Motile responses in outer hair cells

Hearing Research, 22 (1986) 83-90 Elsevier

Motile responses in outer hair cells Hans Peter Zenner Department

of Otolatyngology,

University of Wivzburg, KopJklinikum, D-8700 Wtizburg,

F. R.G.

Motile responses of cochlear hair cells open new perspectives for the understanding of cochlear hearing mechanisms and hearing disorders located in hair cells. Direct visualization of hair cell motility was achieved by a method for the study of living isolated mammalian outer hair cells (OHCs) which has overcome some of the complexities in dealing with the heterogenous organ of Corti. Electrophysiological giga-seal whole-cell recordings of single OHC prepared by this approach had revealed negative cell potentials ranging from -32 mV to -70 mV (Gitter et al. (1986) Oto-Rhino-Laryngol. in press). Elucidation of HC motility has come from two lines of experiments. One follows from the observation that exposure of the lateral and basal membrane parts of living OHCs to increasing bath K+ concentrations resulted in a sustained reversible depolarization of the cell. Here, we report that by depolarization of the cell membrane in the presence of 25-125 mM K+/Cla sustained contraction of OHC was induced. This was followed by relaxation in the presence of artificial perilymph containing 5.4 mM K+/Cl-. By alternating these procedures OHCs were made to undergo as many as five cycles of contraction and relaxation. External Ca2+ was not required for the initial contraction but was essential for relaxation. Following repeated contraction/relaxation cycles the cytoplasm of individual OHCs exhibited a filamentous network, correlating with a new infracuticular anti-actin binding capacity. The second series of experiments originates in the observation that permeabilized OHCs contracted in the presence of ATP. No response was seen in the presence of control nucleotides. ATP-induced contraction was inhibited by cytochalasin B, indicating the presence of an actin-dependent motile mechanism. Studies using fluorescent labelled phalloidin (Zenner and Drenckhahn, in prep.) allowed monitoring of an actin-polymerizing process during ATP-induced contraction of OHCs.

Introduction The concept that outer hair cells of the cochlea possess an active motor capacity confers new dimensions on the models of the physiology and pathophysiology of the inner ear. This concept is essentially based on three different elements. Firstly, actin and actin-associated molecules such as myosin, alpha-actinin, fimbrin and tropomyosin have been discovered in the hair cells of numerous species and their morphology characterized in detail (Flock and Cheung, 1977; Zenner, 1980; Tilney et al., 1980; Drenckhahn et al., 1982, 1985). These proteins are responsible for contractile phenomena both in muscle cells and in nonmuscle cells. Secondly, an active process located in the outer hair cells is postulated. This active process is alleged to explain phenomena such as stimulated and spontaneous acoustic emissions, the unusually sharp frequency selectivity of the basilar membrane, the alteration of the elasticity of the cochlear 0378-5955/86/$03.50

0 1986 Elsevier Science Publishers

partition as well as the more intense vibrations of the basilar membrane near the auditory threshold (negative damping) compared to linear calculations (Gold, 1948; Khanna and Leonhard, 1982; LePage and Johnstone, 1980; Sellick et al., 1982; Dallos, 1981; Kemp, 1978; Kemp and Brown, 1983; Davis, 1983; Kim et al., 1980). Thirdly, in the presence of calcium ions and ATP an alteration in the deflectability of stereocilia was observed in the ampullary crista of the frog, which suggests an analogy to a mechanism resembling muscle (Orman and Flock, 1983; Flock and Strelioff, 1984). Recently, we were able to show under direct visual control on living, isolated outer hair cells of the mammal that these depolarized (Zenner et al., 1985) in the presence of endolymphatic potassium concentrations and had the capacity for repeated reversible longitudinal contraction of the entire cell (Zenner, 1985a). Furthermore, following electrical stimulation, a longitudinal contraction of outer hair cells was observed (Brownell et al., 1985).

B.V. (Biomedical

Division)

84

In the present paper, results are presented which suggest the existence of motor properties due to actin in the outer hair cells of the guinea pig (Zenner, 1984; Flock et al., 1984). (i) Permeabilized isolated outer hair cells contract concentration-dependently in the presence of ATP and calcium. Besides this, a movement of the cuticular plate, but not of the stereociliary bundle can be observed. Contraction is inhibited by cytochalasin as well as by bisphosphate. (ii) The movement seen depends on the free calcium concentration. In the hair cell, the existence of the calcium-binding protein calmodulin can be demonstrated. The substance TFP, which inactivates calmodulin, inhibits the contractile response of the hair cell. (iii) By means of monoclonal antibodies against actin as well as by means of rhodamine-phalloidin binding selectively to actin filaments (Zenner and Drenckhahn, in preparation), the existence of a novel infracuticular actin network can be directly seen.

