JOURNAL OF ELECTRON MICROSCOPY TECHNIQUE 15:261-279 (1990)

Organization of Microtubules in Cochlear Hair Cells DAVID N. FURNESS, CAROLE M. HACKNEY, AND PETER S. STEYGER Department of Communication and Neuroscience, University of Keele, Keele, Staffordshire, ST5 5BG England

KEY WORDS

Cochlea, Hair cell, Cytoskeleton, Immunocytochemistry

ABSTRACT The organization of microtubules in hair cells of the guinea-pig cochlea has been investigated using transmission electron microscopy and correlated with the location of tubulinassociated immunofluorescence in surface preparations of the organ of Corti. Results from both techniques reveal consistent distributions of microtubules in inner and outer hair cells. In the inner hair cells, microtubules are most concentrated in the apex. Reconstruction from serial sections shows three main groups: firstly, in channels through the cuticular plate and in a discontinuous belt around its upper perimeter; secondly, forming a ring inside a rim extending down from the lower perimeter of the plate; and thirdly, in a meshwork underlying the main body of the plate. In the cell body, microtubules line the inner face of the subsurface cistern and extend longitudinally through a tubulo-vesicular track between the apex and base. In outer hair cells, the pattern of microtubules associated with the cuticular plate is similar, although there are fewer present than in inner hair cells. In outer hair cells from the apex of the cochlea, microtubules occur around an infracuticular protrusion of cuticular plate material. In the cell body, many more microtubules occur in the region below the nucleus compared with inner hair cells. The possible functions of microtubules in hair cells are discussed by comparison with those found in other systems. These include morphogenesis and maintenance of cell shape; intracellular transport, e.g., of neurotransmitter vesicles; providing a possible substrate for motility; mechanical support of structures associated with sensory transduction. INTRODUCTION: STRUCTURE AND FUNCTION OF MICROTUBULES Microtubules are cylindrical cytoskeletal filaments, ubiquitous in eukaryotic cells and usually constructed from 13 protofilaments, each composed of a chain of tubulin dimers (Cleveland and Sullivan, 1985; Erickson, 1974). They are thus readily recognizable in thin sections in the electron microscope as tubular structures, with a circular profile of about 24 nm diameter when cut transversely (Amos, 1979). More rarely, microtubules with larger and smaller diameters occur which have correspondingly greater or fewer protofilaments (Chalfie and Thomson, 1979,1982; EichenlaubRitter and Tucker, 1984; Saito and Hama, 1982). Microtubules have great functional diversity. They play a role in the development and maintenance of cell shape (Behnke, 1970; Tucker, 1979). Microtubules are responsible for many types of intracellular transport, such as fast axonal transport in squid giant nerve fibers (Allen et al., 1985), saltatory movements of organelles in cultured cells (Freed and Leibowitz, 19701, and spindle elongation and chromosome movement during cell division (e.g., Bajer and Mole-Bajer, 1972; Inoue and Sato, 1967). They are necessary for the movement of cellular extensions such as reticulopodia (Travis and Bowser, 1986) and the ubiquitous ciliary and flagellar axoneme (e.g., Gibbons, 1981; Satir, 1984). Several mechanisms have been suggested to account for microtubule-based motility. These include “treadmilling,” where microtubules move by assembling at

6 1990 WILEY-LISS, INC.

one end and disassembling a t the other (Margolis and Wilson, 1981); the sliding mechanism, where one microtubule slides along another, the basis of ciliary beating (Satir, 1984) and some cytoplasmic movements (Koonce et al., 1987); and saltatory movement where organelles move along microtubular tracks, perhaps by means of translocator proteins like kinesin (Vale et al., 1985). Biochemical evidence also suggests that actin, another protein involved in motility, may interact with microtubules (Griffith and Pollard, 1978; Pollard et al., 1984). Even within the same cell, different types of microtubule motility can coexist, for example the inhibitor vanadate can prevent ciliary beating but does not stop saltatory movements in culture cells (Buckley and Stewart, 1983). Despite the similar appearance of all microtubules (Amos, 1979), and although the basic polypeptide, tubulin, is relatively highly conserved (Luduena, 19791, there are, nevertheless, biochemically different, probably interconvertible, isotypes of tubulin even within a single cell (Gull et al., 1985). Variations in protofilament number could represent different biochemical and functional subtypes of microtubule. Microtubules show differential stability to a variety of disrupting agents (Behnke and Forer, 1967) such that cytoplasmic

Received July 8, 1988; accepted in revised form October 14, 1988. Address reprint requests to Dr. D.N. Furness, Dept. of Communication and Neuroscience, University of Keele, Keele, Staffordshire, ST5 5RG England.

262

D.N. FURNESS ET AL

microtubules are generally relatively unstable compared with ciliary microtubules. Moreover, they occur in many different arrangements, as single microtubules, or as bundles with a variety of packing organizations. Microtubules can thus show considerable biochemical, structural and functional diversity within a cell. Because of their cytoskeletal functions, ranging from morphogenesis and structural support to motility and intracellular transport, the organization of microtubules probably influences the mechanical properties of any tissue, and they are of considerable importance in the nervous system. Indeed, tubulin may constitute 42% of the soluble protein of chick brain tissue (Luduena, 1979). It is thus possible, given the behaviour of microtubules in other systems, that they may significantly influence the mechanical and neural properties of the organ of Corti.

MICROTUBULES IN THE ORGAN OF CORTI In the organ of Corti, microtubules occur in large quantities in the pillar and Deiters’ cells (Angelborg and Engstrom, 1972) where they seem to have mainly 15 protofilaments (Saito and Hama, 1982). In these cells, the microtubules interdigitate with actin filaments in massive highly organized arrays (Angelborg and Engstrom, 1972; Slepecky and Chamberlain, 1983).Microtubules have also been observed in ultrastructural studies of a variety of hair cells from several species. The hair cells are characterized by a n apical bundle of stereocilia, which are not true cilia but are modified microvilli containing parallel arrays of actin filaments (Flock and Cheung, 1977; Tilney et al., 1980). Usually, the hair bundle includes a true cilium or kinocilium which contains the typical 9 + 2 microtubular arrangement with accompanying basal body (see reviews by Hudspeth, 1983; Zenner, 1986a). In the cochlea of adult mammals, however, the kinocilium is only completely present during development (Bredberg e t al., 1972; Kimura, 1966) and is then lost, leaving only the basal body. Apart from forming the basal body, microtubules have been described in other regions of these hair cells. They occur in the apex of inner hair cells (IHCs), where they are associated with the striated organelle (Friedmann body-Friedmann et al., 1965) and occur around the basal body (Engstrom and Engstrom, 1978; Slepecky et al., 1981). In outer hair cells (OHCs) microtubules occur adjacent to the innermost face of the lateral, laminated cisternae (Lubitz, 1981b; Raphael and Wroblewski, 1986). They are also said to pass longitudinally between the apex and base of both IHCs (Engstrom and Engstrom, 1978) and OHCs (Zenner, 1986a) and are associated with synapses (Lim, 1986; Nadol, 1983a,b). Similar distributions of microtubules have been noted in other types of hair cell also. In vestibular hair cells of rat, microtubules occur a t the apex, again associated with striated bodies (Jorgensen, 1982) and pass longitudinally towards the synaptic region (Favre and Sans, 1983; Jorgensen, 1982). This orientation is prominent also during development (Favre et al., 1986; Heywood et al., 1975). In non-mammalian species, microtubules are found in the apex of hair cells of the

