- Email: [email protected]
Immunoreactivity for taurine in the cochlea: its abundance in supporting cells
Hearing Research 109 (1997) 135^142
Immunoreactivity for taurine in the cochlea: its abundance in supporting cells Kathleen C. Horner *, Catherine Aurousseau è rimentale, Universite è Bordeaux II, Bordeaux, France INSERM, Laboratoire d'Audiologie Expe
Received 1 October 1996; revised 31 March 1997; accepted 5 April 1997
Abstract
Taurine is the second most abundant free amino acid in the brain where its osmoregulatory function is well established. Taurinedeprived kittens show retinal pathology leading to blindness. In the inner ear, taurine has been reported to be the most abundant free amino acid although its role in inner ear function is not known. Immunohistochemistry was employed here to investigate the localisation of taurine in normal cochleae of the guinea pig compared with two different conditions: experimentally induced endolymphatic hydrops and after oral administration of glycerol. In normal cochleae, by light microscopy, taurine-like immunoreaction was never observed in the sensory outer hair cells and appeared absent from the inner hair cells. In contrast taurine-like immunolabeling was found to be present in all supporting tissue with the striking exception of the tectorial membrane and the outer pillar cell which had no or little taurine immunoreactivity respectively. In early experimental endolymphatic hydrops, the distribution of taurine-like immunoreactivity appeared similar to that observed for normal cochleae. In long-term hydrops, degenerated outer hair cells were replaced by the swelling of the phalangeal process of the Deiters' cells which became highly immunoreactive to taurine. After glycerol administration, the tectorial membrane became more tightly bound to the apical surface of the sensory hair cells and distinctly immunoreactive to taurine. The localisation of taurine in the organ of Corti shown here is consistent with taurine being involved in the maintenance of osmotic equilibrium in the normal and perhaps also in the restructuration of the pathological organ of Corti. Keywords :
Cochlea; Hair cell; Tectorial membrane; Supporting cell; Guinea pig
1. Introduction
Taurine (2-aminoethane sulfonic acid) is one of the most abundant free amino acids in mammals and in brain tissue and is the most abundant free amino acid after glutamate (Huxtable, 1992). Various functions have been attributed to taurine although its role in osmoregulation appears to be its major function. The rise in extracellular potassium or the decrease in osmolarity tend to cause cell swelling which triggers the release of
* Corresponding author. Laboratoire d'Audiologie Expeèrimentale, Universiteè Bordeaux II, Hoêpital Pellegrin, Place Ameèlie Raba Leèon, 33076 Bordeaux, France. Tel.: (33) 5-56-24-20-47; Fax: (33) 5-56-9629-84. New address after 15 July 1997: Laboratoire d'Otologie and Neuro-Otologie, Faculteè de Medecine Nord, Boulevard Pierre Dramard, 13916 Marseille Cedex 20, France.
taurine from various cell types in culture as part of the normal cell volume regulation mechanism. Following intraperitoneal injection of distilled water in rats, Nagelhus et al. (1993) observed the delocalisation of taurine from cerebellar Purkinje cells into neighboring Golgi epithelial cells. Those authors considered that the cellular redistribution might represent a rapid adjustment to osmotic perturbation in vivo with a higher priority for neuronal compared with glial volume regulation. Taurine is known to be essential for retinal function (Lombardini, 1991) and taurine deprivation is reported to result in retinal degeneration in monkeys (Imaki et al., 1993), rats (Hageman and Schmidt, 1987) and cats (Hayes et al., 1975). Taurine has been shown to be a rod promoting factor in vitro (Altshuler et al., 1993) and this activity might underlie the dependence on taur-
0378-5955 / 97 / $17.00 ß 1997 Elsevier Science B.V. All rights reserved PII S 0 3 7 8 - 5 9 5 5 ( 9 7 ) 0 0 0 5 7 - 9
HEARES 2812 27-8-97
136
K.C. Horner, C. Aurousseau / Hearing Research 109 (1997) 135^142
ine to maintain normal retinal structures in development. Indeed expression of taurine immunoreactivity changes throughout development such that in the rat retina the ganglion and horizontal cells express taurine at birth but this is lost after about 10 days. Taurine persists in the photoreceptors and Muller glial cells up to adulthood (Lake, 1994; Pow et al., 1994). Little is known about taurine and inner ear function. Taurine is reported to be the most abundant free amino acid in the inner ear (Davies and Owen, 1985). Its possible osmoregulatory function and implication in the pathology of Meènieére's disease has been hypothesised (Horner et al., 1993). Meènieére's disease is characterised by a trilogy of symptoms including £uctuant hearing loss, tinnitus and episodes of vertigo and is morphologically identi¢ed at post mortem by a swelling of the endolymphatic spaces of the inner ear. One audiological diagnostic test involves the oral administration of glycerol which, in the early £uctuant phase of the pathology, can provoke a transient hearing improvement thought to be due to the dehydration of the inner ear (Klockho¡ and Lindblom, 1966). In order to shed some light on the pathology, an animal model has been developed in which surgical blocking of the endolymphatic duct interrupts the £ow of endolymph towards the sac (Kimura and Schuknecht, 1965). The animal model has been widely employed and many physiological modi¢cations have been described and reviewed (Horner, 1993). Electrochemical sampling of the endolymph has indicated that the osmotic gradient within the normal cochlea increases by about 2.5% from the apex towards the base. In the animal model and after 9 weeks of hydrops this gradient is abolished since the osmolality at the base of the cochlea is decreased (Sziklai et al., 1992). The endolymph therefore becomes slightly hypo-osmotic relative to that of normal cochleae. It was therefore particularly opportune to investigate the e¡ect of a moderate hypo-osmolality induced by hydrops and a hyper-osmolality induced by glycerol on taurine distribution in the organ of Corti. If taurine has a functional role in inner ear osmoregulation we expected that its localisation or its activity ought to be modi¢ed under di¡erent osmotic stress conditions.
2. Materials and methods
2.1. Animals
This study employed three groups of female pigmented guinea pigs. Following several preliminary experiments on surface preparation of the dissected organ of Corti data were collected from (1) 10 guinea pigs with normal hearing, (2) 10 guinea pigs with experimentally induced endolymphatic hydrops of the inner ear on one side. For ¢ve of these animals the inner ear tissues were observed after 1^2 months and de¢ned here as early hydrops. The other ¢ve animals were sacri¢ed after 5 months and de¢ned here as late hydrops, (3) 10 guinea pigs which were given glycerol (6 g/kg body weight) orally 30 min before being sacri¢ed. The care and experimental procedures carried out on the animals were performed according to the European Community regulations. 2.2. Induction of endolymphatic hydrops
The animals were anaesthetised (1 ml/kg) with an intramuscular injection of Ketalar-Rompun mixture in the ratio 2:1 (ketamine hydrochloride 50 mg/ml; xylazine 2%). An intradural induction of endolymphatic hydrops was carried out on the left ear via the posterior fossa by obstruction of the endolymphatic duct. The technique for surgical induction of experimental hydrops has been well documented since its ¢rst description (Kimura and Schuknecht (1965) and has been employed routinely by us for the past 10 years (see for example Horner and Guilhaume, 1995). In brief, the bone over the posterior fosa was removed, the dura mater was incised along the midline and the two parts folded laterally. The cerebellum was displaced medially and with care to reveal the pigmented endolymphatic sac lying close to the lateral sinus. A pointed needle was employed to open the intra-osseus part of the sac and then dental wax was forced into the opening to obscure the canal. The cerebellum was allow to return to its place and gel foam was placed over the cranial opening. The skin was sutured and the animals allowed to recover. The surgical intervention lasted no more than 20
C
Fig. 1. Taurine-like immunolabeling in the organ of Corti. A: Control section where the primary antibody was omitted. Similar observations were made when the taurine antiserum was pre-absorbed or with normal rabbit serum at high dilution levels. B^D: Normal organ of Corti from control cochlea (c). B: The tectorial membrane (arrow heads) has clearly no taurine-like immunolabeling. C: The tectorial membrane (arrow heads) can be seen to be attached to immunolabeled supporting cells (arrows). D: Taurine-like immunolabeling is seen in the Deiters' cells and absent from the OHCs (asterisks). E^G: Organ of Corti from hydropic cochleae (h). The section presented in (E) is from a cochlea with short-term hydrops. Within the IHC (arrows) the taurine-like immunolabeling appears less intense than within the surrounding supporting cells. The three OHCs have no labeling (asterisks). One Deiters' cell phalangeal process can be seen to be swollen and with some taurine-like immunolabeling (arrow heads). The sections presented in (F) and (G) are from long-term hydropic cochleae. F: The Reissner's membrane is in a vertical position (arrow heads) and the tectorial membrane is elevated (arrows). G: The swollen Deiters' cells appear to have taurine-like immunolabeling. H^J: Organ of Corti from glycerol-treated animals (g). I: The tectorial membrane can be seen to have taurine-like labeling. J: OHCs torn o¡ from the organ of Corti during the tissue preparation can be seen to have some taurine-like immunolabeling associated uniquely with the synaptic pole (large asterisk) and the apex of the cells (asterisks).
