis et al. [1982] Science 215, 1641-1643) concentrated in the basal pole near the synapses (Sento and Furukawa [1987] J. Comp. Neurol. 258, 352-367). The autofluorescence is enhanced by formaldehyde. The level of fluorescence is high enough to interfere with fluorescence microscopy of hair cells and to suggest that the fluorescent substance might have a particular role in hair-cell function. To identify this substance, we extracted a substance with formaldehyde-enhanced fluorescence from the inner ears of goldfish and purified it chromatographically. The substance copurified with FAD and had the same fluorescence emission spectrum. Two further results supported the identity of the endogenous fluorescent substance with FAD. First, as is the case with flavins, the autofluorescence in inner ear tissue examined within a few hours after fixation was reduced by addition of dithionite. Second, as is the case with the formaldehyde-enhanced fluorophore, the fluorescence of FAD was enhanced by formaldehyde. FAD accounted for 90% of flavins in goldfish inner ears; its concentration in the sensory epithelium was estimated to be about 30 nmol/g tissue weight, one of the highest tissue concentrations known. The FAD is probably associated with an unidentified flavoprotein concentrated in the basal, synaptic region of the hair cell.

Effect of ototoxic drug administration to the endolymphatic sac.

Aminoglycosides were applied to the endolymphatic sac of guinea pigs, and the inner ears were studied by light microscopy. In group 1 gentamicin sulfate application on the lateral surface of the endolymphatic sac produced sensory cell atrophy limited to the macula sacculi. In group 2 gentamicin injection into the sac by extradural and intradural approaches often showed lesions in the maculae sacculi and to the sac from a lateral opening resulted in a similar pattern of degeneration but with less severity in comparison to the injection method. Mild to severe endolymphatic hydrops was sometimes observed. The study shows diffusion of drugs taking place against the longitudinal flow of endolymph. This procedure may be applicable for the treatment of Meniere's disease.

[Electron microscopic studies on the glycocalyx of the utricular macula following amikacin in guinea pigs].

Twelve guinea pigs were randomly separated into 4 groups, one group as control, the other groups were given different dosages of Amikacin for 5 days. Ten days later, the animals were killed by decapitation and the right utricular maculae removed. After staining with reuthenium red, the glycocalyx was observed under TEM. In the control group, the glycocalyx was found covering the surface of both cells and sensory hairs and forming cross links with all the sensory hairs in the same hair band. The main changes in the experimental groups were: 1. decrease or disappearance of glycocalyx on the surface of cells and hairs; 2. fusion of sensory hairs and large blebs formed at the side of hairs. The results of the present study suggest that the decrease or disappearance of glycocalyx may be the first change of the damages caused by Amikacin.

Effect of gentamicin on the carbohydrates of the vestibular end organs: an investigation by the use of FITC-lectins.

The effect of gentamicin on the glycoconjugates in the vestibular end organs is demonstrated using FITC-lectins. Five milligrams of gentamicin dissolved in 0.1 ml. saline was injected in a single dose into the middle ear of adult guinea pigs. Seven days after the injection, the fluorescent reactivity of Wheat germ agglutinin (WGA), Abrus precatorius agglutinin (APA), Concanavaline A (Con A), and Latyrus odoratus agglutinin (LOA) was decreased in the apical epithelial cell surface as well as in the gelatinous layer of the otolithic membrane of the maculae. In the cupula of the crista ampullaris, the reactivity for WGA and Con A was reduced. The otoconia, however, showed no detectable changes. This indicates that gentamicin may induce an altered carbohydrate metabolism resulting in a decrease of N-acetyl-glucosamine, mannose, galactose and glucose in the glycocalyx lining the epithelial cells as well as in the gelatinous layer of the otolithic membrane or cupula in the vestibular end organs.

Morphological changes in the vestibular epithelia and ganglion induced by ototoxic drug.

