Hair cell recovery in the vestibular sensory epithelia of mature guinea pigs.

The progression of recovery of the vestibular sensory epithelia of guinea pigs after gentamicin-induced hair cell injury was assessed quantitatively and qualitatively. Evaluations were made of the number of cells bearing hair bundles by using scanning electron microscopy (SEM) and of identifiable hair cells in thin sections. Both assessment procedures showed that an initial loss of hair cells in utricular maculae is followed by significant recovery in the number of hair cells present. SEM also showed recovery in saccules comparable to that in utricles. During the recovery, progressive maturation of hair bundles, which exhibited features similar to those seen during normal ontogenetic development of hair cells, could be identified. The pattern and extent of hair cell loss and subsequent reappearance revealed by SEM corresponded with that derived from analysis of thin sections. This suggests that repair of nonlethally damaged hair cells is unlikely but, rather, that new hair cells are produced. An apparent decrease in supporting cell numbers was observed coincident with the increase in hair cell numbers. This complements previous morphological observations, which have suggested new hair cells arise from direct, nonmitotic transdifferentiation of supporting cells. The quantitative analyses indicate that more than half of the hair cells that are lost are replaced, but the recovery process does not result in complete restoration of the epithelium. Eight months after the end of drug treatment, the number of hair cells present was still significantly less than normal, and several other abnormalities persisted. There was also no evidence of any hair cell recovery in the organ of Corti. Thus, there appear to be limitations on the capacity for spontaneous replacement of lost hair cells in the mammalian inner ear.

Postnatal maturation of the organ of Corti in gerbils: morphology and physiological responses.

The organ of Corti, the sensory epithelium of hearing in mammals, matures postnatally in the gerbil. Quantitative analyses of the postnatal development of the organ of Corti, including supporting cells and the basilar membrane, were carried out. The morphological study confirmed that maturation of the sensory cells proceeds with a base-to-apex gradient, with the outer hair cells appearing to mature before the inner hair cells. Maturation of the supporting cells and the basilar membrane commenced first in the middle turn. Expansion of the second row of Deiters' cells began at 6 days after birth in the middle turn, before enlargement of the pillar cell heads at 8 days postnatally. Pillar cell head enlargement continued until 20 days postnatally in the middle turn. The tunnel of Corti and spaces of Nuel appeared first in the middle turn between 8 and 10 days postnatally. The maturation of the basilar membrane involved the thickening of the central hyaline layer and a reduction in the epithelial cells on the tympanic aspect. This process continued until about 20 days after birth. The cochlear microphonic potential, whole nerve action potential, and stimulus frequency otoacoustic emissions were recorded from 12 days after birth onward and related to changes in organ of Corti morphology. The results show that changes in the accessory structures continue throughout the period of onset and development of cochlear responses between 12 and 20 days after birth, and may therefore influence the micromechanical responses of the organ of Corti to acoustic stimuli during this period.

Postnatal development of membrane specialisations of gerbil outer hair cells.

Using a combination of freeze-fracture and thin sections, this study examines the maturation of the membrane specialisations of the gerbil outer hair cells (OHC) between 2 and 16 days after birth (DAB). The apical membrane, the junctional region around the neck of the cell, and the lateral and basal membranes are described. The results suggest a sequential development of the different components of the lateral wall. Intramembrane protein particles (IMP), the putative OHC motor elements, were found to be present at low density at 2 DAB and increased in density from 2200 IMP/microns 2 at 2 DAB to 4131/ microns 2 at 8 DAB. OHCs have been reported as showing electromotility from 8 DAB onward. IMPs continue to increase in density until mature values are attained at 16 DAB. Sub-surface cisternae did not appear until 8 DAB, with a single layer being complete by 10 DAB. Pillar structures, proposed to be related to the cytoskeletal lattice, first appear at 10 DAB. The apical membrane of the immature hair cell is characterised by the presence of pits related to the endocytosis of vesicles, and tip-links between stereocilia, thought to be associated with sites of ion channel opening, are present at 2 DAB. The junctional region comprises two areas which mature at differing rates: an apical-most region which attains an adult-like appearance by 8 DAB and a basal-ward region which continues to increase in complexity until mature at 16 DAB. The functional significance of the results are discussed in relation to the possible roles of the junctional regions and the proposed sites of the OHC motor elements.

Two modes of hair cell loss from the vestibular sensory epithelia of the guinea pig inner ear.