chick basilar papilla (Cohen and Fermin, 19781, the snake auditory system (Jorgensen, 1982), the ratfish lateral line (Lubitz, 1981a) and the frog crista ampullaris (Flock et al., 1981a). Frequently, they are associated with the basal body a s in mammalian hair cells and possibly with the stereociliary rootlets. Immunocytochemical techniques have also been employed to show tubulin-associated immunoreactivity in the organ of Corti. Strong labelling of the supporting cells correlates well with the presence of microtubules (Flock et al., 1981b, 1982; Slepecky and Chamberlain, 1985, 1987). Immunolabelling of tubulin in hair cells has, however, been more ambiguous. Zenner (1981) showed tubulin-associated immunoreactivity within the bases of stereocilia and the cuticular plate. Slepecky et al. (1988) demonstrated networks of tubulin in the OHC body in sections of freeze-dried and embedded organ of Corti, but have not shown any evidence of a n apical concentration. Together, evidence from electron microscopy and immunocytochemistry suggests that there may be a particular distribution of microtubules in hair cells. In order to investigate this further, we have undertaken a study of microtubular organization in hair cells using primarily transmission electron microscopy (TEM) for reconstruction from serial sections. Immunocytochemical studies using a monoclonal anti-tubulin antibody on surface preparations of the organ of Corti have enabled the pattern of tubulin-associated immunoreactivity in whole cells to be observed a t the light microscope (LM) level.

MATERIALS AND METHODS Pigmented guinea-pigs were anaesthetized with sodium pentobarbitone, decapitated, and the bullae were removed. The cochlea was exposed and fixed by perfusion through the round window and a small hole in the apex. The primary fixative was 2.5% glutaraldehyde in 0.1 M sodium cacodylate buffer containing 2 mM calcium chloride for 2 h, followed by postfixation with 170 osmium tetroxide in the same buffer for 1 h. The cochleae were dissected into segments in 70% ethanol, dehydrated and embedded in Spurr resin, as described previously (Furness and Hackney, 1985). Serial ultrathin sections were cut with a diamond knife, in a plane parallel to the reticular lamina (Fig. l ) , mounted on Formvar-coated slot grids, and stained in 2% uranyl acetate in 70% ethanol, followed by 2% aqueous lead citrate. These were used for three dimensional (3-D) reconstruction of the apex of IHCs. Other sections were cut radially (Fig. 11, mounted on 200 mesh grids and stained as above, to confirm the pattern obtained in the reconstruction. Regions of OHCs were included in the sections and examined for comparison with IHCs. All sections were examined in a JEOL 100 CX EM a t 80 or 100 kV. For reconstruction, all micrographs from the series were printed a t the same magnification, and for each section, outlines of the cell, cuticular plate, and positions of microtubules were traced onto a n acetate sheet. The outlines were superimposed serially to reveal the 3-D shape of the cuticular plate and the distribution of microtubules in the apex.

MICROTUBULES IN HAIR CELLS

263

cific and disregarded in descriptions of the pattern of tubulin-associated immunoreactivity. There is discretely filamentous specific labelling in primary incubated material which has not been observed in any control. The antibody has also been immunoblotted against protein extracts of the cochlea (courtesy of G.P. Richardson of the University of Sussex) and found to be specific for a single protein of molecular weight approx. 50 kD,which is very close to that of tubulin (a fuller account of the immunocytochemistry will appear in papers in preparation).

Fig. 1. Diagram of a n IHC showing the section planes used, either parallel to the reticular lamina (A), or radially (B).

In order to determine whether different types of microtubule occupied different locations, examples were selected and their diameters measured. Microtubules were sampled from radial sections of IHCs, and inner and outer spiral nerve fibers, in the same sections, for comparison. A constant magnification of 66,000 x was used, with all locations on the specimen adjusted for eucentricity, and examples from every microtubular group were taken in each session. At intervals throughout each session a calibration image was recorded, using a catalase crystal with major and minor subunit spacings of 8.75 nm and 6.85 nm respectively. The microscope was calibrated and the degree of fluctuation of magnification both during and between sessions was found to be negligible. For immunocytochemistry, cochleae were obtained as above and fixed with 4% paraformaldehyde in 0.12 M phosphate buffered saline (PBS), pH 7.3, for 2 h. The bony shell was then removed, and the whole cochlea was incubated in 0.25% Triton-X 100 in PBS containing 1%ox serum (OPBS) for 30 min, to permeabilize the tissue, followed by neat ox serum to block nonspecific binding. The cochlea was then incubated a t room temperature with a monoclonal anti-yeast tubulin primary antibody (Sera-Lab, UK) diluted between 1:20 and 1:20,000 in OPBS for 24 h, washed, and subsequently incubated with a n FITC-conjugated secondary antibody diluted 1:20 in OPBS, for 5 h. The cochlea was dissected in the OPBS buffer; mounted in a n antibleaching solution composed of 0.01% p-phenylene-diamine in 30% glycerol and 0.6% polyvinyl alcohol in PBS, pH 8.0; and viewed with a Leitz Dialux 20 EB microscope fitted with epifluorescence optics and a n FITC filter block. Images were recorded on Kodak Technical Pan 2415 film. For controls, cochleae were treated a s above but with either saline or non-immune serum replacing the primary antibody. Saline controls show no staining. Material incubated with the primary antibody, and material incubated with non-immune serum replacing the primary antibody, both show faint, diffuse fluorescence over the stereocilia, which is thus considered non-spe-

RESULTS Microtubules are found in varying quantities in all cytoplasmic regions of both IHCs and OHCs. In IHCs they are particularly concentrated at the apex, whilst in OHCs they are more concentrated at the base of the cells. The microtubules are found in a complex and consistent pattern which is qualitatively similar in both cell types. It was found, by selecting certain focal planes in the immunolabelled surface preparations, that the pattern of fluorescence obtained by LM could be correlated very closely with the distribution of microtubules in appropriate section planes in TEM. In the plates, therefore, the relevant LM views are included with the corresponding TEM sections. The pattern of microtubules found in different regions of the hair cells is intimately related to the structural organization of those regions. Thus, in order to make a n adequate description of the distribution of microtubules, i t is necessary to redescribe some features of the general organization of the different regions prior to presenting the microtubular organization. The inner hair cell apex General organization. The upper surface of the IHC is usually concave (Fig. 2) due to the presence of a wide, shallow groove, from which the stereocilia project, lying along the longitudinal axis of the hair cell row. Fibrous cuticular plate material fills much of the extreme apical part of the cell, except for specific cytoplasmic areas. The upper surface of the plate lies directly beneath the apical plasma membrane except for a few narrow peripheral zones (Figs. 2, 3A). In a single cell, the depth of the cuticular plate is constant over most of its area, but a t the edges it extends to a greater depth, particularly on the side closer to the stria vascularis (strial-facing side) (Figs. 2, 4A, 5A). The difference in depth between the periphery and the centre is most marked in cells from the apical turns of the cochlea, which generally appear to have rather thinner cuticular plates than those of more basal turns. For example, the hair cell shown in Figure 2 is from the apex of the cochlea and has a cuticular plate with a depth (in the centre) of 1.2 pm, whilst in the basal hair cell in Figure 5B, it is 2.6 pm (measurements are taken from comparable section planes in the two cells). The upper perimeter of the cuticular plate is partially encircled by a belt of cytoplasm which separates it from the apical membrane (Figs. 2, 3A). The belt is continuous with a wide channel of usually fixed position, located in the area between the stereociliary bundle and the stria1 side of the cell (Figs. 2A, 3A, 5A,B).