HEARES 2812 27-8-97
K.C. Horner, C. Aurousseau / Hearing Research 109 (1997) 135^142
HEARES 2812 27-8-97
137
138
K.C. Horner, C. Aurousseau / Hearing Research 109 (1997) 135^142
min and the animals were awake and moving about the cage with no sign of vestibular disturbance after about 30 min. 2.3. Glycerol administration
The glycerol solution employed consisted of 600 g of glycerine made up to 1 l with distilled water. The dose level employed was 6 g/kg body weight given orally once and 30 min before the animals were killed. We have shown earlier that at this dose level changes in auditory function in guinea pigs occur which reach maximum after 30 min and recover completely at 24 h (Horner and Cazals, 1987). 2.4. Preparation of tissue for immunocytochemistry
The animals were anaesthetised (1 ml/kg) with an intramuscular injection of Ketalar-Rompun mixture in the ratio 2:1 (ketamine hydrochloride 50 mg/ml; xylazine 2%). Following decapitation the temporal bones removed, perfused via the round and oval windows with 2.5% glutaraldehyde/1% paraformaldehyde in 0.1 M phosphate bu¡er at pH 7.4. In some cases the apex of the cochlea was opened before the perfusion was commenced. The time between decapitation and perfusion was up to 5 min for the two ears. The temporal bones were left bathing in the same ¢xative for 24 h. The inner ears were then perfused with the phosphate bu¡er and left again for 24 h. The organ of Corti was dissected out from the temporal bones. The tissue was embedded in Tissue-Tek (Miles Inc.) and frozen to 350³C in isopentane on dry-ice. Frozen sections of 10 Wm were made at 320³C, collected on glass slides, dried rapidly on a heated plate to 40³C and frozen at 320³C until use. 2.5. Immunohistochemistry
For immunocytochemistry the slides were (1) washed twice, 15 min each, in phosphate bu¡er (0.01 M, pH 7.4) with bovine serum albumen (0.2%) and NaCl (0.9%) (PBS/BSA/NaCl); (2) washed for 5 min in 3% H2 O2 in methanol 100% (1:4) for 5 min in order to inhibit endogenous peroxidases; (3) washed twice for 5 min in PBS/BSA/NaCl; (4) incubated at room temperature for 1 h in PBS/BSA/NaCl with 0.3% Triton-X and 10% normal goat serum (Chemicon International, CA) to permeabilize the tissues and to lower non-speci¢c background activity; (5) exposed for 24 h at 4³C to polyclonal antibody to taurine from the rabbit (Chemicon International CA) in PBS/BSA/NaCl with 0.3% Triton-X and 1% normal goat serum; (6) washed twice, 15 min each, in PBS/BSA/NaCl; (7) incubated at room temperature 1 h in goat anti-rabbit IgG conjugated to peroxidase (dilution 1:50) (Chemicon International,
CA) in PBS/BSA/NaCL with 0.3% Triton-X and 1% normal goat serum; (8) exposed to rabbit peroxidase anti-peroxidase (dilution 1:400) (Sigma) for 1 h; (9) washed twice in PBS; (10) incubated in diaminobenzidine and H2O2 (Sigma fast) in PBS for precisely 8 min; (11) washed twice in PBS; (12) quickly air dried under a cool ventilator; (13) mounted in Entellan (Merck) and coverslipped. Normal rabbit serum for control experiments was also employed (Chemicon International, CA). An initial series of the test batches (n = 10) consisted of two slides (2U5 sections) from each of the two test conditions and the normal condition as well as control slides for which the primary antibody against taurine was omitted. For speci¢city controls additional experiments were carried out (n = 10) in which the diluted taurine antiserum was pre-adsorbed with taurine at different concentrations (250^25 mM). Immunohistochemistry procedures were performed simultaneously on test sections (with the taurine antiserum) and control sections (pre-adsorbed). In addition a series of slides were exposed to normal rabbit serum at di¡erent dilutions (1:300^1:2400) and compared with the immunohistochemical reaction of the taurine antiserum at the same dilutions. 3. Results
3.1. Taurine-like immunolabeling in the normal organ of Corti
In absence of the taurine antiserum no immunoreactivity was observed (Fig. 1A). Similarly pre-adsorption with taurine abolished all labeling. While some weak unspeci¢c background staining was observed when the taurine antiserum was replaced by the normal serum at low dilutions (1:300, 1:600) it was eliminated at higher dilutions (1:1200, 1:1800, 1:2400). At these high dilutions of the taurine antiserum a speci¢c taurine-like immunoreaction was still observable. In test sections of the normal organ of Corti the most intense taurine-like immunolabeling was speci¢cally associated with supporting cells (Fig. 1B). No difference in degree and/or distribution of immunolabeling from di¡erent cochlear turns was detected although observations were more easily made in the three upper turns where the hair cells are longer. Intense taurine-like immunoreactivity was associated with the cup-like portion of the Deiters' cells which is in close contact with the synaptic pole of the outer hair cells (OHC, Fig. 1D) and the Hensen cells towards the outer edge of the organ. Similar intense taurine-like immunoreactivity was seen within the inner phalangeal cells and the inner border cells which together surround the inner hair cells (IHC, Fig. 1C) as well as the inner
HEARES 2812 27-8-97
K.C. Horner, C. Aurousseau / Hearing Research 109 (1997) 135^142
sulcus cells towards the inner edge of the organ (Fig. 1B). There was, in contrast, a remarkable absence of immunolabeling associated with the OHCs (Fig. 1D). Some localised taurine-like immunolabeling could be observed on the reticular lamina but appeared to be associated with the contours of the apex of the cell rather than with the cuticular plates of the OHCs. The present light microscope study could not con¢rm the absence of taurine-immunolabeling from the IHCs since they are surrounded by taurine-like immunolabeling supporting cells, although in some sections there did appear to be less taurine-like immunolabeling in the IHCs (Fig. 1E). The tectorial membrane overlying the apical pole of the sensory hair cells showed no immunoreaction (Fig. 1B,C). The tectorial membrane appeared to be attached to supporting cells on the inner sulcus side of the IHCs which were probably border cells (Fig. 1C). These border cells were highly taurine immunoreactive. In addition there was little or no taurine-like immunolabeling associated with the outer pillar cells. 3.2. E¡ect of endolymphatic hydrops
The presence of hydrops was always con¢rmed at dissection and identi¢ed as the increased volume of the endolymph-¢lled scala media and the congruent diminution of the perilymph-¢lled scala vestibuli. The Reissner's membrane separating the two scala was con¢rmed to be extending into the scala vestibuli. Within the ¢rst 2 months of endolymphatic hydrops the distribution of taurine-like immunoreactivity appeared similar to that observed in the normal organ of Corti. In early hydrops the OHCs remained distinctly taurine free (Fig. 1E). The phalangeal process of the Deiters' cells which, in normal control cochleae were not immunoreactive to taurine and di¤cult to detect by light microscopy, could in hydropic cochleae, appear swollen and taurine immunoreative in particular in association with the outermost row of OHCs (Fig. 1E). In cochleae with hydrops of more than 5 months the Reissner's membrane was clearly in a distended vertical position and the tectorial membrane could also be observed to be elevated (Fig. 1F). In general, the tectorial membrane did not appear to be taurine-immunoreactive. The membrane could appear to be taurine immunoreactive but this was interpreted as artefactual when the tissue was clearly folded. The cells located under the reticular lamina which were normally taurine-free were now intensely immunoreactive. Close microscopic observation of the tissue suggested strongly that the immunoreactivity was associated not with OHCs but rather with the swollen phalangeal processes of the Deiters' cells which replaced the degenerating hair cells (Fig. 1G). This interpretation was reinforced by the
139
observation of the extensive immunoreaction throughout the reticular lamina in long-term hydrops. 3.3. E¡ect of glycerol
Thirty minutes following the oral administration of glycerol the tectorial membrane was observed to become taurine immunoreactive (Fig. 1H). In addition, the tectorial membrane appeared more tightly bound to the reticular lamina (Fig. 1I). The OHCs remained distinctly non immunoreactive at the glycerol dose tested (Fig. 1J). 4. Discussion
The present study has con¢rmed a recent report that in the organ of Corti of the cochlea, taurine-like immunoreactivity is normally associated with supporting cells (Usami and Ottersen, 1995). The observations are in agreement with earlier autoradiographic data (Schwartz and Ryan, 1983) and in con£ict with an earlier report of taurine speci¢cally within the OHCs and very little within supporting structure (Harding and Davies, 1993). A number of possible explanations can be proposed in order to account for the discrepancy. First, it should be recalled that the antibody against taurine employed by Harding and Davies (1993) was a monoclonal antibody while the present study employed a polyclonal antibody. Second, Harding and Davies carried out pre-embedding immunocytochemistry on the whole organ of Corti while in the present study immunocytochemistry was carried out on 10 Wm frozen sections. Third, the ¢xative (4% formaldehyde) employed by Harding and Davies has been reported to retain less than 1% of the free amino acid pool (Storm-Mathisen and Ottersen, 1990). Usami and Ottersen (1995) have pointed out that the greater the loss of free amino acid during ¢xation, the greater is the likelihood that the immunostaining will diverge from the in-vivo distribution. In the present study the absence of taurine from the IHCs as seen in transmission microscopy (Usami and Ottersen, 1995) could not be con¢rmed by light microscopy since the cells are surrounded by taurine-like immunoreactive supporting cells. This, on the other hand, is the ¢rst study demonstrating the localisation of taurine in the swollen phalangeal process of the Deiters's cells following hair cell loss. The present data might have signi¢cant implications for studies on isolated hair cells in vitro where the hair cell/Deiters' cell association is disrupted and the natural environment is highly altered. In addition, the data may provide a basis for a better understanding of the repair mechanism of the organ of Corti following damage which involves the swelling of the phalangeal process of the Deiter's cells.