The morphological changes of the vestibular sensory epithelia and the vestibular ganglions induced by Gentamicin(GM) were investigated using scanning electron microscope, transmission electron microscope and light microscope. The guinea pigs were injected with a single application of 4 mg (0.1ml) of GM into the middle ear through the tympanic membrane. The vestibular organs and the ganglions were observed up to 6 months after the treatment. Four days after the injection, fused, ballooned and missing cilia were observed in the vestibular sensory epithelia. These changes progressed and extended toward the periphery of the crista and the macula. The changes of the vestibular ganglions were first observed one month after the treatment. The degenerative process started from destruction of the mitochondrial cristae and vacuolization of the cytoplasm in the Schwann cell. The next step of the change was dissociation of the myelin sheath around the ganglion cell. The cytoplasmic organelles in the ganglion cell gradually deteriorated. At the later stage, the myelin sheath around the ganglion cell disappeared and the number of the cell reduced. Furthermore, the myelin sheath of the nerve fiber was dissociated. In this study the signs of the vestibular ganglion damage were later than that of the vestibular organ. However, we thought the changes in the ganglion are probably due to direct influence of GM, since the degeneration was found to develop in a relatively short period.

The ototoxic effects of gentamicin on the vestibular maculae in pigmented guinea pigs.

The normal density of the sensory hair cells of the utricular macula in pigmented guinea pigs was examined. The central region had a greater sensory cell density than the peripheral region. The effect of three different doses of gentamicin on the sensory hair cells was quantified. An increase in sensory cell damage was found with increasing doses of gentamicin. A differential regional sensitivity occurred. The central region was more vulnerable to gentamicin than the periphery. Only the central region of the saccular macula was examined, but no sensory cell losses were found with any of the dose regimes used. The utricular macula is therefore more sensitive to gentamicin than the saccular macula. This work was compared with a similar study using albino guinea pigs, and it was found that the pigmented animals are probably more resistant to ototoxic damage.

Chlorphentermine-induced changes of the peripheral vestibular sense organ of the rat.

Chronic feeding of chlorphentermine to young rats induces an accumulation of electron dense conglomerates and vacuoles in intraepithelial afferent nerve fibres of maculae and cristae in the vestibular part of the inner ear. In spite of the fact that preterminal parts of axons are severely swollen due to this accumulation, the nerve terminals maintain their normal size but can sometimes contain inclusions. Efferent fibres always remain free of these changes. Sensory cells, vestibular ganglion cells and occasionally supporting cells contain lamellar and crystalloid-like inclusions which disappear when the treatment is interrupted. After a long-term feeding break the nerve fibre inclusions disappear. A few swollen distended morphologically empty preterminals remain. The preference of inclusion body accumulation in afferent neurons may indicate an especially high lipid turnover in this part of the hair cell innervation.

Vestibular damage from chlorhexidine in guinea pigs.

In the early 1970s the suspicion arose that chlorhexidine, a disinfectant extensively used for preoperative skin disinfection, was ototoxic if introduced into the middle ear cavity. It has previously been shown by the author that chlorhexidine causes damage to the organ of Corti of guinea pigs when introduced into the tympanic cavity. In the present study the toxic effect of chlorhexidine on the vestibular part of the inner ear is investigated. Middle ears of guinea pigs were exposed to chlorhexidine in two different concentrations in two different solvents. The duration of exposure was varied and the animals were sacrificed and examined at various periods. Damaged neuroepithelia were seen in most of the animals. The extent of damage was related to the concentration of chlorhexidine, to the duration of exposure and to the time lapse after exposure.

Streptomycin-induced defects of the otoconial membrane.

Damage to the neuro-epithelium caused by the vestibulotoxic drugs has been well described in previous reports; however, little is known about the effect of such drugs on the otoconia. In this investigation, streptomycin sulfate was found to create a circumscribed defect in the crystalline layer of the otoconial membrane in the utricle of the guinea pig. The defect was secondary to a sharply outlined lesion of the neuroepithelium which corresponded in size and location to the otoconial defect. Many of the otoconia along the margin of the defect showed signs of degeneration and appeared similar to those observed in the saccule of man during aging. In advanced stages of decay, the otoconia were reduced to hollow shells consisting of longitudinal interconnecting strands. All the epithelial lesions occurred with remarkable consistency in the same region of the macula utriculi at the posterior end of the striola.