(iv) Furthermore, it can be shown that the motility of the outer hair cells is not only of contractile nature, but is also associated with polymerization of actin molecules (Zenner and Drenckhahn, in preparation). ATP-induced motility of the entire hair cell Isolated, living outer hair cells of the guinea pig cochlea (Zenner et al., 1985) were permeabilized in the presence of Triton X-100 which largely removes the outer cell membrane (Fig. la) but leaves the cytoskeleton of the cell intact. This allows access of external ATP and calcium to the cytoplasm. This elicits a visible contraction of the hair cell along its longitudinal axis. The degree and rate of contraction depend on the ATP concentration and require micromolar calcium concentrations (Fig. lb).

Fig. 1. Motility of permeabilized isolated OHCs. A living isolated OHC from a guinea pig cochlea is demembranated (a) in the presence of Triton X-100. Following addition of ATP (b) a contraction of the whole cell can be visualized. Furthermore, photographs reveal a tilting movement of the cuticular plate ( + ).

85

Conditions for a motile response

The nucleotide dependence of contraction was tested with control substances. Bisphosphate, which competes with ATP for the specific binding site, but which cannot provide energy, leads to an inhibition of contraction. A nucleotide-free medium was unable to induce a motor response. Since the experiments described primarily suggest that the muscle protein actin might be responsible for the contraction process, this hypothesis was further investigated. The toxin cytochalasin inhibits the polymerization of actin filaments and thus inhibits actin-dependent contraction processes. In the presence of cytochalasin, the ATP-induced contractile process of permeabilized hair cells could be effectively suppressed. Calcium and calmodulin

cochlear hair cells, however, which we used allowed the complete cytoskeleton to be visualized in a single preparation (Zenner, 1983; Zenner et al., 1985). By means of monoclonal antibodies, besides the actin in the cuticular plate and stereocilia the immunofluorescence showed up an additional actin network which (Figs. 2, 4) extends from the lower side of the cuticular plate to the caudal end of the hair cell. When the cell is focussed under the microscope (cannot be represented photographically) parts of this network appear to be located near to the cell membrane. In the meantime, electron microscopic investigations have yielded similar results (Flock et al., 1984). Furthermore, the whole-cell investigation technique enables direct three-dimensional visualization of an actin ring running circularly around the central region of the cuticular plate (Zenner et al., 1985), as was postulated on the basis of electron

In the absence of calcium (adjusted with EGTA), ATP was no longer able to elicit a motor response of the hair cell. A modulation of the intracellular calcium level must hence have a signal effect for elicitation of a mechanical hair cell response. As a possible intracellular binding site, the calcium”bin~ng protein calmodulin could be localized in outer hair cells by means of an immunofluorescent antibody. In order to elucidate further a possible role of calmodulin as modulator protein in the calcium- and ATP-dependent contractile response of the hair cells, the sensory cell was treated with TFP, which inhibits calmodulin. This led to an inhibition of the contractile response. An additional infra-cuticular actin network

Apart from in our laboratory (Zenner, 1980), in the past actin could be detected only in the region of the cuticular plate and stereocilia of the hair cells (Flock and Cheung, 1977; Tilney et al., 1980; Drenckhahn et al., 1982, 1985). The infracuticular space, in which the observed contraction essentially occurs, seemed to be free of actin. If this concept were correct, then it would refute the assumption of an actin-dependent contractile capacity of hair cells along their longitudinal axis described above. The whole-cell preparation of

Fig. 2. Immunost~~ng of whole outer hair cells with antiactin antibodies. The antibodies stain in an isolated OHC from the guinea pig stereocilia, cuticular plate and an additiohal (+ +) infracuticular network. +, Actin near the cell membrane. Insert: circular arrangement of actin surrounding the cuticular plate. ( + + + , ‘contractile ring’) in another cell (Zenner et al., 1985).

86

Fig. 3. Single-frame analysis of the contraction of a living OHC during initiation of a motile response. Visualization of a longitudinal contraction

K+ intoxication. (a) Just before stimulation; (b) 60 s after (Zenner, 1985) and a tilting movement of the cuticular plate.