HEARES 2812 27-8-97
140
K.C. Horner, C. Aurousseau / Hearing Research 109 (1997) 135^142
4.1. No
taurine-like
immunolabeling
in
the
outer
hair
cells of the organ of Corti
The absence of taurine from healthy OHCs described here by light microscopy is particularly striking. This absence from the OHCs is certainly not due to a possible leakage of taurine from the hair cells due to the triton treatment employed here. Similar distribution of taurine in the organ of Corti has been reported recently by Usami and Ottersen (1995) who did not employ triton. The possible leakage of taurine due to a delay to ¢xation can also be discounted. Our rapid ¢xation technique ensures that the round and the oval windows are fully exposed and guarantees a rapid and e¤cient ¢xation of both ears within 5 min maximum of decapitation. In the present study opening of the apex of the cochlea was not found to in£uence the taurine-like immunolabeling. An alternate ¢xation technique of invivo perfusion of the cochleae, is not beyond criticism. Deep and relatively prolonged anaesthesia is required in order to e¤ciently expose the windows by a ventral approach to the bullae and so the possible complications due to the e¡ect of anaesthesia and surgical stress cannot be ruled out. Having said that, such an in-vivo ¢xation procedure has been employed by Usami and Ottersen (1995) and they have reported the absence of taurine from the sensory hair cells. The observed localisation of taurine uniquely in supporting cells is probably related to the distribution of the many gap junctions between all supporting cells of the organ of Corti while being absent between supporting and sensory cells (Iurato et al., 1976; Kikuchi et al., 1995). 4.2. Hair
cell
motility
requires
an
e¤cient
protection
against the osmotic stress
The OHCs of the organ of Corti, change shape on depolarisation/repolarisation (Dallos et al., 1993), at least in vitro, and therefore belong to a specialised category of cells whose osmoregulating facilities can cope with cell motility. A `slow motile response' of hair cells associated with swelling can be provoked by exposing the cells in vitro to a milieu rich in K or by lowering the osmolarity. The shortening is reported to be associated with a volume increase of up to 20% (Dulon et al., 1988; Ulfendahl, 1988). However, it should be recalled that this volume change is observed for hair cells isolated from their supporting cells in an arti¢cial medium whose composition is experimentally modi¢ed. The in-vivo situation is not likely to present such extreme conditions. A `fast motile response' with maintenance of cell volume, involves cyclic elongation^contraction when electrically stimulated in vitro (Brownell et al., 1985) and is presumed to mimick the hair cell response to sound stimulation in vivo. The precise
mechanisms responsible for the fast motile response have not been established but considered at this time to be `molecular motors' in the plasma membrane (Kalinec et al., 1992) working in association with the underlying cortical lattice involving the interaction of actin and spectrin (Holley and Ashmore, 1990) and with the underlying lateral cisternae (Brownell, 1986). It is also assumed that the elastic restoring force is dependent on the intracellular turgor pressure since the fast motility deteriorates when the intracellular pressure is reduced (Brownell and Shehata, 1990). Indeed it has been speculated that modi¢cation of intracellular pressure in vivo may provide a mechanism for controlling the gain of the mammalian `cochlear ampli¢er' (Kakehata and Santos-Sacchi, 1995). Harding and Davies (1993) who observed taurine within the hair cells postulated that taurine might be implicated in hair cell motility. The present study demonstrated the absence of taurine from hair cells and so the data here suggest that hair cells do not rely directly on taurine as an osmolyte in order maintain the intracellular pressure. The shunting of taurine probably occurs over a prolonged time scale which would appear more appropriate for the supporting cells, functioning as a bu¡ering protective system for the hair cells. 4.3. The swelling of the phalangeal process of the Deiters' cells is associated with taurine-like immunolabeling
We have demonstrated that in the normal organ of Corti the cup-like body of Deiters' cells has high taurine-like immunolabeling while the phalangeal process appears not to have taurine. Within the ¢rst 2 months of hydrops the localisation of taurine does not appear to be considerably modi¢ed. This apparent constancy of taurine localisation is observed despite a moderate decrease in the osmolality (Sziklai et al., 1992) and the accumulation of electron dense bodies in the subcuticular cytoplasm, thought to be lipofuschin, which can be observed already 1 month after induction of hydrops (Horner and Guilhaume, 1995). In long-term hydrops we have observed an accumulation of taurine in the swollen phalangeal process of the Deiter's cells and in particular the outermost cells are a¡ected ¢rst. It is should be recalled that in hydropic cochleae there is a selective loss of short and middle stereocilia in the three upper cochlear turns (Horner et al., 1988) which begins with the outermost row (Rydmarker and Horner, 1990). It has been speculated that the particular pattern of atrophy of stereocilia starting from the outermost row is due to the abnormal elevation of the tectorial membrane which would a¡ect the outermost hair cells ¢rst (Rydmarker and Horner, 1990). The loss of hair cells is delayed relative to the stereocilia loss but follows the same pattern with the outermost hair cells a¡ected ¢rst. The swelling of the
HEARES 2812 27-8-97
K.C. Horner, C. Aurousseau / Hearing Research 109 (1997) 135^142
141
phalangeal process of the Deiter's cells which begins
motic stress conditions, which would be facilitated by
with the outermost cells is therefore almost certainly
the coupling of the tectorial membrane to the inner hair
related to the damage to the associated hair cell.
cell border cells as observed here. Such a coupling in
Swelling of cells
was
the phalangeal
¢rst
following
observed
antibiotic
in
process of
damaged
treatment
the
organs
Deiters' of
vivo, however, remains a subject of debate (Lawrence
Corti
and Burgio, 1980), since most histological procedures
and
are known to a¡ect the ¢ber composition of the tecto-
(Hunter-Duvar
Mount, 1978 ; Forge, 1985). The swelling of the phalan-
rial membrane (Hasko and Richardson, 1988).
geal process of the Deiter's cell is clearly a means of
The observed tighter coupling between the tectorial
¢lling the space left by a dying hair cell in order to
membrane and the organ of Corti following intrave-
conserve the structure of the organ of Corti and could
nous injection of 2 g/kg glycerol has been reported ear-
therefore be considered as participating in programmed
lier (Meyer zum Gottesberge and Tsujikawa, 1993) to
cell death. The demonstration here of taurine accumu-
be related to a glycerol-induced decrease in Ca
lation within the swollen phalangeal processes of the
lier studies have demonstrated that the un¢xed tectorial
Deiter's
to
membrane maintained in arti¢cial endolymph is highly
hydropic ears but might represent a mechanism which
sensitive to calcium-swelling in solution containing the
initiates or accompanies swelling. Whether the £ux of
Ca
taurine is due to passive di¡usion or active uptake by
ing
the phalangeal processes cannot be resolved here. As
1979 ; Freeman et al., 1994). The taurine immunolabel-
mentioned earlier following intraperitoneal injection of
ing of the tectorial membrane following glycerol admin-
distilled water in rats, there is decreased taurine immu-
istration might thus represent a protective measure in
nolabeling of cerebellar Purkinje cells and increased im-
order to augment the local concentration of Ca
munolabeling in neighboring glial cells which become
has
swollen (Nagelhus et al., 1993).
Ca
An
cells
is
probably
osmoregulatory
not
role
for
a
feature
taurine
particular
in
2
2
chelator ethylenediamine-tetraacetate and shrink-
on
the
been
2
. Ear-
re-introduction
established
availability
association
protects
with the Deiter's cells is in keeping with recent identi-
high Ca
against
2
that
under Ca
2
of
Ca
2
taurine
conditions overload
(Kronester-Frei,
can of
both low
under
2
. It
increase
2
Ca
and
conditions
of
(Huxtable, 1992).
¢cation of a structure unique to these cells named the
The tighter coupling of the tectorial membrane to the
`rosette complex' (Spicer and Schulte, 1993). Immuno-
organ of Corti following glycerol administration might
histochemistry has revealed ci¢c
to
the
meshwork has
led
to
rosette
which the
Q
complex
is
muscle actin isoform spe-
account to some extent for the reported temporary im-
and
è nie é re's patients. Howprovement in hearing in some Me
enclosed
immunoreactive
proposal
that
the
to
in
a
loose
vimentin
Deiter's
cells
and may
è nie é re's disease the tectoever in the animal model of Me rial
membrane
has
been
reported
to
be
elevated
play a role in regulating ion homeostasis and/or motile
(Rydmarker and Horner, 1990 ; Meyer zum Gottesberge
response properties of the organ of Corti (Nakazawa et
and Tsujikawa, 1993) and Ca
al., 1995). In addition the restructuration of the organ
noyu and Meyer zum Gottesberge, 1986), yet adminis-
of Corti almost certainly involves a modi¢cation of the
tration of glycerol at di¡erent dose levels including 6 g/kg
e¡erent feedback simultaneously to supranuclear termi-
failed to improve the compound action potential audio-
nals outer
de-
gram (Horner and Cazals, 1987). It remains to be deter-
scribed in the cat (Liberman et al., 1990) and in the
¨mined whether the animal model is representative of Me
human organ of Corti (Nadol and Burgess, 1994).
é re's patients or of a particular sub-group of patients nie
hair
cells
and to
the
Deiters'
cells
as
2
is highly elevated (Ni-
who present no improvement or a hearing deterioration
4.4. Glycerol induces taurine-like immunolabeling within the tectorial membrane
following glycerol administration (Barbara et al., 1994).
Acknowledgments
The present study has shown that in the normal organ of Corti there is no taurine-like immunolabeling associated with the tectorial membrane. Following the
This
research
received
¢nancial
support
from
the
ègional d'Aquitaine. Conseil Re
oral administration of 6 g/kg glycerol some taurine-like immunolabeling
was
brane
tectorial
and
the
seen
within
the
membrane
tectorial appeared
memto
be
more closely bound to the organ of Corti. Two possible interpretations
of
the
immunolabeling
can
be
made.
