Movement of monovalent ions across the membranes of marginal cells of the stria vascularis in the guinea pig cochlea

The resting potential and monovalent ions in the marginal cells and scala media were measured before and 20 min after the onset of anoxia using ion-sensitive microelectrodes. The resting potential of the marginal cells decreased from 62.7 to -2.4 mV. The K+ activity decreased from 77.7 to 53.2 mEq/l, while the Na+ activity increased from 2.6 to 24.7 mEq/l. The Cl- activity did not change significantly. In the scala media, the endocochlear potential decreased rapidly from 80.9 to -28.0 mV after the onset of anoxia. The K+ activity decreased from 119.0 to 96.5 mEq/l, the Na+ activity increased from 1.3 to 9.5 mEq/l and that of Cl- decreased from 127.0 to 115.1 mEq/l. The electrochemical gradients determined for each ion based on the ionic changes in the scala media and marginal cells, suggested the existence of an Na/K pump and Na-K-2Cl cotransport at the basolateral membrane of the marginal cells, and a rheogenic K pump and Na-K-2Cl transport at the luminal membrane of the marginal cells. The Na+ and K+ must be recycled at the basolateral membrane and luminal membrane of marginal cells, respectively.

N-ethylmaleimide-inhibited electrogenic K+ secretion in the ampulla of the frog semicircular canal

1. The mechanisms of K+ secretion into endolymph were studied on a preparation of isolated semicircular canal with different pharmacological inhibitors. Three periods of 5 or 30 min were performed, the first as control, the second in the presence of the drugs added to the apical or the basolateral bathing solution, and the third as recovery. Apical fluid was sampled at the beginning and the end of each period, transepithelial potential was recorded, Na+, K+, and Cl- concentrations, and K+ efflux, with 86Rb+ as a tracer, were measured and K+ fluxes were calculated. 2. When both sides of the epithelium were bathed with perilymph-like solution, the epithelium absorbed Na+, secreted K+, and generated a lumen positive potential. 3. The ATPases inhibitors, ouabain (10(-5) and 10(-3) M) and N-ethylmaleimide (10(-4) and 10(-3) M) inhibited the electrogenic K+ secretion when added to the basolateral fluid. N-ethylmaleimide (10(-3) M) applied to the apical fluid during a 5 min period decreased the K+ influx by 43% and the transepithelial potential by 66%. Other ATPase inhibitors, harmaline (10(-3) M), omeprazole (10(-4) M), vanadate (10(-4) M and 10(-3) M), N,N'-dicyclohexylcarbodiimide (DCC, 10(-5) M), 7-chloro-4-nitrobenz-2-oxa-1,3-diazole (NBD-Cl, 5 x 10(-6) M and 5 x 10(-5) M), and bafilomycin (10(-7) M) did not affect the K+ transport nor the transepithelial potential when they were added to the apical fluid. 4. The Na(+)-K(+)-Cl- co-transporter inhibitor, bumetanide, decreased both the transepithelial potential and the K+ transport when added to the basolateral solution but not to the apical one. At 10(-6) M, bumetanide maximally decreased the K+ influx by about 60%. 5. K+ channel blockers, quinine (10(-4) M), TEA (5 x 10(-3) M), added to the apical solution and barium (2 x 10(-3) M) added to either the apical or the basolateral solutions, did not affect the K+ transport and the transepithelial potential. 6. The carbonic anhydrase inhibitor acetazolamide (10(-3) M) added to both apical and basolateral solutions did not affect the K+ transport and the transepithelial potential. 7. It is concluded that, in the ampulla of the semicircular canal, a basolateral Na(+)-K(+)-Cl- co-transporter energized by the Na+, K(+)-ATPase was involved for 60% in the K+ secretion into endolymph. The electrogenic K+ transport would partly depend on a N-ethylmaleimide-sensitive protein possibly located at the apical plasma membrane or intracellularly.

[Contribution of cell culture to the study of cochlear endolymph]

Investigation of the mechanisms involved in the secretion of cochlear endolymph is hindered by the heterogeneity, the fragility, and the difficulty of access to the secretory structures of the inner ear, i.e. stria vascularis and Reissner's membrane. Cell culture may overcome these limitations and allow the study of the epithelial transport processes. Cultured cells derived from gerbil's explants of stria vascularis and Reissner's membrane presented morphological and immunohistochemical properties similar to those of the native cells. Investigation by the patch clamp technique of the apical membranes of marginal-like cells showed the presence of non specific cationic channels that could allow passive K+ flow towards endolymph and, thus, play a major role in the production of endolymph. Biochemical studies demonstrated that a cAMP production by these marginal-like cells. Primary culture of stria vascularis and Reissner's membrane is a suitable model to further investigate the physiology of endolymph production.