1C-a

IC -b -d

-e

Fig. 4. Model of the actin skeleton in auditory outer hair cells (based on data from Flock and Cheung, 1977; Zenner, 1980, 1984; Tilney et al., 1980; Drenckhahn et al., 1982, 1985; Flock

microscopic studies (Flock et al., 1981; Hirokawa and Tilney, 1982). Control experiments with rhodamine-labelled phalloidin which is able to bind selectively to polymerized actin, provided an identical pattern of an infracuticularly situated actin filament network (Zenner and Drenckhahn, in preparation).

et al., 1984). (a) Stiff actin bundles in the stereocilia are rigidly held together in parallel register by fimbrin. (b,c) Actin filaments and myosin in a random three-dimensional actin network in the cuticular plate (b) and in an actin ring surrounding it (c). a-Actinin crosslinks actin filaments. An infracuticular actin network extends from the cuticular plate downwards forming a central (d) and a peripheral (e) network. Myosin can be found adjacent to the cortical actin network.

87

Tbe possible role of myosin

PoIyme~~a~on of actin

An actin-mediated contraction requires myosin as a molecular partner. The existence of myosin in hair cells was demonstrated immunohistochemitally by Drenckhahn et al. (1982) in the region of the cuticular plate as well as near the cell membrane extending in direction of the cell base. Besides actin in our experiments micro-SDS gel electrophoreses of isolated, outer hair cells from the cochlea of the guinea pig reveal also proteins of molecular weights of about 2~~ and about 20000 which comigrate with the heavy chains and with the light chains of myosin. In addition, the competitive inhibition of ATP contraction by bisphosphate suggests the involvement of the corresponding ATP binding sites of the myosin light chain kinase. The i~bition of contraction by the calmodulin inhibitor TFP also indicates the possible calcium regulation of a myosin light chain kinase. Furthermore, the rate of contraction of hair cells of 500 rim/s (Zenner, 1984) measured after 10 s by means of a video-assisted system corresponds to the lower ovation rates of myosin microspheres on actin filaments as observed by Sheetz and Spudich (1983). Thus, the experiments described make it probable that myosin might be involved in the process of #ntra~tion observed in hair cells, but this is not yet definitively proved.

As the result of studies with anti-actin antibodies on hair cells in 5-10 pm thick sections of the guinea pig cochlea, a pool of depolymerized actin was suggested to be present in both the cuticular plate and in the region between the cuticular plate and the nucleus (Zenner, 1980). It is a fundamental property of actin filaments that they are unstable structures which can be built up from such a pool of individual subunits and can also be depolymerized once more. In cooperation with D. Drenckhahn, the rhodamine-labelled phalloidin in the whole-cell preparation of the hair cell allowed us (Zenner and Drenckhahn, in preparation) to examine the polymerization state of actin before and after a contraction It was shown that after an ATP-induced contraction of the cell body, the contracted hair cell displayed a marked enhancement of rhodamine-phalloidin fluorescence. In the presence of an excess of unlabeled phalloidin, this fluorescence is no longer visible. Since rhod~ne-phalloidin binds selectively to polymerized actin, this observation allows the conclusion that during the contraction process of the hair cell not only an actin-dependent contraction process, but at the same time a polymerization process is induced. In addition under ~ght-scatte~ng conditions, singleframe analysis of contracting outer hair cells revealed a cytoplasm& flow in an apico-basal direction in occasional cases.

The possible role of IP, as second messenger

Investigations done with J. Schacht (Zenner and Schacht, in preparation; see also Schacht, 1986) could demonstrate that permeabilized isolated OHCs contracted in a calcium-free ([Ca”] < 10e6 M) medium in the presence of IP, (inositoltrisphosphate) and ATP. No motile response was seen when 1s was replaced by IP or inositol, suggesting a specific role for IP,. IP, might act as a second messenger by releasing Ca2’ from intracellular Ca2+-binding sites, which increases the intracellular Ca” level. The rise of [Ca”] is followed by the motile response of the OHC. Thus, the experiments suggest a possible key role for IP, as second messenger during mechanical stimulation of an OHC.