First glycerol might modify the constitution of the tectorial membrane and so, as in any damaged tissue, it might become more immunoreactive. The other interpretation would favour a delocalisation of taurine from the supporting cells into the tectorial membrane, in os-
References Altshuler, D., Lo-Turco, J.J., Rush, J. and Cepko, C. (1993) Taurine promotes the di¡erentiation of a vertebrate retinal cell type in vitro. Development 119, 1317^1328. Barbara, M., Monini, S., Attanasio, G. and Filipo, R. (1994) Glycerol test : considerations on the prognostic value in the long-term apè nie é re's patients. In : R. Filipo and M. Barbara (Eds.), proach to Me è nie é re's Disease, Kugler, Amsterdam, pp. 127-129. Me Brownell, W.E. (1986) Outer hair cell motility and cochlear frequency
HEARES 2812 27-8-97
K.C. Horner, C. Aurousseau / Hearing Research 109 (1997) 135^142
142
selectivity. In : B.C.J. Moore and R.D. Patterson (Eds.), Auditory Frequency Selectivity, Plenum, New York, pp. 109-116.
Kalinec, F., Holley, M.C., Iwasa, K.H., Lim, D.J. and Kachar, B. (1992) A membrane-based force generation mechanism in auditory
Brownell, W.E., Bader, C.R., Bertrand, D. and de Ribaupierre, Y. (1985) Evoked mechanical responses of isolated cochlear hair cells. Science 227, 194^196.
sensory cells. Proc. Natl. Acad. Sci. 89, 8671^8675. Kikuchi, T., Kimura, R.S., Paul, D.L. and Adams, J.C. (1995) Gap junctions in the rat cochlea : immunohistochemical and ultrastruc-
Brownell, W.E., and Shehata, W.E. (1990) The e¡ect of cytoplasmic
tural analysis. Anat. Embryol. Berl. 191, 101^118.
turgor pressure on the static and dynamic mechanical properties of
Kimura, R.S. and Schuknecht, H.F. (1965) Membranous hydrops in
outer hair cells. In : P. Dallos, C.D. Geisler, J.W. Matthews, M.A.
the inner ear of the guinea pig after obliteration of the endolym-
Ruggero, and C.A. Steele (Eds.), Mechanics and Biophysics of Hearing. Springer, New York, pp. 52-59.
phatic sac. Pract-Otorhinolaryngol. 27, 343^354. Klockho¡, I. and Lindblom, U. (1966) Endolymphatic hydrops as
Dallos, P., Hallworth, R. and Evans, B.N. (1993) Theory of electrically driven shape changes of cochlear outer hair cells. J. Neurophysiol. 70, 299^323.
revealed
by
the
glycerol
test.
Acta.
Otolaryngol.
(Stockh.)
61,
459^462. Kronester-Frei, A. (1979) The e¡ect of changes in endolymphatic ion
Davies, W.E., and S. Owen (1985) The nature of neurotransmitters in the mammalian lower auditory system. In : D.G. Drescher (Ed.), Auditory Biochemistry, CC Thomas, Spring¢eld, IL, pp. 244-257.
concentrations on the tectorial membrane. Hear. Res. 1, 81^94. Lake, N. (1994) Taurine and GABA in the rat retina during postnatal development. Vis. Neurosci. 11, 253^260.
Dulon, D., Aran, J.M. and Schacht, J. (1988) Potassium-depolarisa-
Lawrence, M. and Burgio, P.A. (1980) Attachment of the tectorial
tion induces motility in isolated outer hair cells by an osmotic
membrane revealed by scanning microscopy. Ann. Otol. Rhinol.
mechanism. Hear. Res. 32, 123^130.
Laryngol. 89, 325^330.
Forge, A. (1985) Outer hair cell loss and supporting cell expansion following chronic gentamicin treatment. Hear. Res. 19, 171^182. Freeman, D.M., Cotanche, D.A., Ehsani, F. and Weiss, T.F. (1994) The osmotic responses of the isolated tectorial membrane of the
,
chick to isosmotic solutions : e¡ect of Na
K
2
and Ca
concen-
tration. Hear. Res. 79, 197^215.
Liberman, M.C., Dodds, L.W. and Pierce, S. (1990) A¡erent and e¡erent innervation of the cat cochlea : quantitative analysis with light and electron microscopy. J. Comp. Neurol. 301, 443^460. Lombardini, J.B. (1991) Taurine : retinal functions. Brain Res. Rev. 16, 151^169. Meyer
Hageman, G.S. and Schmidt, S.Y. (1987) Taurine de¢cient pigmented and albino rats : early retinal abnormalities and di¡erential rates of photoreceptor degeneration. Prog. Clin. Biol. Res. 247, 497^515. Harding, N.J. and Davies, W.E. (1993) Cellular localisation of taurine in the organ of Corti. Hear. Res. 65, 211^215.
zum
Gottesberge,
A.M.
and
Tsujikawa, S.
(1993)
Glycerol
e¡ect on the guinea pig tectorial membrane. Eur. Arch. Otorhinolaryngol. 250, 88^91. Nadol, J.B. and Burgess, B.J. (1994) Supranuclear e¡erent synapses on outer hair cells and Deiters' cells in the human organ of Corti. Hear. Res. 81, 49^56.
Hasko, J.A. and Richardson, G.P. (1988) The ultrastructural organ-
Nagelhus, E.A., Lehmann, A. and Ottersen, O.P. (1993) Neuronal-
isation and properties of the mouse tectorial membrane matrix.
glial exchange of taurine during hypo-osmotic stress : a combined
Hear. Res. 35, 21^38.
immunocytochemical
Hayes, K.C., Carey, R.E. and Schmidt, S.Y. (1975) Retinal degeneration association with taurine de¢ciency in the cat. Science 188, 947^951.
the
biochemical
analysis in
rat
cerebellar
Nakazawa,
K.,
Schultz,
B.A.
and
Spicer,
S.S.
(1995)
The
rosette
complex in gerbil Deiters cells contain Y actin. Hear. Res. 89,
Holley, M.C. and Ashmore, J. (1990) Spectrin, actin and the structure of
and
cortex. Neuroscience 54, 615^631.
cortical
lattice
in
mammalian
cochlear
outer
hair
cells.
J. Cell Sci. 96, 283^291.
121^129. Ninoyu, O. and Meyer zum Gottesberge, A.M. (1986) Changes in
2
Ca
Horner, K.C. (1993) Review : functional changes associated with ex-
activity and DC potential in experimentally induced endo-
lymphatic hydrops. Eur. Arch. Otorhinolaryngol. 243, 106^107.
perimentally induced endolymphatic hydrops. Hear. Res. 58, 1^18.
Pow, D., Crook, D.K. and Wong, R.O. (1994) Early appearance and
Horner, K.C. and Cazals, Y. (1987) Glycerol induced changes in the
transient expression of putative amino acid neurotransmitters and
Arch.
related molecules in the developing rabbit retina : an immunocy-
Horner, K.C. and Guilhaume, A. (1995) Ultrastructural changes in
Rydmarker, S. and Horner, K.C. (1990) Morphological changes of
the hydropic cochlea of the guinea pig. Eur. J. Neurosci. 7, 1305^
hair cell stereocilia and tectorial membrane in guinea pig with
cochlear
responses
of
the
guinea
pig
hydropic
ear.
Eur.
tochemical study. Vis. Neurosci. 11, 1115^1134.
Otorhinolaryngol. 244, 49^54.
experimentally induced hydrops. Scan. Microsc. 4, 705^714.
1312. Horner, K.C., Guilhaume, A. and Cazals, Y. (1988) Atrophy of mid-
Schwartz, I.R. and Ryan, A.F. (1983) Di¡erential labeling of sensory
dle and short stereocilia on outer hair cells of guinea pig labeling
cell and neural populations in the organ of Corti following amino
with experimentally induced hydrops. Hear. Res. 32, 41^48. Horner, K.C., Huang, W. and Erre, J.P. (1993) The e¡ect of taurine modi¢ed diet on normal and hydropic
ears of the guinea pig.
Hear. Res. 70, 1^8. Hunter-Duvar, I. and Mount, R.J. (1978) The organ of Corti following ototoxic antibiotic treatment. Scan. Elect. Microsc. 11, 423^430. Huxtable, R.J. (1992) Physiological actions of taurine. Physiol. Rev. 72, 101^163.
acid incubations. Hear. Res. 9, 185^200. Spicer, S.S. and Schulte, B.A. (1993) Cytologic structures unique to Deiters cells of the cochlea. Anat. Rec. 237, 421^430. Storm-Mathisen, J. and Ottersen, O.P. (1990) Antibodies and ¢xatives for the immunocytochemical localisation of glycine. In : J. StormMathisen and O.P. Ottersen (Eds.), Glycine Neurotransmission, John Wiley, Chichester, pp. 281-301. Sziklai, I., Ferrary, E., Horner, K.C., Sterkers, O. and Amiel, C.
Imaki, H., Jacobson, S.G., Kemp, C.M., Knighton, R.W., Neuringer,
(1992) Time-related alterations of endolymph composition in an
M. and Sturman, J. (1993) Retinal morphology and visual pigment
experimental model of endolymphatic hydrops. Laryngoscope 102,
levels in 6 and 12 month old rhesus monkeys fed a taurine free human infant formula. J. Neurosci. Res. 36, 290^304. Iurato, S., Franke, K., Luciano, L., Wermbter, G., Pannese, E. and Reale, E. (1976) Intercellular junctions in the organ of Corti as re-
431^438. Ulfendahl, M. (1988) Volume and length changes in outer hair cells of the
guinea
pig
after
potassium-induced
shortening.
Eur.
Arch.
Otorhinolaryngol. 245, 237^243.
vealed by freeze fracturing. Acta. Otolaryngol. (Stockh.) 82, 58^69.
Usami, S. and Ottersen, O.P. (1995) The localization of taurine-like
Kakehata, S. and Santos-Sacchi, J. (1995) Membrane tension directly
immunoreactivity in the organ of Corti : a semiquantitative, post-
shifts voltage dependence of outer hair cell motility and associated
embedding immuno-electron microscopic analysis in the rat with
gating charge. Biophys. J. 68, 2190^2197.
some observations in the guinea pig. Brain Res. 676, 277^284.