Pharmacokinetics of furosemide in endolymph

To investigate the pharmacokinetics of organic anions in the endolymph of the guinea pig, 100 mg/kg furosemide, an organic anion, was intravenously given to measure the concentration in the cochlear endolymph by high-performance liquid chromatography with fluorescence detection. In the endolymph, the concentration of the furosemide increased slowly for 1 hr to 1.6 micrograms/ml and gradually declined thereafter. Pretreatment with 200 mg/kg probenecid, an anion transport inhibitor, had no effect on the furosemide elimination in the endolymph except on the concentration at 2 hr. This was contrary to the drastic change observed in the perilymph of the scala tympani by the same pretreatment. Analogous to the effect in the endolymph, probenecid showed no change in the concentration of the serum, while a pronounced gradient of furosemide concentration existed between them. The present results suggest that the furosemide passively transfers from blood to the endolymph at a relatively low penetrability.

Histochemical localization of calcium ATPase in the cochlea of the guinea pig

The activity of Ca(2+)-ATPase in the inner ear of the guinea pig was studied ultracytochemically by the lead citrate reaction. The electron-dense reaction products as an expression of Ca(2+)-ATPase activity were localized in endolymphatic cells of Reissner's membrane, in outer and inner hair cells and in some supporting cells. The main finding was the difference in the localization of Ca(2+)-ATPase in outer and inner hair cells. In the latter cells the activity sites were mainly intracellular and in apical membrane specializations, whereas in the outer hair cells the enzyme was localized in the apical membrane specializations and the basolateral plasma membrane.

Secretion of endolymph by the isolated frog semicircular canal

Secretion of endolymph is localized in some structures of the inner ear, namely the stria vascularis in the cochlea and the dark cells in the vestibule and in the lower vertebrate inner ear. In isolated semicircular canal it is possible to study separately the endolymphatic composition in the ampulla, which contains the dark cells, and in its non-ampullar part, which is devoid of these cells. Further, in vitro preparation of the semicircular canal provides access to both faces of the epithelium so that different agents can be applied separately to the apical or to the basolateral membranes of the epithelium. In this structure, the following results were obtained: i) in vitro, the semicircular canal secreted a K-rich, positively polarized fluid; ii) this fluid was secreted only in the ampulla of the semicircular canal; iii) the secretion of endolymph was dependent on basolateral Na+, K(+)-ATPase, inhibited by ouabain, and basolateral Na-K-Cl co-transporter, inhibited by bumetanide; iv) approximately 60% of luminal Na absorption occurred across a luminal Na channel inhibited by amiloride; v) the permeability of the paracellular pathway of the semicircular canal epithelium was 7.10(-7) cm/s. These results indicate that endolymph secretion involves basolateral Na+, K(+)-ATPase and Na-K-Cl co-transporter. An Na channel has been shown at the apical membrane.

Primary culture of strial marginal cells of guinea pig cochlea: growth, morphologic features, and characterization

To further investigate the cellular mechanisms involved in the formation of endolymph, primary cultures of marginal cells of guinea pig were established. Minute explants obtained by mechanical dissociation of stria vascularis were plated on collagen type I precoated impermeable substrate in serum-free, hormone-supplemented medium. A confluent layer of epithelial-like cells was obtained within 2 weeks. The cultured cells formed domes, demonstrating that they retain some of their transepithelial properties. Polarization was also suggested by electron microscopic observation of apical microvilli and tight junctions. Immunohistochemical methods revealed that the cultured cells coexpressed cytokeratin and vimentin, demonstrating their epithelial origin, although some degree of dedifferentiation occurred. Thus, a primary culture of marginal cells can be established that may be a suitable model for an in-depth investigation of the function of the marginal cells.

[Mechanisms of development of endolymphatic hydrops following secondary immune response in the endolymphatic sac of guinea pigs]

Previously the author reported the immediate development of endolymphatic hydrops (e. hydrops) following direct challenge of secondary antigen to the endolymphatic sac (ES) in guinea pigs, during the early phases of postchallenge, ranging from 1 to 5 weeks. The present study reports the results of specimens taken up to 10 weeks postchallenge, and correlation of e.hydrops to perilymph in antigen-specific antibody levels. From the present results, mechanisms of e.hydrops induced ES immune reaction are suggested as follows. 1) In the early stage of e.hydrops, an acute inflammatory reaction in the ES may produce endolymph by an increased vascular permeability of the inner ear and may impair endolymph absorption from the ES. 2) In the latter stage of e.hydrops, the moderate cellular infiltration in the ES may cause chronic impairment of endolymph absorption in the ES.