264

D.N. FURNESS ET AL

Fig. 2. Radial sections of a n IHC. The direction of the stria vascularis is indicated by arrowheads. A: Section through the centre of the apical region showing the curved upper surface and cuticular plate (Cp), the main channel (Ch) through it, and lateral extensions downwards (E) on the stria1 side. The channel contains a basal body (bb), an area of electron-dense membrane specializations (*I, and is lined by many microtubules (arrows). Microtubules also occur in the upper cytoplasmic belt (l),in the lower ring (21, in both cases sectioned more or less transversely because of the orientation, and in a meshwork ( 3 ) below the main area of the cuticular plate. Microtu-

bules extend down into the cell body. Note the many mitochondria, and the tubulo-cisternal system. Scale bar = 1 pm. B: Section through a more lateral plane of the cell shown in A. The upper cytoplasmic belt is sectioned on both modiolar and strial-facing sides of the cells and contains many microtubules (1).Microtubules of the lower ring (2) and meshwork ( 3 ) can also be seen. The inside of the lateral lower extension (E) of the cuticular plate appears striated, with microtubules running parallel to the striations. Note also the tubulo-cisternal and vesicular system (TV). Scale bar = 1 pm.

266

D.N. FURNESS E T AL.

Fig. 4. A Section from the same series as Figure 3A, in a plane below the main body of the cuticular plate. The lateral downward extension of the plate (El is lined by a ring (2) of parallel microtubules. A meshwork (3) of microtubules encircled by the ring occurs directly beneath the cuticular plate. Note also the many dense granules ( G )and mitochondria in this area. Scale bar = 2 km. B: Tubulin-

associated immunoreactivity in IHCs, a t a level corresponding with the section shown in A. Here the lower ring ( 2 )and meshwork (3) can be seen correlating with features seen in the TEM section. Note that the meshwork shows variability in intensity of staining. The cell bodies of the hair cells contain a filamentous pattern of label. Cells A and B are the same cells as those indicated in Figure 3B. Bar = 2 p,m.

MICROTUBULES IN HAIR CELLS

267

Fig. 5 . A Detail of the main channel through the cuticular plate of a n IHC. The basal body (bb) is located directly beneath the apical membrane. Microtubules line the channel and run through the centre (arrows) where there are many vesicles and partially-coated membraneous tubules (arrowheads). The microtubules of the lower ring (2) are cut transversely. Note the patch of membrane density associated with four regular sub-membraneous dense bodies (*). Scale bar = 0.5 ym. B: Low magnification view of a n IHC apex showing the main

channel through the plate (Ch) and a smaller channel (Sch) both of which are lined by microtubules. The basal body (bb) is displaced from the apex and is associated with dense accumulations in the channel (arrows). The apical membrane shows the area of specialization over the channel (*I. Scale bar = 2 pm. C: Cross section of the upper cytoplasmic belt, showing a bundle of microtubules, some of which are linked together by inter-microtubular arms (arrow). Scale bar = 0.2 Fm.

This channel passes between the upper and the lower surfaces of the plate widening progressively, and contains the basal body which usually occurs directly beneath the apical plasma membrane (Fig. 5A), but sometimes is displaced downwards (Fig. 5B). An interesting feature of the apical membrane overlying the channel is a small area of membrane specialization, approx. 500-600 nm across. The membrane is smooth and relatively thick compared with the adjacent area, and beneath it there are dense, 75-100 nm wide regular structures (Figs. 2A, 5A,B). There are also several smaller channels within the cuticular plate, near the periphery, which pass between the sub-cuticular region and the upper belt of cytoplasm (Fig. 3A). The appearance of cuticular plate material is very similar in all IHCs. The overall matrix of the plate consists of material of intermediate staining density which is filamentous. Within this matrix there are dense patches, arranged in several rough layers which lie parallel with the apical surface, the uppermost two or three of which are composed of larger patches than the remainder (Figs. 2, 5B). At the edges of the cuticular plate there are laminated areas where dense strands and 6 nm diameter microfilaments form paral-

lel arrays. The arrays are generally orientated circumferentially and can occupy an extensive area (called the Friedmann body) (Fig. 3A). The lamination can also be quite prominent on the inner side of the peripheral downward extension of cuticular plate material (Fig. 2B). The stereocilia have dense rootlets which extend down into the cuticular plate. These are ensheathed by areas of very lightly stained material. The cytoplasm immediately surrounding the cuticular plate contains many organelles. There is a welldeveloped tubulo-cisternal system of membranes (Fig. 2B) and large numbers of vesicles, which vary greatly in diameter but are usually between 80 and 200 nm. The vesicles and tubules can also be found in the cytoplasmic channel and belt, right up to the apical surface (Figs. 2A, 5A). There are many mitochondria, Golgi bodies, and large numbers of dense, 70 nm diameter granules just beneath the lower surface of the cuticular plate. Microtubular organization. There are many microtubules enclosing the main body of the cuticular plate and following its contours. They have not been observed to penetrate directly into the cuticular plate material, even around the stereociliary rootlets, al-

268

D.N. FURNESS E T AL

though they pass through the plate where cuticle-free cytoplasmic channels occur. In Figure 6, the distribution of microtubules in the apex of a single IHC has been mapped by superimposing several adjacent sections in a group a t different levels through the cuticular plate region. The distribution determined from serial sections parallel with the reticular lamina has been confirmed in radial sections. The microtubules occur in at least three main groups. The first (group 1) occupies the cytoplasmic zones which pass through the cuticular plate, that is the upper cytoplasmic belt, where they form a discontinuous ring, and the channels, particularly the main channel (Figs. 2,3A, 5A, 6A). Those in the cytoplasmic belt occur in a bundle with up to 40 microtubules orientated in parallel along the modiolar-facing side of the cell (Figs. 2, 5C). There are fewer on the strialfacing side (Fig. 2B). The microtubules in the bundle show relatively poorly organized packing, although some may be linked together by intermicrotubular arms (Fig. 5C). Microtubules in the wide channel occur mainly lining the edge with a few passing through the centre (Fig. 5A). The second (group 2) forms a more-or-less complete ring encircling the cell, just below the main body of the plate (Figs. 2,4A, 6C). The microtubules in the ring are also largely orientated in parallel with each other. They occupy a narrow strip along the inside edge of the rim of cuticular plate material which extends down the edges of the cell. The third (Group 3) forms a meshwork directly beneath the centre of the plate, lying in a plane parallel with the lower surface. Within the plane they show no strongly preferred orientation, although microtubules of similar orientation tend to occur in groups (Figs. 4A, 6B). There are many of the 70 nm granules in the region of the meshwork, and below. These microtubular groups completely enclose the cuticular plate on all sides except where the apical and lateral surfaces lie directly adjacent to the plasma membrane. In the map shown in Figure 6B the microtubules of the meshwork and the lower ring are difficult to distinguish, partly because the maps are compressed into two dimensions. In Figures 2 and 6C, the two appear spatially distinct, though they may share a common pool of microtubules. The upper ring microtubules also extend through the channels in the cuticular plate to enter the zones occupied by the ring and meshwork, suggesting continuity of all the groups. To determine whether the different groups which can be observed are different in origin, structure, and perhaps function, representatives from each group have been selected and characterized. Measurements of their diameters are shown in Table 1, with, for comparison, measurements of microtubule diameters from the outer and inner spiral nerve fibres. The microtubules within the groups of the IHC and in the nerve fibers are all quite similar in diameter (no significant differences were found between any two of the means when compared using a two-tailed T-test). Microtubules from each group also seem to extend deep into the cell body. Others line the Friedmann body region of the cuticular plate (Fig. 3A) with the same orientation as the strands within the body. The immunocytochemical studies show patterns of tubulin-associated immunoreactivity which correlate with the patterns of microtubules reconstructed from