Immunoreactivity for taurine in the cochlea: its abundance in supporting cells Kathleen C. Horner *, Catherine Aurousseau è rimentale, Universite è Bordeaux II, Bordeaux, France INSERM, Laboratoire d'Audiologie Expe
Received 1 October 1996; revised 31 March 1997; accepted 5 April 1997
Abstract
Taurine is the second most abundant free amino acid in the brain where its osmoregulatory function is well established. Taurinedeprived kittens show retinal pathology leading to blindness. In the inner ear, taurine has been reported to be the most abundant free amino acid although its role in inner ear function is not known. Immunohistochemistry was employed here to investigate the localisation of taurine in normal cochleae of the guinea pig compared with two different conditions: experimentally induced endolymphatic hydrops and after oral administration of glycerol. In normal cochleae, by light microscopy, taurine-like immunoreaction was never observed in the sensory outer hair cells and appeared absent from the inner hair cells. In contrast taurine-like immunolabeling was found to be present in all supporting tissue with the striking exception of the tectorial membrane and the outer pillar cell which had no or little taurine immunoreactivity respectively. In early experimental endolymphatic hydrops, the distribution of taurine-like immunoreactivity appeared similar to that observed for normal cochleae. In long-term hydrops, degenerated outer hair cells were replaced by the swelling of the phalangeal process of the Deiters' cells which became highly immunoreactive to taurine. After glycerol administration, the tectorial membrane became more tightly bound to the apical surface of the sensory hair cells and distinctly immunoreactive to taurine. The localisation of taurine in the organ of Corti shown here is consistent with taurine being involved in the maintenance of osmotic equilibrium in the normal and perhaps also in the restructuration of the pathological organ of Corti. Keywords :
Cochlea; Hair cell; Tectorial membrane; Supporting cell; Guinea pig
1. Introduction
Taurine (2-aminoethane sulfonic acid) is one of the most abundant free amino acids in mammals and in brain tissue and is the most abundant free amino acid after glutamate (Huxtable, 1992). Various functions have been attributed to taurine although its role in osmoregulation appears to be its major function. The rise in extracellular potassium or the decrease in osmolarity tend to cause cell swelling which triggers the release of
* Corresponding author. Laboratoire d'Audiologie Expeèrimentale, Universiteè Bordeaux II, Hoêpital Pellegrin, Place Ameèlie Raba Leèon, 33076 Bordeaux, France. Tel.: (33) 5-56-24-20-47; Fax: (33) 5-56-9629-84. New address after 15 July 1997: Laboratoire d'Otologie and Neuro-Otologie, Faculteè de Medecine Nord, Boulevard Pierre Dramard, 13916 Marseille Cedex 20, France.
taurine from various cell types in culture as part of the normal cell volume regulation mechanism. Following intraperitoneal injection of distilled water in rats, Nagelhus et al. (1993) observed the delocalisation of taurine from cerebellar Purkinje cells into neighboring Golgi epithelial cells. Those authors considered that the cellular redistribution might represent a rapid adjustment to osmotic perturbation in vivo with a higher priority for neuronal compared with glial volume regulation. Taurine is known to be essential for retinal function (Lombardini, 1991) and taurine deprivation is reported to result in retinal degeneration in monkeys (Imaki et al., 1993), rats (Hageman and Schmidt, 1987) and cats (Hayes et al., 1975). Taurine has been shown to be a rod promoting factor in vitro (Altshuler et al., 1993) and this activity might underlie the dependence on taur-
0378-5955 / 97 / $17.00 ß 1997 Elsevier Science B.V. All rights reserved PII S 0 3 7 8 - 5 9 5 5 ( 9 7 ) 0 0 0 5 7 - 9
HEARES 2812 27-8-97
136
K.C. Horner, C. Aurousseau / Hearing Research 109 (1997) 135^142
ine to maintain normal retinal structures in development. Indeed expression of taurine immunoreactivity changes throughout development such that in the rat retina the ganglion and horizontal cells express taurine at birth but this is lost after about 10 days. Taurine persists in the photoreceptors and Muller glial cells up to adulthood (Lake, 1994; Pow et al., 1994). Little is known about taurine and inner ear function. Taurine is reported to be the most abundant free amino acid in the inner ear (Davies and Owen, 1985). Its possible osmoregulatory function and implication in the pathology of Meènieére's disease has been hypothesised (Horner et al., 1993). Meènieére's disease is characterised by a trilogy of symptoms including £uctuant hearing loss, tinnitus and episodes of vertigo and is morphologically identi¢ed at post mortem by a swelling of the endolymphatic spaces of the inner ear. One audiological diagnostic test involves the oral administration of glycerol which, in the early £uctuant phase of the pathology, can provoke a transient hearing improvement thought to be due to the dehydration of the inner ear (Klockho¡ and Lindblom, 1966). In order to shed some light on the pathology, an animal model has been developed in which surgical blocking of the endolymphatic duct interrupts the £ow of endolymph towards the sac (Kimura and Schuknecht, 1965). The animal model has been widely employed and many physiological modi¢cations have been described and reviewed (Horner, 1993). Electrochemical sampling of the endolymph has indicated that the osmotic gradient within the normal cochlea increases by about 2.5% from the apex towards the base. In the animal model and after 9 weeks of hydrops this gradient is abolished since the osmolality at the base of the cochlea is decreased (Sziklai et al., 1992). The endolymph therefore becomes slightly hypo-osmotic relative to that of normal cochleae. It was therefore particularly opportune to investigate the e¡ect of a moderate hypo-osmolality induced by hydrops and a hyper-osmolality induced by glycerol on taurine distribution in the organ of Corti. If taurine has a functional role in inner ear osmoregulation we expected that its localisation or its activity ought to be modi¢ed under di¡erent osmotic stress conditions.
2. Materials and methods
2.1. Animals
This study employed three groups of female pigmented guinea pigs. Following several preliminary experiments on surface preparation of the dissected organ of Corti data were collected from (1) 10 guinea pigs with normal hearing, (2) 10 guinea pigs with experimentally induced endolymphatic hydrops of the inner ear on one side. For ¢ve of these animals the inner ear tissues were observed after 1^2 months and de¢ned here as early hydrops. The other ¢ve animals were sacri¢ed after 5 months and de¢ned here as late hydrops, (3) 10 guinea pigs which were given glycerol (6 g/kg body weight) orally 30 min before being sacri¢ed. The care and experimental procedures carried out on the animals were performed according to the European Community regulations. 2.2. Induction of endolymphatic hydrops
The animals were anaesthetised (1 ml/kg) with an intramuscular injection of Ketalar-Rompun mixture in the ratio 2:1 (ketamine hydrochloride 50 mg/ml; xylazine 2%). An intradural induction of endolymphatic hydrops was carried out on the left ear via the posterior fossa by obstruction of the endolymphatic duct. The technique for surgical induction of experimental hydrops has been well documented since its ¢rst description (Kimura and Schuknecht (1965) and has been employed routinely by us for the past 10 years (see for example Horner and Guilhaume, 1995). In brief, the bone over the posterior fosa was removed, the dura mater was incised along the midline and the two parts folded laterally. The cerebellum was displaced medially and with care to reveal the pigmented endolymphatic sac lying close to the lateral sinus. A pointed needle was employed to open the intra-osseus part of the sac and then dental wax was forced into the opening to obscure the canal. The cerebellum was allow to return to its place and gel foam was placed over the cranial opening. The skin was sutured and the animals allowed to recover. The surgical intervention lasted no more than 20
C
Fig. 1. Taurine-like immunolabeling in the organ of Corti. A: Control section where the primary antibody was omitted. Similar observations were made when the taurine antiserum was pre-absorbed or with normal rabbit serum at high dilution levels. B^D: Normal organ of Corti from control cochlea (c). B: The tectorial membrane (arrow heads) has clearly no taurine-like immunolabeling. C: The tectorial membrane (arrow heads) can be seen to be attached to immunolabeled supporting cells (arrows). D: Taurine-like immunolabeling is seen in the Deiters' cells and absent from the OHCs (asterisks). E^G: Organ of Corti from hydropic cochleae (h). The section presented in (E) is from a cochlea with short-term hydrops. Within the IHC (arrows) the taurine-like immunolabeling appears less intense than within the surrounding supporting cells. The three OHCs have no labeling (asterisks). One Deiters' cell phalangeal process can be seen to be swollen and with some taurine-like immunolabeling (arrow heads). The sections presented in (F) and (G) are from long-term hydropic cochleae. F: The Reissner's membrane is in a vertical position (arrow heads) and the tectorial membrane is elevated (arrows). G: The swollen Deiters' cells appear to have taurine-like immunolabeling. H^J: Organ of Corti from glycerol-treated animals (g). I: The tectorial membrane can be seen to have taurine-like labeling. J: OHCs torn o¡ from the organ of Corti during the tissue preparation can be seen to have some taurine-like immunolabeling associated uniquely with the synaptic pole (large asterisk) and the apex of the cells (asterisks).