Potassium-induced ceil

contraction

of the entire hair

As the third form of the motility of hair cells, we were able to observe a response to intoxication with endolymphatic potassium concentrations (Zenner, 1985a). Such intoxications are discussed for example in the pathogenesis of M&G&e’s disease (Jahnke, 1980). After incubation of the intact living hair cell in the presence of 25-125 mM potassium chloride or potassium gluconate, a slower but compared to ATP contraction more pronounced contraction of the sensory cell can be observed directly (Fig. 3). Electrophysiological controls in the giga seal whole cell recording configuration carried out to-

88 gether with A. Gitter and E. Fromter (Zenner et al., 1985) were able to show that the potassium intoxication is associated with a simultaneous measurable depolarization of the hair cell (genuine voltage 70 mV). Following a decrease of the potassium concentration to 5.4 mM within 60 s, the outer hair cell relaxed. Under in vitro conditions, an individual living hair cell could undergo five contractions and five relaxations.

b

Motility of the cuticular plate When the hair cell is placed directly on a lens by means of a 500 pm thick foil as well as immersion oil, optical refraction effects in the region of the cuticular plate and stereocilia can be utilized for observation at more than 1 OOO-fold magnification under light microscopic conditions. If singleframe analyses are performed, two different motile mechanisms can be often observed in the cuticular plate of an outer hair cell. They are only manifested in the presence of contraction conditions for the entire cell, predominantly in the presence of high [K+]. In observations along the axis of symmetry of the cuticular plate, a teasing out of the outermost edge of the cuticular plate to caudal becomes visible. This process (Fig. 5a) appears to be associated with a folding up of the upper side of the cuticular plate to cranial. In lateral observation of the hair cell, a tilting movement to caudal of the part of the cuticular plate oriented to the modiolus can be visualized although (Fig. 5b) the angle of the stereociliary bundle to the cuticular plate does not change. Consistent with this are investigations on the arrangement of actin in the microfilaments of the stereocilia of the alligator lizard, which does not change in the presence of magnesium, calcium and ATP (Tilney et al., 1980). In addition, however, in our investigation, the bundle of stereocilia can be seen to be moved as a whole with the movement of the cuticular plate. In our experiments the microscopic resolution did not allow distinction of individual stereocilia. Since the apical surface of the living hair cell displays a concavity parallel to the axis of symmetry of the cell (Zenner, 1985b; Hudspeth, 1983) it was assumed physiologically that stereocilia are pushed together along the axes of symmetry. The

Fig. 5. Motility of the cuticular plate in outer hair cells. In observations along the axis of symmetry (a) the cuticular plate shows a bulging down of the outermost edge of the cuticular plate. This process appears to be associated with a folding up of the upper side of the cuticular plate. In lateral observation of the hair cell (b), a tilting movement to caudal of the part of the cuticular plate oriented to the modiolus can be visualized, although the angle of the stereociliary bundle to the cuticular plate does not change. However, the bundle of stereocilia can be seen to be moved as a whole with the movement of the cuticular plate. Observations (Zenner. 1984, 1985a) were made with planapochromatic bright field or phase contrast optics (Leitz) supplemented by a real time video system (Philips) or an electronically controlled film camera (Bauer) with 16 fps to 4 fpm. This set up allowed time dependent measurements of cell length and width with a resolution of 0.4-0.8 pm under real-time and time-lapse conditions as well as single-frame analysis.

observed alteration of the concavity of the cuticular plate should therefore lead to an alteration of the spatial arrangement of the stereocilia and thus of their mechanical properties (Fig. 5a). This might be involved in the process of the transduction system (Zenner, 1985b; Hudspeth, 1983). Implications Outer hair cells of the are evidently capable of ments can be induced 1985a), electric stimulation as well as by ATP and Flock et al.. 1984).

cochlea of the guinea pig movement. These moveby potassium (Zenner, (Brownell et al., 1985) calcium (Zenner, 1984;