Ch

A

B

Fig. 6. Maps showing the distribution of microtubules at different levels through the apex, determined from serial sections of a single IHC. Each “map” is obtained by superimposing several consecutive drawings of serial sections, and plotting the positions of the microtubules in the resulting slice to produce a two-dimensional representation. Note that this method compresses three dimensional information into two, resulting in some blurring of the spatial distinction between groups. Scale bar = 5 IJ-m.A: Apical slice composed of six sections. Outlines of five of the sections are included to create a contour map of the upper surface, which shows the longitudinal groove (arrow) in the centre. The contours are numbered *1-*5 in descending order through the cell, and are spaced at a section thickness of 80-100 nm. The microtubules within this slice are concentrated into the larger (Ch)and smaller channels (arrows) through the cuticular plate, and pass by the Friedmann body (FB) and around the upper perimeter (1).Position in cell 600-1,200 nm below uppermost section. B: Subapical slice composed of the five sections immediately subsequent to those used for A. Only a single border is shown because the sections have very similar outlines. The meshwork of microtubules is most clearly represented a t this level (31, although circumferentially oriented microtubules comprising the lower ring can be discerned (2). Position in cell approx. 1,200-1,700 nm below uppermost section. C: Slice composed of five sections immediately subsequent to those used for B. At this level, the meshwork is less concentrated, and the lower ring is more apparent. Note that the ring is best represented on the strial side of the cell, where it lies more deeply than on the modiolar side. Position in cell 1,700-2,200 nm below uppermost section.

MICROTUBULES IN HAIR CELLS TABLE 1. T h e comparative diameters o f microtubules from different groups'

Microtubule proup IHC 1 IHC 2 IHC 3 Inner spiral fibers Outer spiral fibers (Combined IHC data) (Combined neurotubule data)

Mean diameter (nm) 26.9 ~~

~

26.5 26.5 26.5 25.9 (26.6) (26.3)

S.D.

n

f 1.5 ~.

5n ..

1.6 1.7 1.4 1.8 1.6) 1.5)

52 52 40 21 (154) (61)

f

+f

t (+(?

'Data showing mean diameters of samples of microtubules, taken from radial sections of the organ of Corti.

TEM. By focusing down through a surface preparation, labelling can be identified in the main channel and upper ring (Fig. 3b) and in the lower ring and meshwork (Fig. 4B). The upper ring is rarely continuous around the entire apex, and there are small patches of fluorescence around the periphery of the cuticular plate. These patches often connect the upper and lower rings, and correspond to the small channels of cytoplasm through the cuticular plate, which contain microtubules (Fig. 3B). The lower ring is usually complete, and encircles the meshwork. In both the TEM and immunofluorescence the meshwork is often well represented in some cells, and the ring may appear to be simply the edge of the meshwork pressed into a circular pattern by the edges of the cell and the extension of the cuticular plate. In other cases, however, the meshwork is rather poorly represented (Fig. 4B), whilst the ring is well represented.

The inner hair cell body General organization. The flask shaped cell body has a more-or-less central nucleus, many mitochondria, and fairly consistently located areas of endoplasmic reticulum which are common to most IHCs. Arrays of rough endoplasmic reticulum (RER) occur around the nucleus, associated with accumulations of mitochondria (Fig. 7A,B). A tubulo-vesicular system of mainly smooth endoplasmic reticulum (SER) and associated coated and uncoated vesicles forms a track which extends from the apical Golgi body zone to the synaptic region at the base of the cell (Fig. 7A). A single layer of subsurface cisternae is located around the periphery of much of the cell (Fig. 7A), associated with a layer of mitochondria (Fig. 7B). Microtubular organization. Microtubules are found generally within the cytoplasm and have mainly longitudinal orientations. They are more specifically located directly inside the sub-surface cisternal layer (Fig. 7C), and in the track of smooth endoplasmic reticulum, often as bundles of several parallel microtubules (Fig. 7D). Apart from these particular locations, the microtubules generally occur as a network ramifying throughout the cell body which can be seen most clearly in the immunocytochemical preparations (Figs. 3B, 4B).

269

The outer hair cell apex General organization. The apical surface of OHCs is usually rather flatter than that of IHCs (Fig. 8A). The upper surface of the cuticular plate lies directly beneath the apical plasma membrane except for peripheral zones of cytoplasm (Fig. 9A). A channel of cytoplasm on the strial-facing side of the cell extends between the upper and lower surface of the cuticular plate, widening progressively towards the lower surface (Figs. 8B, 9A) and frequently contains a basal body. The area of apical membrane overlying the channel has an area of membrane specialization and dense patches similar to that of IHCs (Fig. 8B). There are numerous small channels of cytoplasm through the periphery of the cuticular plate (Fig. 9A). The peripheral edges of the cuticular plate extend down the lateral surfaces of all OHCs. In OHCs of more apical turns, the lower surface of the cuticular plate gives rise to a projection which extends deeply into the cell and is surrounded by Golgi bodies and other organelles (see below) (Figs. 8A,C, 10A).The projection is laminated, with striations orientated roughly parallel to the long axis of the cell and is composed of microfilaments and dense strands (Fig. 8C). In more basal turns the lower surface of the cuticular plate is more regular than in more apical turns. Microtubular organization. The organization of microtubules is qualitatively similar to that of IHCs, but there are far fewer present in the apex of OHCs. Representatives of each of the three groups described for IHCs have been observed. The main channel has many microtubules lining it (Fig. 8B) and many more extend into the cell body, orientated longitudinally in association with the projection of the cuticular plate when present (Fig. 8C). The smaller channels and upper cytoplasmic belt have few microtubules in them. A few microtubules form a lower ring and meshwork (Fig. 11A). Immunofluorescent staining reveals a large patch and several smaller spots of label which correlate with the large and small channels through the cuticular plate (Fig. 9B). In cells from the apical region of the cochlea, there is a central concentration of tubulin-associated immunoreactivity located beneath the cuticular plate which represents the microtubules adjacent to the projection (Fig. 10B). In all OHCs, there is a thin lower ring and meshwork (Fig. 11B) but an upper ring is seen only infrequently. The outer hair cell body General organization. The cell body is cylindrical and has a nucleus located towards the basal end. The cytoplasm in the centre has few organelles, unless there is a projection from the base of the cuticular plate. When this is present, Golgi bodies, mitochondria, and stacks of laminated membranes accompany the projection (Figs. 8A,C, 10A). At the periphery of the cell, there are several layers of laminated cisternae, although these usually reduce to a single layer towards the base, and are largely absent in the synaptic region (Fig. 12A). Mitochondria are associated with the cis-

270

D.N. FURNESS ET AL

Fig. 7. A: Radial section of an IHC showing the features of the cell body. Mitochondria are concentrated at the apex, in association with Golgi bodies, and in loose clusters below the nucleus, associated with RER. A track of mainly SER stretches from the Golgi region to the base (arrows). The area in the box is shown in greater detail in D. Subsurface cisternae occur around most of the cell periphery (appearing as a denser area of membrane a t this magnification, arrowheads). Scale bar = 5 pm. B: Radial section viewed by LM (2 pm thick resin section, stained with 1%cresyl violet in 70%, ethanol). In this thicker

section, mitochondria can be seen to form a layer a t the periphery of the cell, associated with the areas likely t o have subsurface cisternae 5 pm. C: Tangential section of the periphery of (arrows). Scale bar an IHC, showing the subsurface cisternal system (SSC) and associated microtubules (arrows).Scale bar = 1 pm. D: Detail of the tubulocisternal track extending between the apex and base (region outlined in A). The track contains coated and uncoated vesicles, and tubules of SER. Many microtubules, some in bundles, are orientated longitu0.5 pm. dinally through the track (arrows). Scale bar y

7

MICROTUBULES IN HAIR CELLS

Fig. 8. Radial sections of the apical region of OHCs. A: Section showing OHCs from rows 2 and 3. Note the flat upper surface and irregular lower surface of the cuticular plate. Some of the stereociliary rootlets penetrate completely through the plate. The cuticular plate material projects deeply into the cell body (P) with associated mitochondria and other organelles. Mitochondria can also be seen to form a layer around the periphery of the cells, associated with the

271

subsurface cisternae. Scale bar = 5 pm. B: Detail of the channel region of an OHC showing the lining of microtubules (arrows) and an area of membrane specialization similar to that of IHCs (*I. Scale bar = 1 pm. C: Detail of the deep projection showing the highly ordered array of microfilaments (arrowheads) and dense material. Many microtubules occur around the protrusion. Scale bar = 1 pm.