HEARES 2812 27-8-97
K.C. Horner, C. Aurousseau / Hearing Research 109 (1997) 135^142
HEARES 2812 27-8-97
137
138
K.C. Horner, C. Aurousseau / Hearing Research 109 (1997) 135^142
min and the animals were awake and moving about the cage with no sign of vestibular disturbance after about 30 min. 2.3. Glycerol administration
The glycerol solution employed consisted of 600 g of glycerine made up to 1 l with distilled water. The dose level employed was 6 g/kg body weight given orally once and 30 min before the animals were killed. We have shown earlier that at this dose level changes in auditory function in guinea pigs occur which reach maximum after 30 min and recover completely at 24 h (Horner and Cazals, 1987). 2.4. Preparation of tissue for immunocytochemistry
The animals were anaesthetised (1 ml/kg) with an intramuscular injection of Ketalar-Rompun mixture in the ratio 2:1 (ketamine hydrochloride 50 mg/ml; xylazine 2%). Following decapitation the temporal bones removed, perfused via the round and oval windows with 2.5% glutaraldehyde/1% paraformaldehyde in 0.1 M phosphate bu¡er at pH 7.4. In some cases the apex of the cochlea was opened before the perfusion was commenced. The time between decapitation and perfusion was up to 5 min for the two ears. The temporal bones were left bathing in the same ¢xative for 24 h. The inner ears were then perfused with the phosphate bu¡er and left again for 24 h. The organ of Corti was dissected out from the temporal bones. The tissue was embedded in Tissue-Tek (Miles Inc.) and frozen to 350³C in isopentane on dry-ice. Frozen sections of 10 Wm were made at 320³C, collected on glass slides, dried rapidly on a heated plate to 40³C and frozen at 320³C until use. 2.5. Immunohistochemistry
For immunocytochemistry the slides were (1) washed twice, 15 min each, in phosphate bu¡er (0.01 M, pH 7.4) with bovine serum albumen (0.2%) and NaCl (0.9%) (PBS/BSA/NaCl); (2) washed for 5 min in 3% H2 O2 in methanol 100% (1:4) for 5 min in order to inhibit endogenous peroxidases; (3) washed twice for 5 min in PBS/BSA/NaCl; (4) incubated at room temperature for 1 h in PBS/BSA/NaCl with 0.3% Triton-X and 10% normal goat serum (Chemicon International, CA) to permeabilize the tissues and to lower non-speci¢c background activity; (5) exposed for 24 h at 4³C to polyclonal antibody to taurine from the rabbit (Chemicon International CA) in PBS/BSA/NaCl with 0.3% Triton-X and 1% normal goat serum; (6) washed twice, 15 min each, in PBS/BSA/NaCl; (7) incubated at room temperature 1 h in goat anti-rabbit IgG conjugated to peroxidase (dilution 1:50) (Chemicon International,
CA) in PBS/BSA/NaCL with 0.3% Triton-X and 1% normal goat serum; (8) exposed to rabbit peroxidase anti-peroxidase (dilution 1:400) (Sigma) for 1 h; (9) washed twice in PBS; (10) incubated in diaminobenzidine and H2O2 (Sigma fast) in PBS for precisely 8 min; (11) washed twice in PBS; (12) quickly air dried under a cool ventilator; (13) mounted in Entellan (Merck) and coverslipped. Normal rabbit serum for control experiments was also employed (Chemicon International, CA). An initial series of the test batches (n = 10) consisted of two slides (2U5 sections) from each of the two test conditions and the normal condition as well as control slides for which the primary antibody against taurine was omitted. For speci¢city controls additional experiments were carried out (n = 10) in which the diluted taurine antiserum was pre-adsorbed with taurine at different concentrations (250^25 mM). Immunohistochemistry procedures were performed simultaneously on test sections (with the taurine antiserum) and control sections (pre-adsorbed). In addition a series of slides were exposed to normal rabbit serum at di¡erent dilutions (1:300^1:2400) and compared with the immunohistochemical reaction of the taurine antiserum at the same dilutions. 3. Results
3.1. Taurine-like immunolabeling in the normal organ of Corti
In absence of the taurine antiserum no immunoreactivity was observed (Fig. 1A). Similarly pre-adsorption with taurine abolished all labeling. While some weak unspeci¢c background staining was observed when the taurine antiserum was replaced by the normal serum at low dilutions (1:300, 1:600) it was eliminated at higher dilutions (1:1200, 1:1800, 1:2400). At these high dilutions of the taurine antiserum a speci¢c taurine-like immunoreaction was still observable. In test sections of the normal organ of Corti the most intense taurine-like immunolabeling was speci¢cally associated with supporting cells (Fig. 1B). No difference in degree and/or distribution of immunolabeling from di¡erent cochlear turns was detected although observations were more easily made in the three upper turns where the hair cells are longer. Intense taurine-like immunoreactivity was associated with the cup-like portion of the Deiters' cells which is in close contact with the synaptic pole of the outer hair cells (OHC, Fig. 1D) and the Hensen cells towards the outer edge of the organ. Similar intense taurine-like immunoreactivity was seen within the inner phalangeal cells and the inner border cells which together surround the inner hair cells (IHC, Fig. 1C) as well as the inner
HEARES 2812 27-8-97
K.C. Horner, C. Aurousseau / Hearing Research 109 (1997) 135^142
sulcus cells towards the inner edge of the organ (Fig. 1B). There was, in contrast, a remarkable absence of immunolabeling associated with the OHCs (Fig. 1D). Some localised taurine-like immunolabeling could be observed on the reticular lamina but appeared to be associated with the contours of the apex of the cell rather than with the cuticular plates of the OHCs. The present light microscope study could not con¢rm the absence of taurine-immunolabeling from the IHCs since they are surrounded by taurine-like immunolabeling supporting cells, although in some sections there did appear to be less taurine-like immunolabeling in the IHCs (Fig. 1E). The tectorial membrane overlying the apical pole of the sensory hair cells showed no immunoreaction (Fig. 1B,C). The tectorial membrane appeared to be attached to supporting cells on the inner sulcus side of the IHCs which were probably border cells (Fig. 1C). These border cells were highly taurine immunoreactive. In addition there was little or no taurine-like immunolabeling associated with the outer pillar cells. 3.2. E¡ect of endolymphatic hydrops
The presence of hydrops was always con¢rmed at dissection and identi¢ed as the increased volume of the endolymph-¢lled scala media and the congruent diminution of the perilymph-¢lled scala vestibuli. The Reissner's membrane separating the two scala was con¢rmed to be extending into the scala vestibuli. Within the ¢rst 2 months of endolymphatic hydrops the distribution of taurine-like immunoreactivity appeared similar to that observed in the normal organ of Corti. In early hydrops the OHCs remained distinctly taurine free (Fig. 1E). The phalangeal process of the Deiters' cells which, in normal control cochleae were not immunoreactive to taurine and di¤cult to detect by light microscopy, could in hydropic cochleae, appear swollen and taurine immunoreative in particular in association with the outermost row of OHCs (Fig. 1E). In cochleae with hydrops of more than 5 months the Reissner's membrane was clearly in a distended vertical position and the tectorial membrane could also be observed to be elevated (Fig. 1F). In general, the tectorial membrane did not appear to be taurine-immunoreactive. The membrane could appear to be taurine immunoreactive but this was interpreted as artefactual when the tissue was clearly folded. The cells located under the reticular lamina which were normally taurine-free were now intensely immunoreactive. Close microscopic observation of the tissue suggested strongly that the immunoreactivity was associated not with OHCs but rather with the swollen phalangeal processes of the Deiters' cells which replaced the degenerating hair cells (Fig. 1G). This interpretation was reinforced by the
139
observation of the extensive immunoreaction throughout the reticular lamina in long-term hydrops. 3.3. E¡ect of glycerol
Thirty minutes following the oral administration of glycerol the tectorial membrane was observed to become taurine immunoreactive (Fig. 1H). In addition, the tectorial membrane appeared more tightly bound to the reticular lamina (Fig. 1I). The OHCs remained distinctly non immunoreactive at the glycerol dose tested (Fig. 1J). 4. Discussion
The present study has con¢rmed a recent report that in the organ of Corti of the cochlea, taurine-like immunoreactivity is normally associated with supporting cells (Usami and Ottersen, 1995). The observations are in agreement with earlier autoradiographic data (Schwartz and Ryan, 1983) and in con£ict with an earlier report of taurine speci¢cally within the OHCs and very little within supporting structure (Harding and Davies, 1993). A number of possible explanations can be proposed in order to account for the discrepancy. First, it should be recalled that the antibody against taurine employed by Harding and Davies (1993) was a monoclonal antibody while the present study employed a polyclonal antibody. Second, Harding and Davies carried out pre-embedding immunocytochemistry on the whole organ of Corti while in the present study immunocytochemistry was carried out on 10 Wm frozen sections. Third, the ¢xative (4% formaldehyde) employed by Harding and Davies has been reported to retain less than 1% of the free amino acid pool (Storm-Mathisen and Ottersen, 1990). Usami and Ottersen (1995) have pointed out that the greater the loss of free amino acid during ¢xation, the greater is the likelihood that the immunostaining will diverge from the in-vivo distribution. In the present study the absence of taurine from the IHCs as seen in transmission microscopy (Usami and Ottersen, 1995) could not be con¢rmed by light microscopy since the cells are surrounded by taurine-like immunoreactive supporting cells. This, on the other hand, is the ¢rst study demonstrating the localisation of taurine in the swollen phalangeal process of the Deiters's cells following hair cell loss. The present data might have signi¢cant implications for studies on isolated hair cells in vitro where the hair cell/Deiters' cell association is disrupted and the natural environment is highly altered. In addition, the data may provide a basis for a better understanding of the repair mechanism of the organ of Corti following damage which involves the swelling of the phalangeal process of the Deiter's cells.