89

The remarkable degree of contraction of outer hair cells was observed under cell culture conditions. In vivo, a contraction of the hair cells may be prevented by numerous adjacent structures, so that an isometric contraction with a change of hair cell stiffness results. In contrast to other epithelial cells, however, outer hair cells are surrounded by large extracellular spaces so that they are only fixed at the cell base and the cell apex. This might allow outer hair cells a certain free motility even in vivo. The molecular mechanisms underlying both the potassium-induced and the electrically elicited contractile response have not been elucidated. The observed contraction of outer hair cells in the presence of increased potassium concentration can be regarded primarily as a pathophysiological step during into~cation of the pe~lymphatic space (Zenner, 198%). Such a process is discussed for example for Meniere’s disease (Jahnke, 1980). In contrast, evidence can be provided suggesting that ATP- and calcium-induced motile responses of hair cells are produced by actin together with myosin. This process appears to be mediated by a rise of the intracellular calcium level which is modulated by calmodulin. The presence of calmodulin, actin, myosin as well as the involvement of ATP and calcium have been demonstrated in hair cells, but the proof of the interaction of these molecular components during a contractile process of the hair cell is still lacking. The ATP- and calcium-induced contractions of permeabilized isolated outer hair cells suggest a physiological mechanism. Accordingly, outer hair cells must be peripheral effector cells. The hypothetical model comprises an influence of OHC motility on the micromechanics of the basilar membrane and the stereociliary region of outer hair cells in order to attain an adaptation to high sound pressure. Moreover, control of the damping characteristics of the basilar membrane appears reasonable. The possibility of an involvement in the postulated active process of the inner ear near auditory threshold with an influence on the inner hair cells must also be included in the discussion. Control of OHC motility could be achieved by the efferent nerves as well as by a direct regulation by the micromechanics of the basilar membrane via the deflection of the stereocilia of the outer hair

cells. However, knowledge of the motility of outer hair cells does not yet appear sufficient to establish a final conclusive model. Summary

The concept that outer hair cells of the cochlea possess an active motor capacity confers new dimensions on the models of the physiology and pathophysiology of the inner ear. Experiments are reported which suggest the existence of motor properties due to actin in the outer hair cells of the guinea pig. (i) Permeabilized isolated outer hair cells contract concentration-dependently in the presence of ATP and calcium. Besides this, a movement of the cuticular plate, but not of the stereociliary bundle can be observed. Contraction is in~bited by cytoch~asin as well as by bisphosphate. (ii) The movement seen depends on the free calcium concentration. In the hair cell, the existence of the calcium-binding protein calmodulin can be demonstrated. The substance TFP, which inactivates calmodulin, inhibits the contractile response of the hair cell. (iii) By means of monoclonal antibodies against actin as well as by means of rhodamine-phalloidin binding selectively to actin filaments, the existence of a novel infracuticular actin network can be directly seen. (iv) Furthermore, it can be shown that the motility of the outer hair cell is not only of contractile nature, but is also associated with polymerization of actin molecules. Thus, OHCs are evidently capable of movement. The ATP- and calcium-induced contractions of pe~eabi~zed isolated outer hair cells suggest a physiological actin-dependent mechanism. Accordingly, outer hair cells must be peripheral effector cells. The hypothetical model comprises an influence of OHC motility on the micromechanics of the basilar membrane and the stereociliary region of outer hair cells: (i) in order to attain an adaptation to high sound pressure; (ii) moreover, control of the damping characteristics of the basilar membrane appears reasonable.