272

D.N. FURNESS ET AL.

Fig. 9. A Section through the apical region of OHCs (row 2 ) in a plane parallel to the reticular lamina. Note the single main channel (Ch) and several peripheral small channels which penetrate the cuticular plate (arrows). In some cells a thin, discontinuous belt of cytoplasm occurs around the upper region of the plate (e.g., arrowheads). Scale bar = 5 pm. B Tubulin-associated immunoreactivity in the upper region of the apex of OHCs (row 2). Note the prominent staining in the large channel (Ch), and the few small spots of label correlating with some of the small channels (arrows). The supporting cell apices, around the hair cells, are strongly labelled (S).Scale bar = 5 pm.

Fig. 10. A Section through OHCs of row 3, in a plane parallel to the reticular lamina, a t a level below the main cuticular plate. The protrusion into the infracuticular region is visible (P) in the centre of the cells. Scale bar = 5 pm. B: Tubulin-associated immunoreactivity in a region corresponding to that shown in A. Note the central patch of weak fluorescence which has bright edges, indicating that the label is mostly concentrated around the region rather than in it. Scale bar = 5 km.

MICROTUBULES IN HAIR CELLS

273

Fig. 11. A Section through the apex of an OHC, cut parallel to the reticular lamina, in a plane towards the lower surface of the cuticular plate. Microtubules corresponding to the meshwork (3) and lower ring (2) of IHCs can be discerned but are fewer in number. Scale bar = 2 pm. B: Tubulin-associated immunoreactivity at a level corresponding

to that of the section shown in A. A thin lower ring of fluorescence (2) encloses the meshwork (3),just below the cuticular plate in the OHCs. The phalangeal processes of the surrounding Deiters’ cells (Dc) are strongly labelled. Scale bar = 5 pm.

ternae, and are concentrated at the extreme apical and basal ends of the cell (Figs. 8A, 12A). Microtubular organization. Microtubules are present throughout the cytoplasm of the cell body, but in relatively small numbers above the nucleus. They occur along the central axis of the cell, in particular extending beyond the apical protrusion when present, perhaps as far as the nucleus. They also occur at the periphery inside the innermost layer of the cisternae. The microtubules are rather more concentrated around and below the nucleus (Fig. 12A). Here two groups can be distinguished on the basis of orientation. The microtubules at the periphery generally lie in parallel arrays with a common orientation. Frequently, though not exclusively, they are orientated circumferentially along the inside face of the single cisternal layer (Figs. 12A-C). The central microtubules form a halo around the nucleus (Fig. 12D) and are present in large numbers, but are randomly orientated, particularly below the nucleus, among the many mitochondria (Fig. 12A). Immunostaining reveals tubulin-associated immunoreactivity forming a network in the cell body, often concentrated in a central longitudinal band above the nucleus, and an intense concentration below the nucleus (Fig. 13), consistent with the pattern seen in the

TEM. However, the different groups of microtubules below the nucleus have not been distinguished by this technique. The consistent features of microtubular distribution in the inner and outer hair cells of the guinea pig cochlea are summarized schematically in Figures 14 and 15.

DISCUSSION The distribution of microtubules described here using TEM correlates well with our immunofluorescent localization of tubulin using a monoclonal antibody, and forms consistent patterns in guinea-pig cochlear hair cells. Our results extend considerably previous ultrastructural and immunocytochemical studies of this important component of the cytoskeleton. For example, the concentration of microtubules in the apex of IHCs, noted by Engstrom and Engstrom (1978) and present in other hair cell types (e.g., Jgrgensen, 1982), is very highly organized, with well defined groups within it. The microtubules in this region form a network enclosing the plate. Although some authors have observed that microtubules approach the stereociliary rootlets (Flock et al., 1981a; Jorgensen, 1982) and perhaps enter the plate, they rarely directly penetrate the cuticular plate matrix elsewhere. It is possible that micro-

274

D.N. FURNESS ET AL.

Fig. 12. Radial sections through the base of an OHC. A: The region below the nucleus has a concentration of microtubules within which two groups can be observed. The first group of microtubules occur apparently randomly orientated in the centre, amongst the cluster of mitochondria (arrows). At the edge, the second group of microtubules line the inner face of the cisternal system, and in this cell are orientated largely in a transverse ring (outlined in the boxes on either side). The box marked (*I is shown in detail in B. Scale bar = 1 pm.

B: Detail of the box (*I from A. The microtubules line the inner face of the cisternal system and are cut transversely (arrows) indicating their orientation is probably circumferential. Scale bar = 0.25 pm. C: Tangential section of the periphery of the hair cell, showing the band of parallel microtubules internal to the cisbernal layer. Scale bar = 1 pm. D: Section to one side of the nucleus. The nucleus is often surrounded by a halo of microtubules (arrows). Scale bar = 1 pm.

MICROTUBULES IN HAIR CELLS

Fig. 13. Tubulin-associated immunoreactivity in the cell body of

OHCs. The nucleus is unstained (N) but is surrounded by a faint halo of label. Above the nucleus, there is sometimes a thin central strand of label running longitudinally ( e g , arrow), whilst below the staining

tubules may only enter the less dense regions around the rootlets, or at places where the matrix is penetrated by cytoplasmic channels. If so, the microtubules may have the spatial identities described here, a s a result of their exclusion from specific areas of the cuticular plate material. In OHCs, a similar pattern exists in the apex, although there are fewer microtubules present. However, the infracuticular protrusion of the cuticular plate, described previously by Angelborg and Engstrom (19731, is also surrounded by microtubules (see also Thorne et al., 1987) like the rest of the cuticular plate, and so there appears to be a n intimate association of microtubules with cuticular plate material. The protrusion may correspond with the infracuticular actin-containing network seen preferentially in OHCs from the apical turns of the cochlea (Steyger et al., 1988; Thorne et al., 1987). In the IHC cell body, microtubules occur in a special track of SER and vesicles, rather than being distributed more evenly throughout the cytoplasm. In OHCs, relatively few microtubules occur in the cell body, although there is a basal concentration, noted e.g., by Lim (1986), which has a previously unrecognized organization, in which two apparently separate groups can be defined. In both cell types, microtubules can be associated with subsurface cisternal membranes, a fea-

275

is intense but there is little detail visible (arrowheads). Note that the Deiters’ cells also show staining around the OHC base. Scale bar = 5 IJ-m.

ture described previously only for OHCs (Lubitz, 1981b; Raphael and Wroblewski, 1986). The results of the immunocytochemistry also differ in some respects from previous studies. This could be because immunostaining may reveal more than one form of tubulin. The filamentous staining described here contrasts sharply with the diffuse staining of stereociliary bases and cuticular plate shown by Zenner (1981) which was thought to represent unpolymerised tubulin. Slepecky et al. (1988) have shown a fibrillar network of tubulin-associated immunoreactivity in the OHC body and a concentration at the OHC base, which is in agreement with our observations, but have not demonstrated the apical organization observed here. These results may differ as a consequence of preparation method. Our method involves chemical fixation and observation of surface preparations, whilst, most recently, Slepecky et al. (1988) have used freeze-drying followed by embedding and sectioning. Since cold temperatures can disassemble microtubules, some more readily than others (Behnke and Forer, 19671, cryofixation may not be the most appropriate technique for the optimal preservation of intact microtubules. Moreover, the surface preparation technique used here has the advantage over sectioning that the distribution of label is visible in 3-D in many intact cells simultaneously.