HEARES 2812 27-8-97
140
K.C. Horner, C. Aurousseau / Hearing Research 109 (1997) 135^142
4.1. No
taurine-like
immunolabeling
in
the
outer
hair
cells of the organ of Corti
The absence of taurine from healthy OHCs described here by light microscopy is particularly striking. This absence from the OHCs is certainly not due to a possible leakage of taurine from the hair cells due to the triton treatment employed here. Similar distribution of taurine in the organ of Corti has been reported recently by Usami and Ottersen (1995) who did not employ triton. The possible leakage of taurine due to a delay to ¢xation can also be discounted. Our rapid ¢xation technique ensures that the round and the oval windows are fully exposed and guarantees a rapid and e¤cient ¢xation of both ears within 5 min maximum of decapitation. In the present study opening of the apex of the cochlea was not found to in£uence the taurine-like immunolabeling. An alternate ¢xation technique of invivo perfusion of the cochleae, is not beyond criticism. Deep and relatively prolonged anaesthesia is required in order to e¤ciently expose the windows by a ventral approach to the bullae and so the possible complications due to the e¡ect of anaesthesia and surgical stress cannot be ruled out. Having said that, such an in-vivo ¢xation procedure has been employed by Usami and Ottersen (1995) and they have reported the absence of taurine from the sensory hair cells. The observed localisation of taurine uniquely in supporting cells is probably related to the distribution of the many gap junctions between all supporting cells of the organ of Corti while being absent between supporting and sensory cells (Iurato et al., 1976; Kikuchi et al., 1995). 4.2. Hair
cell
motility
requires
an
e¤cient
protection
against the osmotic stress
The OHCs of the organ of Corti, change shape on depolarisation/repolarisation (Dallos et al., 1993), at least in vitro, and therefore belong to a specialised category of cells whose osmoregulating facilities can cope with cell motility. A `slow motile response' of hair cells associated with swelling can be provoked by exposing the cells in vitro to a milieu rich in K or by lowering the osmolarity. The shortening is reported to be associated with a volume increase of up to 20% (Dulon et al., 1988; Ulfendahl, 1988). However, it should be recalled that this volume change is observed for hair cells isolated from their supporting cells in an arti¢cial medium whose composition is experimentally modi¢ed. The in-vivo situation is not likely to present such extreme conditions. A `fast motile response' with maintenance of cell volume, involves cyclic elongation^contraction when electrically stimulated in vitro (Brownell et al., 1985) and is presumed to mimick the hair cell response to sound stimulation in vivo. The precise
mechanisms responsible for the fast motile response have not been established but considered at this time to be `molecular motors' in the plasma membrane (Kalinec et al., 1992) working in association with the underlying cortical lattice involving the interaction of actin and spectrin (Holley and Ashmore, 1990) and with the underlying lateral cisternae (Brownell, 1986). It is also assumed that the elastic restoring force is dependent on the intracellular turgor pressure since the fast motility deteriorates when the intracellular pressure is reduced (Brownell and Shehata, 1990). Indeed it has been speculated that modi¢cation of intracellular pressure in vivo may provide a mechanism for controlling the gain of the mammalian `cochlear ampli¢er' (Kakehata and Santos-Sacchi, 1995). Harding and Davies (1993) who observed taurine within the hair cells postulated that taurine might be implicated in hair cell motility. The present study demonstrated the absence of taurine from hair cells and so the data here suggest that hair cells do not rely directly on taurine as an osmolyte in order maintain the intracellular pressure. The shunting of taurine probably occurs over a prolonged time scale which would appear more appropriate for the supporting cells, functioning as a bu¡ering protective system for the hair cells. 4.3. The swelling of the phalangeal process of the Deiters' cells is associated with taurine-like immunolabeling
We have demonstrated that in the normal organ of Corti the cup-like body of Deiters' cells has high taurine-like immunolabeling while the phalangeal process appears not to have taurine. Within the ¢rst 2 months of hydrops the localisation of taurine does not appear to be considerably modi¢ed. This apparent constancy of taurine localisation is observed despite a moderate decrease in the osmolality (Sziklai et al., 1992) and the accumulation of electron dense bodies in the subcuticular cytoplasm, thought to be lipofuschin, which can be observed already 1 month after induction of hydrops (Horner and Guilhaume, 1995). In long-term hydrops we have observed an accumulation of taurine in the swollen phalangeal process of the Deiter's cells and in particular the outermost cells are a¡ected ¢rst. It is should be recalled that in hydropic cochleae there is a selective loss of short and middle stereocilia in the three upper cochlear turns (Horner et al., 1988) which begins with the outermost row (Rydmarker and Horner, 1990). It has been speculated that the particular pattern of atrophy of stereocilia starting from the outermost row is due to the abnormal elevation of the tectorial membrane which would a¡ect the outermost hair cells ¢rst (Rydmarker and Horner, 1990). The loss of hair cells is delayed relative to the stereocilia loss but follows the same pattern with the outermost hair cells a¡ected ¢rst. The swelling of the
HEARES 2812 27-8-97
K.C. Horner, C. Aurousseau / Hearing Research 109 (1997) 135^142
141
phalangeal process of the Deiter's cells which begins
motic stress conditions, which would be facilitated by
with the outermost cells is therefore almost certainly
the coupling of the tectorial membrane to the inner hair
related to the damage to the associated hair cell.
cell border cells as observed here. Such a coupling in
Swelling of cells
was
the phalangeal
¢rst
following
observed
antibiotic
in
process of
damaged
treatment
the
organs
Deiters' of
vivo, however, remains a subject of debate (Lawrence
Corti
and Burgio, 1980), since most histological procedures
and
are known to a¡ect the ¢ber composition of the tecto-
(Hunter-Duvar
Mount, 1978 ; Forge, 1985). The swelling of the phalan-
rial membrane (Hasko and Richardson, 1988).
geal process of the Deiter's cell is clearly a means of
The observed tighter coupling between the tectorial
¢lling the space left by a dying hair cell in order to
membrane and the organ of Corti following intrave-
conserve the structure of the organ of Corti and could
nous injection of 2 g/kg glycerol has been reported ear-
therefore be considered as participating in programmed
lier (Meyer zum Gottesberge and Tsujikawa, 1993) to
cell death. The demonstration here of taurine accumu-
be related to a glycerol-induced decrease in Ca
lation within the swollen phalangeal processes of the
lier studies have demonstrated that the un¢xed tectorial
Deiter's
to
membrane maintained in arti¢cial endolymph is highly
hydropic ears but might represent a mechanism which
sensitive to calcium-swelling in solution containing the
initiates or accompanies swelling. Whether the £ux of
Ca
taurine is due to passive di¡usion or active uptake by
ing
the phalangeal processes cannot be resolved here. As
1979 ; Freeman et al., 1994). The taurine immunolabel-
mentioned earlier following intraperitoneal injection of
ing of the tectorial membrane following glycerol admin-
distilled water in rats, there is decreased taurine immu-
istration might thus represent a protective measure in
nolabeling of cerebellar Purkinje cells and increased im-
order to augment the local concentration of Ca
munolabeling in neighboring glial cells which become
has
swollen (Nagelhus et al., 1993).
Ca
An
cells
is
probably
osmoregulatory
not
role
for
a
feature
taurine
particular
in
2
2
chelator ethylenediamine-tetraacetate and shrink-
on
the
been
2
. Ear-
re-introduction
established
availability
association
protects
with the Deiter's cells is in keeping with recent identi-
high Ca
against
2
that
under Ca
2
of
Ca
2
taurine
conditions overload
(Kronester-Frei,
can of
both low
under
2
. It
increase
2
Ca
and
conditions
of
(Huxtable, 1992).
¢cation of a structure unique to these cells named the
The tighter coupling of the tectorial membrane to the
`rosette complex' (Spicer and Schulte, 1993). Immuno-
organ of Corti following glycerol administration might
histochemistry has revealed ci¢c
to
the
meshwork has
led
to
rosette
which the
Q
complex
is
muscle actin isoform spe-
account to some extent for the reported temporary im-
and
è nie é re's patients. Howprovement in hearing in some Me
enclosed
immunoreactive
proposal
that
the
to
in
a
loose
vimentin
Deiter's
cells
and may
è nie é re's disease the tectoever in the animal model of Me rial
membrane
has
been
reported
to
be
elevated
play a role in regulating ion homeostasis and/or motile
(Rydmarker and Horner, 1990 ; Meyer zum Gottesberge
response properties of the organ of Corti (Nakazawa et
and Tsujikawa, 1993) and Ca
al., 1995). In addition the restructuration of the organ
noyu and Meyer zum Gottesberge, 1986), yet adminis-
of Corti almost certainly involves a modi¢cation of the
tration of glycerol at di¡erent dose levels including 6 g/kg
e¡erent feedback simultaneously to supranuclear termi-
failed to improve the compound action potential audio-
nals outer
de-
gram (Horner and Cazals, 1987). It remains to be deter-
scribed in the cat (Liberman et al., 1990) and in the
¨mined whether the animal model is representative of Me
human organ of Corti (Nadol and Burgess, 1994).
é re's patients or of a particular sub-group of patients nie
hair
cells
and to
the
Deiters'
cells
as
2
is highly elevated (Ni-
who present no improvement or a hearing deterioration
4.4. Glycerol induces taurine-like immunolabeling within the tectorial membrane
following glycerol administration (Barbara et al., 1994).
Acknowledgments
The present study has shown that in the normal organ of Corti there is no taurine-like immunolabeling associated with the tectorial membrane. Following the
This
research
received
¢nancial
support
from
the
ègional d'Aquitaine. Conseil Re
oral administration of 6 g/kg glycerol some taurine-like immunolabeling
was
brane
tectorial
and
the
seen
within
the
membrane
tectorial appeared
memto
be
more closely bound to the organ of Corti. Two possible interpretations
of
the
immunolabeling
can
be
made.