The expert technical assistance of Ursula Schmitt, Ulrike Zimmermann and Barbara

90

Motschenbacher is gratefully acknowledged. This work was supported by D.F.G. Grant Ze 149/2-2. References Brownell, W.E., Bader, C.R., Bertrand, D. and de Ribaupierre, Y. (1985): Evoked mechanical responses of isolated cochlear outer hair cells. Science 227, 194-196. Dallos, P. (1981): Cochlear physiology. Annu. Rev. Physiol. 32, 153-190. Davis, H. (1983): An active process in cochlear mechanics. Hearing Res. 9, 79-90. Drenckhahn, D., Kellner, J., Mannherz, H.G., Groschel-Steward, U., Kendrick-Jones, J. and Scholey, J. (1982): Absence of myosin-like immuno-reactivity in stereocilia of cochlear hair cells. Nature (London) 300, 531-532. Drenckhahn, D., Schafer, T. and Prinz, M. (1985): Actin, myosin and associated proteins in the vertebrate auditory and vestibular organs. In: Auditory Biochemistry. Editor: D. Drescher. Academic Press, New York (in press). Flock, A and Cheung, H. (1977): Actin filaments in sensory hairs of inner ear receptor cells. J. Cell. Biol. 75, 339-343. Flock, A. and Strelioff, D. (1984): Graded and non linear mechanical properties of sensory hairs in the mammalian hearing organ. Nature (London) 310, 597-599. Flock, A., Cheung, H.C., Flock, B. and Utter, G. (1981): Three sets of actin filaments in sensory cells of the inner ear. J. Neurocytol. 10, 133-147. Flock, A., Flock, B. and Ulfendahl, M. (1984): A new view of hearing. Acta Physiol. Stand. 121, C256 (Abstr.). Gold, T. (1948): Hearing. The physical bases of the action of the cochlea. Proc. R. Sot. Edinb. B 135, 492-498. Hirokawa, N. and Tilney, L. (1982): Interactions between actin filaments and between actin filaments and membrane in quickly frozen and deep etched hair cells of the chick ear. J. Cell Biol. 95, 249-261. Hudspeth, A.J. (1983): Mechanoelectrical transduction by hair cells in the acoustico-lateralis system. Annu. Rev. Neurosci. 6, 187-215. Jahnke, K. (1980): Elektronenmikroskopie in Forschung und Diagnose des HNO-Fachgebietes. Arch. Otorhinolaryngol. 235, 484-498. Kemp, D.T. (1978): Stimulated acoustic emission from within the human auditory system. J. Acoust. Sot. Am. 64, 138661391. Kemp, D.T. and A.M. Brown (1983): A comparison of mechanical nonlinearities in the cochleae of man and gerbil from ear canal measurements. In: Hearing - Physiological Bases and Psychophysics, p. 82. Editors: R. Klinke and R. Hartmann. Springer-Verlag, Berlin.

Khanna, S.M. and Leonhard, D.G.B. (1982): Laser interferometric measurements of basilar membrane vibrations in cats. Science 215, 305-306. Kim, D.O., Neeley, S.T., Molnar, C.E. and Matthews, J.W. (1980): In: Psychophysical, Physiological and Behavioural Studies in Hearing, pp. 7-14. Editors: G. van den Brink and F.A. Bilsen. Delft University Press, Delft, The Netherlands. LePage, E.W. and Johnstone, B.M. (1980): Non-linear mechanical behaviour of the basilar membrane in the basal turn of the guinea pig cochlea. Hearing Res. 2, 183-189. Orman, S. and Flock, A. (1983): Active control of sensory hair mechanics implied by susceptibility to media that induce contraction in muscle. Hearing Res. 11, 261-266. Schacht, J. (1986): Molecular mechanisms of drug-induced hearing loss. Hearing Res. 22, 297-304. Sellick, P.M., Patuzzi, R. and Johnstone, B.M. (1982): Measurements of basilar membrane motion in the guinea pig using the Mi%bauer technique. J. Acoust. Sot. Am. 72, 131-141. Sheetz, M.P. and Spudich, J.A. (1983): Movement of myosincoated fluorescent beads on actin cables in vitro. Nature (London) 303, 31-35. Tilney, L.C., DeRosier, D.J. and Mulroy, M.J. (1980): The organization of actin filaments in the stereocilia of co&ear hair cells. J. Cell Biol. 86, 244259. Zenner, H.P. (1980): Cytoskeletal and muscle like elements in cochlear hair cells. Arch. Otorhinolaryngol. 230, 82-92. Zenner, H.P. (1983): Biochemical approaches to single outer hair cells. In: Cochlear Research, pp. 17-21. Editors: L.P. Lobe and P. Lotz. Halle (Saale) University Press, Halle (Saale). Zenner, H.P. (1984): Contractility of isolated hair cells from the guinea pig cochlea. In: Abstr. Inner Ear Biol., XXI Workshop, Taormina, 1984, p. 56. Zenner, H.P. (1986): Molecular structure of hair cells. In: Neurobiology of Hearing: The Cochlea. Editors: D.E. Hoffmann, R.A. Altschuler and R.P. Bobbin. Raven Press, New York (in press). Zenner, H.P. (1986): K’-induced motility and depolarization of cochlear hair cells - direct evidence for new pathophysio logical steps in Met&e’s disease. Arch. Otorhinolaryngol. (in press). Zenner, H.P., Gitter, A., Zimmermann, U., Schmitt, U. and Fromter, E. (1985a): Die isolierte, lebende Haarzelle - ein neues Model1 zur Untersuchung der Hbrfunktion. Laryngol. Rhinol. Otol. 64, 642-648. Zenner, H.P., Zimmermann, U. and Schmitt, U. (1985b): Reversible contraction of isolated mammalian cochlear hair cells. Hearing Res. 18, 127-133.