276

D.N. FURNESS ET AL.

. Fig. 14. Schematic diagram showing the approximate locations and orientations of different microtubular groups in the IHCs. Microtubules are concentrated at the apex where they fall into the three main groups: 1)microtubules lining the upper cytoplasmic belt and the channels through the cuticular plate (Cp); 2) microtubules in a

Microtubular function in hair cells

lower ring inside the rim extending beneath the edge of the plate; and 3) microtubules forming a meshwork beneath the centre of the plate. In the cell body microtubules are found throughout, but are concentrated in the tubulo-vesicular track (4Tv)and around the edges (5) lining the subsurface cistern. The nucleus (N) is shown in the centre.

may subserve a slower, efferent control on cochlear sensitivity (Brownell et al., 1985; Kim, 1984). It has, so Before discussing the possible functions of microtu- far, not been possible to show motility of isolated IHCs. There is no direct evidence available to enable us to bules in inner and outer hair cells, it is worth considering briefly the respective contributions of these cells assign particular roles to the microtubules in hair cells. to cochlear function. The two hair cell types of the Nevertheless, evidence from other systems allows us to mammalian organ of Corti have differences in position, suggest three main possibilities. Morphogenesis. Microtubules are thought to play a ultrastructure and innervation (e.g., Angelborg and Engstrom, 1973; Spoendlin, 1969). Numerous experi- role in the morphogenesis and maintenance of cell ments have suggested that they may have different shape in certain erythrocytes (Behnke, 1970) and in functions (e.g., Kiang et al., 1986; Kim, 1984). Most the development of muscle cells, where the shape is current concepts now assign a more or less passive re- important for alignment of actin filaments (Warren, ceptor role to the IHCs, which send information along 1969, 1974). Heywood et al. (1975) describe microtuthe majority (85--95%) of cochlear afferents. The bules in developing vestibular hair cells, which may OHCs, which are innervated by only 5-15% of affer- partially correlate with those observed here in cochlear ents (Morrison et al., 1975; Spoendlin, 1969), are hair cells even though the cuticular plate does not apthought to contribute largely to cochlear mechanics by pear complete in their micrographs. The “transverse” a n active cycle-by-cycle feedback process of amplifica- microtubules they describe may correspond with ring tion, and are responsible for the sharply tuned response microtubules, the “diagonal” ones with those of the of the cochlea (e.g., Ashmore, 1987; Mountain, 1986). meshwork. Since hair cells in general may have a ring The fact that OHCs, when isolated, are capable of cy- of aligned actin filaments (Hirokawa and Tilney, 1982; clical movements at acoustic frequencies (Ashmore, Zenner, 1986a1, near to the rings of microtubules, it is 1987) supports this possibility. The OHCs also exhibit possible that the distribution of the actin may be afother, slower forms of motility (Flock et al., 1986; Zen- fected during development by the cell shape, which is ner, 1986b; Zenner et al., 19851, a behaviour which in t u r n influenced by the pattern of microtubules. If

MICROTUBULES IN HAIR CELLS

277

and Stebbings, 1979). In hair cells, the sub-apical microtubules may function in the accumulation and maintenance of the population of organelles in the apex. The longitudinally orientated microtubules may transport vesicles produced by the Golgi region, via the track of SER, to the synaptic region, a function similar to that provided by neurotubules in axons (Allen et al., 1985; Wuerker and Kirkpatrick, 1972). There are more microtubules between the Golgi region and the base in IHCs than in OHCs. Since the majority of the afferent innervation is associated with IHCs (Spoendlin, 1969) it is likely that transmitter turnover and release would be greater in these cells, thus requiring a more extensive intracellular transport system (Siege1 and Brownell, 1986). Mechanicalproperties. The hair cells may be very sensitive (Hudspeth, 1985) and might thus respond to a wide range of stimulus energies. This implies a sensitive but well-protected transduction mechanism which is located a t the apex (Hudspeth, 1983). The high concentration of microtubules and other cytoskeletal proteins in this area (see also Flock et al., 1982, and others) suggests a particular involvement of the cytoskeleton in transduction, and the larger number of microtubules in the IHCs implies greater importance of that cytoskeletal function compared with OHCs. The concentration may simply be a structural adaptation to provide mechanical support, or it may function to maximise the sensitivity of the cell. Microtubules are frequently associated with motility. For example, ciliary beating is produced by adjacent parallel microtubules which slide alongside one another by means of the breaking and reforming of ATPase-containing intermicrotubular arms (Satir, 1984). If sliding between the parallel microtubules of the rings in the hair cells were to occur, the diameter of the rings could be increased or reduced. This might result in movement, or changes in the orientation, of apical structures, or in a n alteration of any tension generated in the area by the rings. If the whole cell motility of the isolated OHCs occurs in vivo, special mechanical support and coupling between OHCs and other cells and structures in the organ of Corti may be required. The presence of the concentration of microtubules towards the base of the Fig. 15. Schematic diagram showing the approximate locations OHC may be involved in preventing possible deformaand orientations of different microtubular groups in the outer hair tion of these cells whilst undergoing movements in cells. This diagram is based on a n OHC from the apex of the cochlea and so contains a protrusion (PI from the cuticular plate (Cp) which situ, although at their apices, the cells are probably would not occur in hair cells from basal turns. Microtubules are found held by junctional complexes. The microtubules of the in the apex in similar groups as in IHCs (1-3) and around the prosurrounding supporting cells may also be involved in trusion (4J. In the cell body they are found throughout but are especially concentrated in the region around and below the nucleus (N).In this function. The slow contraction of isolated OHCs, induced by the very base of the cell, there may he two groups, those lining the periphery (5) and those in the centre (6). application of ATP or potassium, may involve a cytoskeletal mechanism (Zenner, 198613; Zenner e t al., 1985). Slow motility based on a microtubular mechathe main function of the microtubules was simply mor- nism is exhibited by another type of sensory cell, the phogenesis, however, then it is interesting that there cone photoreceptor of the teleost retina, which underare more in the apex of IHCs than OHCs. This fact goes elongation and shortening during dark and light implies that some other function may require either adaptation (Warren and Burnside, 1978). The mechathe retention, or addition, of more microtubules in nism which generates the fast motile response of OHCs is unknown, but according to Ashmore (19871, it is IHCs. Intracellular transport. Microtubules are in- probably too rapid to be a cytoskeletal process. Nevervolved in intracellular transport of organelles (Hyams theless the cytoskeleton may indirectly influence the

278

D.N. FURNESS ET AL.

motility by providing a framework against which the generator operates; or it could mediate the effects, if any, which this response may have on the organ of Corti. Attempts to use the microtubule inhibitor, colchicine, on isolated OHCs have failed to show any effects on either motility or cell shape (Zenner, 1986a; Zenner e t al., 1985). Since, however, microtubules of different kinds show differential stability to colchicine (Behnke, 1970; Behnke and Forer, 1967), and since the state of the microtubules was not determined in the colchicinetreated hair cells, no firm conclusion can be drawn about the role of microtubules.