First glycerol might modify the constitution of the tectorial membrane and so, as in any damaged tissue, it might become more immunoreactive. The other interpretation would favour a delocalisation of taurine from the supporting cells into the tectorial membrane, in os-
References Altshuler, D., Lo-Turco, J.J., Rush, J. and Cepko, C. (1993) Taurine promotes the di¡erentiation of a vertebrate retinal cell type in vitro. Development 119, 1317^1328. Barbara, M., Monini, S., Attanasio, G. and Filipo, R. (1994) Glycerol test : considerations on the prognostic value in the long-term apè nie é re's patients. In : R. Filipo and M. Barbara (Eds.), proach to Me è nie é re's Disease, Kugler, Amsterdam, pp. 127-129. Me Brownell, W.E. (1986) Outer hair cell motility and cochlear frequency
HEARES 2812 27-8-97
K.C. Horner, C. Aurousseau / Hearing Research 109 (1997) 135^142
142
selectivity. In : B.C.J. Moore and R.D. Patterson (Eds.), Auditory Frequency Selectivity, Plenum, New York, pp. 109-116.
Kalinec, F., Holley, M.C., Iwasa, K.H., Lim, D.J. and Kachar, B. (1992) A membrane-based force generation mechanism in auditory
Brownell, W.E., Bader, C.R., Bertrand, D. and de Ribaupierre, Y. (1985) Evoked mechanical responses of isolated cochlear hair cells. Science 227, 194^196.
sensory cells. Proc. Natl. Acad. Sci. 89, 8671^8675. Kikuchi, T., Kimura, R.S., Paul, D.L. and Adams, J.C. (1995) Gap junctions in the rat cochlea : immunohistochemical and ultrastruc-
Brownell, W.E., and Shehata, W.E. (1990) The e¡ect of cytoplasmic
tural analysis. Anat. Embryol. Berl. 191, 101^118.
turgor pressure on the static and dynamic mechanical properties of
Kimura, R.S. and Schuknecht, H.F. (1965) Membranous hydrops in
outer hair cells. In : P. Dallos, C.D. Geisler, J.W. Matthews, M.A.
the inner ear of the guinea pig after obliteration of the endolym-
Ruggero, and C.A. Steele (Eds.), Mechanics and Biophysics of Hearing. Springer, New York, pp. 52-59.
phatic sac. Pract-Otorhinolaryngol. 27, 343^354. Klockho¡, I. and Lindblom, U. (1966) Endolymphatic hydrops as
Dallos, P., Hallworth, R. and Evans, B.N. (1993) Theory of electrically driven shape changes of cochlear outer hair cells. J. Neurophysiol. 70, 299^323.
revealed
by
the
glycerol
test.
Acta.
Otolaryngol.
(Stockh.)
61,
459^462. Kronester-Frei, A. (1979) The e¡ect of changes in endolymphatic ion
Davies, W.E., and S. Owen (1985) The nature of neurotransmitters in the mammalian lower auditory system. In : D.G. Drescher (Ed.), Auditory Biochemistry, CC Thomas, Spring¢eld, IL, pp. 244-257.
concentrations on the tectorial membrane. Hear. Res. 1, 81^94. Lake, N. (1994) Taurine and GABA in the rat retina during postnatal development. Vis. Neurosci. 11, 253^260.
Dulon, D., Aran, J.M. and Schacht, J. (1988) Potassium-depolarisa-
Lawrence, M. and Burgio, P.A. (1980) Attachment of the tectorial
tion induces motility in isolated outer hair cells by an osmotic
membrane revealed by scanning microscopy. Ann. Otol. Rhinol.
mechanism. Hear. Res. 32, 123^130.
Laryngol. 89, 325^330.
Forge, A. (1985) Outer hair cell loss and supporting cell expansion following chronic gentamicin treatment. Hear. Res. 19, 171^182. Freeman, D.M., Cotanche, D.A., Ehsani, F. and Weiss, T.F. (1994) The osmotic responses of the isolated tectorial membrane of the
,
chick to isosmotic solutions : e¡ect of Na
K
2
and Ca
concen-
tration. Hear. Res. 79, 197^215.
Liberman, M.C., Dodds, L.W. and Pierce, S. (1990) A¡erent and e¡erent innervation of the cat cochlea : quantitative analysis with light and electron microscopy. J. Comp. Neurol. 301, 443^460. Lombardini, J.B. (1991) Taurine : retinal functions. Brain Res. Rev. 16, 151^169. Meyer
Hageman, G.S. and Schmidt, S.Y. (1987) Taurine de¢cient pigmented and albino rats : early retinal abnormalities and di¡erential rates of photoreceptor degeneration. Prog. Clin. Biol. Res. 247, 497^515. Harding, N.J. and Davies, W.E. (1993) Cellular localisation of taurine in the organ of Corti. Hear. Res. 65, 211^215.
zum
Gottesberge,
A.M.
and
Tsujikawa, S.
(1993)
Glycerol
e¡ect on the guinea pig tectorial membrane. Eur. Arch. Otorhinolaryngol. 250, 88^91. Nadol, J.B. and Burgess, B.J. (1994) Supranuclear e¡erent synapses on outer hair cells and Deiters' cells in the human organ of Corti. Hear. Res. 81, 49^56.
Hasko, J.A. and Richardson, G.P. (1988) The ultrastructural organ-
Nagelhus, E.A., Lehmann, A. and Ottersen, O.P. (1993) Neuronal-
isation and properties of the mouse tectorial membrane matrix.
glial exchange of taurine during hypo-osmotic stress : a combined
Hear. Res. 35, 21^38.
immunocytochemical
Hayes, K.C., Carey, R.E. and Schmidt, S.Y. (1975) Retinal degeneration association with taurine de¢ciency in the cat. Science 188, 947^951.
the
biochemical
analysis in
rat
cerebellar
Nakazawa,
K.,
Schultz,
B.A.
and
Spicer,
S.S.
(1995)
The
rosette
complex in gerbil Deiters cells contain Y actin. Hear. Res. 89,
Holley, M.C. and Ashmore, J. (1990) Spectrin, actin and the structure of
and
cortex. Neuroscience 54, 615^631.
cortical
lattice
in
mammalian
cochlear
outer
hair
cells.
J. Cell Sci. 96, 283^291.
121^129. Ninoyu, O. and Meyer zum Gottesberge, A.M. (1986) Changes in
2
Ca
Horner, K.C. (1993) Review : functional changes associated with ex-
activity and DC potential in experimentally induced endo-
lymphatic hydrops. Eur. Arch. Otorhinolaryngol. 243, 106^107.
perimentally induced endolymphatic hydrops. Hear. Res. 58, 1^18.
Pow, D., Crook, D.K. and Wong, R.O. (1994) Early appearance and
Horner, K.C. and Cazals, Y. (1987) Glycerol induced changes in the
transient expression of putative amino acid neurotransmitters and
Arch.
related molecules in the developing rabbit retina : an immunocy-
Horner, K.C. and Guilhaume, A. (1995) Ultrastructural changes in
Rydmarker, S. and Horner, K.C. (1990) Morphological changes of
the hydropic cochlea of the guinea pig. Eur. J. Neurosci. 7, 1305^
hair cell stereocilia and tectorial membrane in guinea pig with
cochlear
responses
of
the
guinea
pig
hydropic
ear.
Eur.
tochemical study. Vis. Neurosci. 11, 1115^1134.
Otorhinolaryngol. 244, 49^54.
experimentally induced hydrops. Scan. Microsc. 4, 705^714.
1312. Horner, K.C., Guilhaume, A. and Cazals, Y. (1988) Atrophy of mid-
Schwartz, I.R. and Ryan, A.F. (1983) Di¡erential labeling of sensory
dle and short stereocilia on outer hair cells of guinea pig labeling
cell and neural populations in the organ of Corti following amino
with experimentally induced hydrops. Hear. Res. 32, 41^48. Horner, K.C., Huang, W. and Erre, J.P. (1993) The e¡ect of taurine modi¢ed diet on normal and hydropic
ears of the guinea pig.
Hear. Res. 70, 1^8. Hunter-Duvar, I. and Mount, R.J. (1978) The organ of Corti following ototoxic antibiotic treatment. Scan. Elect. Microsc. 11, 423^430. Huxtable, R.J. (1992) Physiological actions of taurine. Physiol. Rev. 72, 101^163.
acid incubations. Hear. Res. 9, 185^200. Spicer, S.S. and Schulte, B.A. (1993) Cytologic structures unique to Deiters cells of the cochlea. Anat. Rec. 237, 421^430. Storm-Mathisen, J. and Ottersen, O.P. (1990) Antibodies and ¢xatives for the immunocytochemical localisation of glycine. In : J. StormMathisen and O.P. Ottersen (Eds.), Glycine Neurotransmission, John Wiley, Chichester, pp. 281-301. Sziklai, I., Ferrary, E., Horner, K.C., Sterkers, O. and Amiel, C.
Imaki, H., Jacobson, S.G., Kemp, C.M., Knighton, R.W., Neuringer,
(1992) Time-related alterations of endolymph composition in an
M. and Sturman, J. (1993) Retinal morphology and visual pigment
experimental model of endolymphatic hydrops. Laryngoscope 102,
levels in 6 and 12 month old rhesus monkeys fed a taurine free human infant formula. J. Neurosci. Res. 36, 290^304. Iurato, S., Franke, K., Luciano, L., Wermbter, G., Pannese, E. and Reale, E. (1976) Intercellular junctions in the organ of Corti as re-
431^438. Ulfendahl, M. (1988) Volume and length changes in outer hair cells of the
guinea
pig
after
potassium-induced
shortening.
Eur.
Arch.
Otorhinolaryngol. 245, 237^243.
vealed by freeze fracturing. Acta. Otolaryngol. (Stockh.) 82, 58^69.
Usami, S. and Ottersen, O.P. (1995) The localization of taurine-like
Kakehata, S. and Santos-Sacchi, J. (1995) Membrane tension directly
immunoreactivity in the organ of Corti : a semiquantitative, post-
shifts voltage dependence of outer hair cell motility and associated
embedding immuno-electron microscopic analysis in the rat with
gating charge. Biophys. J. 68, 2190^2197.
some observations in the guinea pig. Brain Res. 676, 277^284.