In the vestibular and auditory neurosensory epithelia of poikilothermic vertebrates and of birds, damaged sensory "hair" cells are often deleted by extrusion from the apical surface. In contrast, in the adult mammalian auditory epithelium (the organ of Corti), the bodies of damaged hair cells degenerate within the epithelium. To determine whether this apparent difference is species related or is associated with the differing structural organisation of the epithelia, hair cell deletion in the mammalian vestibular end-organs was examined. The structural organisation of these tissues is closer to that of the inner ear epithelia of lower vertebrates than to the organ of Corti. Hair cell loss was induced by chronic, systemic treatment of guinea pigs with the ototoxic aminoglycoside antibiotic gentamicin. The vestibular sensory epithelia were examined at various times after treatment via scanning electron microscopy, thin sectioning, and staining f-actin with fluorescently labelled phalloidin. Two distinct modes of hair cell loss were identified: 1) degeneration of hair cells within the epithelium, which often showed morphological features consistent with those described for apoptosis, and 2) extrusion of intact cells from the apical surface. Neither process caused the formation of obvious lesions through the epithelial surfaces. Expansion of adjacent supporting cells during hair cell deletion resulted in repair that appeared to preserve permeability barriers. There was also no evidence of inflammation accompanying hair cell removal. Thus, with both modes of hair cell loss, it appeared that deletion of hair cells was achieved without disruption of tissue architecture or integrity. This may be important for subsequent repair and regeneration processes to operate.

Ultrastructural evidence for hair cell regeneration in the mammalian inner ear.

It has long been thought that hair cell loss from the inner ears of mammals is irreversible. This report presents scanning electron micrographs and thin sections of the utricles from the inner ears of guinea pigs that show that, after hair cell loss caused by treatment with the aminoglycoside gentamicin, hair cells reappeared. Four weeks after the end of treatment, a large number of cells with immature hair bundles in multiple stages of development could be identified in the utricle. Thin sections showed that lost type 1 hair cells were replaced by cells with a morphology similar to that of type 2 hair cells. These results indicate an unexpected capacity for hair cell regeneration in vivo in the mature mammalian inner ear.

Structural variability of the sub-surface cisternae in intact, isolated outer hair cells shown by fluorescent labelling of intracellular membranes and freeze-fracture.

The intracellular membrane systems in intact, isolated outer hair cells were visualised using the fluorescent membrane probe 3,3'-dihexyloxacarbocyanine iodide (DiOC6) and by freeze-fracture, and f-actin distribution was examined with rhodamine-phalloidin. DiOC6 stained the sub-surface cisternal membranes in the lateral wall and revealed a membrane system running in the centre of the cell from the nucleus to the sub-cuticular region. In optical sections of the lateral wall of fluorescently labelled cells, obtained by scanning laser confocal microscopy, the sub-surface membrane appeared as a fenestrated sheet or a fine network of tubules. Freeze-fracture replicas of rapidly-frozen, unfixed outer hair cells also showed the sub-surface membrane as a fenestrated sheet in some cells or as a network of tubules in others. These combined studies indicate that the interruptions within the cisternal membranes as seen in normal thin sections of outer hair cells are not fixation artefacts but may reflect the dynamic and plastic properties of this membrane system. Double staining of cells with rhodamine-phalloidin and DiOC6 showed substantial co-localisation of intracellular membranes and f-actin. The results suggest there may be a continuous, dynamic endoplasmic reticulum system, forming a core in the centre of the cell, broadening in the subcuticular region and extending down the lateral wall, that may have a role in the turnover and distribution of cytoskeletal assemblies within the outer hair cell.

Scanning electron microscopy of the mammalian organ of Corti: assessment of preparative procedures.

Different fixation, drying and coating procedures have been applied in preparation of the organ of Corti for scanning electron microscopy (SEM), and structural features of the apical surface of the tissue in unfixed, freeze-fractured preparations used in assessing their effects on morphology. Fixation with glutaraldehyde alone or osmium tetroxide alone causes artefacts that are substantially avoided when tissue is doubly fixed in glutaraldehyde followed by osmium. Significant improvements in preservation are also obtained when tissue is additionally processed through thiocarbohydrazide-osmium (TOTO) processing. In addition to providing a conducting coat, it stabilises the tissue against deformations that might otherwise occur during drying, and reduces the extent of tissue shrinkage. Freeze-drying of TOTO processed tissue produces less tissue distortion than critical point drying (CPD) but is not so easy to apply routinely. The distortions of structure in TOTO-processed CPD tissue are not significant and this may be the preferred procedure for routine use, but air drying from hexamethyldisilazane is a useful alternative, producing results as good as those from CPD samples if TOTO processing is applied beforehand. One particular advantage of freeze-drying, though, is that after freezing, brittle fracture through the tissue can occur making examination of intracellular structure by SEM relatively easy. However, again, TOTO processing prior to freezing is of value as this appears to prevent the formation of large ice-crystal during freezing. Examination of isolated outer hair cells by SEM shows that isolation procedures do not cause significant damage to the stereociliary bundles.