Microtubules in other cells of the organ of Corti The microtubules found in the hair cells are similar in diameter to microtubules present in the majority of eukaryotic cells studied, and also in nerve fibers throughout the organ of Corti where they probably have functions similar to those of neurotubules in general. However, the supporting cells have massive bundles of microtubules composed of 15 protofilaments rather than the normal 13 (Saito and Hama, 1982) which may imply a special function. These are similar to those of the mechanoreceptive neurones of a nematode (Chalfie and Thomson, 1982) where the largesized microtubules are essential for mechanosensitivity and smaller microtubules (11 protofilaments) cannot substitute for them. This suggests a possibility for mechanosensitivity of supporting cells, which also contain actin in anti-parallel arrays (Slepecky and Chamberlain, 1983). Although myosin has not yet been successfully demonstrated in these cells (Slepecky and Chamberlain, 1987), they could, nevertheless, have a more active function than simple mechanical support. It is possible that there may be local self-regulation of stiffness properties residing in the supporting cells, by a type of mechanism integral to the cochlea. There can be little doubt that microtubules play a n important role in both the development and function of the cochlea. They are essential for cell division and important in morphogenesis and maintenance of cell shape, and they can provide a substrate for various types of motility. In these respects, they are certainly important in the supporting cells in which they are heavily concentrated, and in the synaptic and neuronal functions of the nerve fibers, and probably of the hair cells (Favre and Sans, 1983; Favre et al., 1986; Nadol, 1983a,b). Whether they play a significant role in the passive or active mechanical properties of hair cells, however, remains to be established. ACKNOWLEDGMENTS This work is supported by the Wellcome Trust and the MRC. We would like to thank Mr. S. Murray for technical assistance. REFERENCES Allen, R.D., Weiss, D.G., Hayden, J.H., Brown, D.T., Fujikawe, H., and Simpson, M. (1985) Gliding movement of and bidirectional transport along single native microtubules from squid axoplasm: Evidence for an active role of microtubules in cytoplasmic transport. J. Cell Biol., 100:1736-1752. Amos, L.A. (1979) Structure of microtubules. In: Microtubules K. Roberts and J.S. Hyams, eds. Academic Press, London, pp. 1-64.

Angelborg, C., and Engstrom, H. (1972) Supporting elements in the organ of Corti. I. Fibrillar structures in the supporting cells of the organ of Corti of mammals. Acta Otolaryngol. ISuppl.1 (Stockh.), 301:49-60. Angelborg, C., and Engstrom, H. (1973)The normal organ of Corti. In: Basic Mechanisms in Hearing. A. M$ller, ed. Academic Press, London, pp. 125-183. Ashmore, J.F. (1987) A fast motile response in guinea-pig outer hair cells: The cellular basis of the cochlear amplifier. J. Physiol. (Lond.), 388:323-347. Baier, A S . . and MolB-Baier, J. (1972) Spindle dynamics and chromogome movements. Int. Rev. Cytol. [~;pp1.1,3:i-271. Behnke, 0. (1970) A comparative study of microtubules of diskshaped blood cells. J. Ultrastruct. Res., 31:61-75. Behnke, O., and Forer, A. (1967) Evidence for four classes of microtubules in individual cells. J. Cell Sci., 2:169-192. Bredberg, G., Ades, H.W., and Engstrom, H. (1972) Scanning electron microscopy of the normal and pathologically altered organ of Corti. Acta Otolaryngol. ISuppl.1 (Stockh.), 301:3-48. Brownell, W.E., Bader, C.R., Bertrand, D., and de Rihaupierre, Y. (1985) Evoked mechanical responses of isolated cochlear outer hair cells. Science, 227:194-196. Buckley, L., and Stewart, M. (19831 Ciliary hut not saltatory movements are inhibited by vanadate microinjected into living culturedcells. Cell Motil., 3167-184. Chalfie, M., and Thomson, J.N. 11979) Organization of neuronal microtubules in the nematode Caenorhabditis elegans. J. Cell Biol., 82:278-289. Chalfie, M., and Thomson, J.N. (1982) Structural and functional diversity in the neuronal microtubules of Caenorhabditis elegans. J. Cell Biol., 93:15-23. Cleveland, D.W., and Sullivan, K.F. (19851 Molecular biology and genetics of tuhulin. Annu. Rev. Biochem., 54:331-365. Cohen, G.M., and Fermin, C.D. (1978) The development of hair cells in the embryonic chick’s hasilar papilla. Acta Otolaryngol. (Stockh.), 86:342-358. Eichenlaub-Ritter, U., and Tucker, J.B. (1984) Microtubules with more than thirteen protofilaments in the dividing nuclei of ciliates. Nature, 307:60-62. Engstrom, H., and Engstrom, B. (1978) Structure of the hairs on cochlear sensory cells. Hear. Res., 1:49-66. Erickson, H.P. (1974) Microtubule surface lattice and subunit structure and observations on reassembly. J . Cell Biol., 60:153-167. Favre, D., Dememes, D., and Sans, A. (1986) Microtubule organization and synaptogenesis in the vestibular sensory cells. Dev. Brain Res., 25:137-142. Favre. D.. and Sans. A. (1983) Organization and densitv of microtubules in the vestibular sensory cells in the cat. Acta’btolaryngol. (Stockh.). 96:15-20. Flock, A.,Bretscher, A., and Weber, K. (19821 Immunohistochemical localization of several cytoskeletal proteins in inner ear sensory and supporting cells. Hear. Res., 7:75-89. Flock, A.,and Cheung, H.C. (1977) Actin filaments in sensory hairs of the inner ear receptor cells. J . Cell Biol., 75:339-343. Flock, A,, Cheung, H.C., Flock, B., and Utter, G. (1981a)Three sets of actin filaments in sensory cells of the inner ear. Identification and functional orientation determined by gel electrophoresis, immunofluorescence and electron microscopy. J . Neurocytol., 10:133-147. Flock, A,, Flock, B., and Ulfendahl, M. (1986) Mechanisms of movement in outer hair cells and a possible structural basis. Arch. Otorhinolaryngol., 24333-90. Flock, A,, Hoppe, Y., and Wei, X. (1981b) Immunofluorescence localization of proteins in semithin 0.2-1 pm frozen sections of the ear. Arch. Otorhinolaryngol., 23355-66. Freed, J.J., and Leibowitz, M.M. (1970) The association of a class of saltatory movements with microtubules in cultured cells. J. Cell Biol., 45:334-354. Friedmann, I., Cawthorne, T., and Bird, E.S. (1965) The laminated cytoplasmic inclusions in the sensory epithelium of the human macula. J. Ultrastruct. Res., 12:92-103. Furness, D.N., and Hackney, C.M. (1985) Cross-links between stereocilia in the guinea pig cochlea. Hear. Res., 18:177-188. Gibbons, I.R. (1981) Cilia and flagella of eukaryotes. J. Cell Biol., 91:107s124s. Griffth, L.M., and Pollard, T.D. (1978) Evidence for actin filamentmicrotubule interaction mediated by microtubule-associated proteins. J. Cell Biol., 78:958-965. Gull, K., Hussey, D.J., Sasse, R., Schneider, A,, Seeheck, T., and Sher-

MICROTUBULES IN HAIR CELLS win, T. (1985) Tubulin isotypes: Generation of diversity in cells and microtubular organelles. J. Cell Sci. I Suppl.], 5243-255. Heywood, P., Van de Water, T., Hilding, D.A., and Ruben, J . (1975) Distribution of microtubules and microfilaments in developing vestibular sensory epithelium of mouse otocysts grown in uitro. J. Cell Sci., 17:171-189. Hirokawa, N., and Tilney, L.G. (1982) Interactions between actin filaments and between actin filaments and membranes in quickfrozen and deeply etched hair cells of chick ear. J. Cell Biol., 95: 249-261. Hudspeth, A.J. (1983) Mechanoelectrical transduction by hair cells in the acousticolateralis sensory system. Annu. Rev. Neurosci., 6:187215. Hudspeth, A.J. (1985) The cellular basis of hearing: The biophysics of hair cells. Science, 230:745-752. Hyams, J.S., and Stehbings, H. (1979) Microtubule associated cytoplasmic transport. In: Microtubules. K. Roberts and J.S. Hyams, eds. Academic Press, London, pp. 487-530. Inoue, S., and Sato, H. (1967) Cell motility by labile association of molecules: The nature of mitotic spindle fibers and their role in chromosome movement. J. Gen. Physiol., 50:259-292. JZrgensen, J.M. (1982) Microtubules and laminated structures in inner ear hair cells. Acta Otolaryngol. (Stockh.), 94:241-248. Kiang, N.Y.S., Liberman, M.C., Sewell, W.F., andGuinan, J.J. (1986) Single unit clues to cochlear mechanisms. Hear. Res., 22:171-182. Kim, D.O. (1984) Functional roles of the inner- and outer-hair-cell subsystems in the cochlea and brainstem. In: Hearing Sciences. C.I. Berlin, ed. College Hill Press, San Diego, pp. 241-262. Kimura, R.S. (1966) Hairs of the cochlear sensory cells and their attachment to the tectorial membrane. Acta Otolaryngol. (Stockh.), 61:55-72. Koonce, M.P., Tong, J., Euteneur, U., and Schliwa, M. (1987) Active sliding between cytoplasmic microtubules. Nature, 328:737-739. Lim. D.J. (1986) Functional structure of the orean of Corti: A review. Hear. Res., 22:117-146. Lubitz. D.K.J.. Ekstrom von (1981a) Ultrastructure of lateral line sense organs of the ratfish, Chimaera monstrosa. Cell Tissue Res., 215:651-665. Lubitz, D.K.J., Ekstrom von (1981b)Subsurface tubular system in the outer sensory cells of the rat cochlea. Cell Tissue Res., 220:787-795. Luduena, R.F. (1979) Biochemistry of tubulin. In: Microtubules. K. Roberts and J.S. Hyams, eds. Academic Press, London, pp. 65-116. Margolis, R.L., and Wilson, L. (1981) Microtubule treadmills-possible molecular machinery. Nature, 293:705-711. Morrison, D., Schindler, R.A., and Wersall, J . (1975) A quantitative analysis of the afferent innervation of the organ of Corti in guinea pig. Acta Otolaryngol. (Stockh.), 79:ll-23. Mountain, D.C. (1986) Electromechanical properties of hair cells. In: Neurobiology of Hearing: The Cochlea. R.A. Altschuler, D.W. Hoffmann, and R.P. Bobbin, eds. Raven Press, New York, pp. 77-90. Nadol, J.B. (1983a) Serial section reconstruction of the neural poles of hair cells in the human organ of Corti. I. Inner hair cells. Laryngoscope, 93599-614. Nadol, J.B. (198313)Serial section reconstruction of the neural poles of hair cells in the human organ of Corti. 11. Outer hair cells. Laryngoscope, 93:780-791. Pollard, T.D., Selden, S.C., and Maupin, P. (1984) Interaction of actin filaments with microtubules. J . Cell Biol., 99:33~-37s. Raphael, Y., and Wroblewski, R. (1986) Linkage of sub-membranecisterns with the cytoskeleton and the plasma membrane in cochlear outer hair cells. Submicrosc. Cytol., 18:731-737. Y

279

Saito, K., and Hama, K. (1982)Structural diversity of microtubules in supporting cells of the sensory epithelium of the guinea pig organ of Corti. J . Electron Microsc. (Tokyo), 31:278-281. Satir, P. (1984) The generation of ciliary motion. J. Protozool., 31: 8-12. Siegel, J.H., and Brownell, W.E. (1986) Synaptic and Golgi membrane recycling in cochlear hair cells. J . Neurocytol., 15:311-328. Slepecky, N., and Chamberlain, S.C. (1983) Distribution and polarity of actin in inner ear supporting cells. Hear. Res., 10:359-370. Slepecky, N., and Chamberlain, S.C. (1985) Immunoelectron microscopic and immunofluorescent localization of cytoskeletal and muscle-like contractile proteins in inner ear sensory hair cells. Hear. Res., 20245-260. Slepecky, N., and Chamberlain, S.C. (1987) Tropomyosin co-localizes with actin microfilaments and microtubules within supporting cells of the inner ear. Cell Tissue Res., 248:63-66. Slepecky, N., Hamernik, R., and Henderson, D. (1981) The consistent occurrence of a striated organelle (Friedmann body) in the inner hair cells of the normal chinchilla. Acta Otolaryngol. (Stockh.), 91: 189-198. Slepecky, N., Ulfendahl, M., and Flock, A. (1988) Effects of caffeine and tetracaine on outer hair cell shortening suggest intracellular calcium involvement. Hear. Res., 32:ll-22. Spoendlin, H. (1969) Innervation patterns in the organ of Corti of the cat. Acta Otolaryngol. (Stockh.), 67:239-254. Steyger, P.S., Furness, D.N., Hackney, C.M., and Richardson, G.P. (1988)The distribution of cytoskeletal proteins in the organ of Corti demonstrated by immunocytochemistry. Br. J. Audiol., 22:126,127. Thorne, P.R., Carlisle, L., Zajic, G., Schacht, J., and Altschuler, R.A. (1987) Differences in the distribution of F-actin in outer hair cells along the organ of Corti. Hear. Res., 30:253-266. Tilney, L.G., DeRosier, D.J., and Mulroy, M.J. (1980) The organization of actin filaments in the stereocilia of cochlear hair cells. J. Cell Biol., 86:244-259. Travis, J.L., and Bowser, S.S. (1986) A new model of reticulopodial motility and shape: Evidence for a microtubule based motor and an actin skeleton. Cell Motil., 6:2-14. Tucker, J.B. (1979) Spatial organization of microtubules. In: Microtubules. K. Roberts and J.S. Hyams, eds. Academic Press, London, pp. 315-357. Vale, R.D., Reese, T.S., and Sheetz, M.P. (1985) Identification of a novel force-generating protein, kinesin, involved in microtubulebased motility. Cell, 42:39-50. Warren, R.H. (1969) The effect of colchicine on myogenesis in vivo in Rann pipiens and Rhodnius prolixus (Hemiptera).J . Cell Biol., 39: 544-555. Warren, R.H. (1974) Microtubule organization in elongating myogenic cells. J . Cell Biol., 63550-566. Warren, R.H., and Burnside, B. (1978) Microtubules in cone myoid elongation in the teleost retina. J. Cell Biol., 78:247-259. Wuerker, B.R., and Kirkpatrick, J.B. (1972) Neuronal microtubules, neurofilaments and microfilaments. Int. Rev. Cytol., 33:45-75. Zenner, H.P. (1981) Cytoskeletal and muscle-like elements in cochlear hair cells. Arch. Otorhinolaryngol., 230:81-92. Zenner, H.P. (1986a) Molecular structure of hair cells. In: Neurobiology of Hearing: The Cochlea. R.A. Altschuler, D.W. Hoffmann, and R.P. Bobbin, eds. Raven Press, New York, pp. 1-21. Zenner, H.P. (1986b) Motile responses in outer hair cells. Hear. Res., 22183-90. Zenner, H.P., Zimmerman, U., and Schmitt, U. (1985) Reversible contraction of isolated mammalian cochlear hair cells. Hear. Res., 18: 127-133.