The potassium concentration in the cochlear fluids of the embryonic and post-hatching chick

The potassium concentration was measured in the perilymphatic and endolymphatic spaces of the chick's cochlea using ion-specific microelectrodes. The mean values in post-hatching chicks were 8.1 and 161 mM/l, respectively and are thus the same as in mammals. The high potassium concentration is reached at the latest at stage E42, and thus before the tegmentum vasculosum is fully developed. The endocochlear potential of chicks was maximally + 13 mV and thus somewhat below the published values for birds. Positive potentials in the endolymphatic space of 5 mV or more were already present in animals of age E41 and E42 and the developmental process is complete at the latest one day after hatching.

Transport of 5-hydroxydopamine and horseradish peroxidase through the perilymph-endolymph barrier

The distribution of 5-hydroxydopamine (5-OHDA) and horseradish peroxidase (HRP) in the cochlea was examined by light and electron microscopy after perilymphatic perfusion. Reaction product of 5-OHDA was observed along the endolymphatic surface of the cochlear duct in both 10-minute and 30-minute perfusions, while that of HRP was observed only in the 30-minute perfusion. Both tracers were found in pinocytotic vesicles of Reissner's membrane, Claudius' cells, and Hensen's cells. No labeling was seen in tight junctions of the epithelial cells lining the scala media. The epithelial cells of the spiral prominence showed various degrees of degeneration in the 5-OHDA experiment. These morphologic findings suggest that 5-OHDA and HRP can enter the endolymphatic space from the perilymphatic space by pinocytosis and that the degree of transport depends on the size of the particles.

Pilocarpine elicits the interdental cell secretory activity of the inner ear

Subcutaneous injection of pilocarpine in guinea pigs resulted in the following ultrastructural changes: 1) the apical cavities of the interdental cells were filled with a substance indistinguishable from the overlying amorphous layer of the TM; 2) a great number of spherical structures appeared over the limbal portion of the tectorial membrane. In TEM photomicrographs these structures displayed the same appearance as the amorphous layer of the TM and were usually continuous to it; 3) the number of holes that decorate the upper surface of the limbal portion of the TM was dramatically increased and it was found that they connect the endolymphatic space to the apical cavities of the interdental cells; 4) there was an increase in the number of the small extracellular vesicles found in the clear spaces of the tectorial membrane. These facts suggest that pilocarpine stimulates the secretion of the interdental cells, confirming the existence of the secretory processes previously described (Prieto et al., 1990). These findings can be related to the turnover of the TM in the adult animal and, perhaps, to the secretion of some organic compound to the endolymph. We postulate that the actions of pilocarpine on the interdental cells are most probably mediated by the activation of muscarinic acetylcholine receptors in these cells.

Carbohydrates in the guinea pig stria vascularis demonstrated with lectin-gold techniques

Soybean agglutinin (SBA), Helix pomatia agglutinin (HPA), Ricinus communis agglutinin-II (RCA-II), Limax flavus agglutinin (LFA) and wheat germ agglutinin (WGA) were employed to determine the localization of specific carbohydrates on thin sections of lowicryl K4M embedded guinea pig striae vasculares using the lectin-gold and glycoprotein-gold techniques. SBA, HPA and RCA-II gold labeling was observed in many of the cytoplasmic vesicles, endosomes and apical tubules located in the supranuclear region as well as on the microvilli and micropinocytotic invaginations of the luminal plasma membrane of the marginal cells. LFA labeling was found on the basal plasma membrane of the marginal cells as well as in the basement membrane of the perivascular spaces. WGA binding sites were detected along the plasma membrane of all types of cells constituting the stria vascularis. Our present results revealed that the membranes of internalization and many of the cytoplasmic vesicles, endosomes and apical tubules in the supranuclear region of the marginal cells are associated together and it is suggested that these structures may be related to the regulation of endolymph.

Ototoxic effect of potassium canrenoate on the guinea pig cochlea

Potassium canrenoate (PC) is a diuretic with antialdosterone action, reducing the reabsorption of sodium in the efferent renal tubules. This drug has been reported to damage only the marginal cells of the stria vascularis in the cochlea. The perilymphatic space of the guinea pig cochlea was perfused with an artificial perilymph containing 5 x 10(-3) M potassium canrenoate. Following the onset of perfusion, the endocochlear dc potential (EP) gradually declined to around 10 mV but did not become negative even when the perfusion was continued. A similar decrease in EP was observed in kanamycin-deafened guinea pigs. When the respirator was turned off and perfusion discontinued, a large negative EP appeared during 5 min of anoxia in normal guinea pigs but not in the kanamycin-deafened guinea pigs. The decreased EP recovered to preanoxic level after resumption of ventilation. The cochlear microphonics (CM) also gradually declined in parallel with the EP. The summating potential (SP) showed only a minor change. The potassium ion activity in the endolymph decreased slightly but the sodium ion activity remained unchanged during perfusion. These findings suggest that the main target of the PC is the cells of the stria vascularis.

[Immunohistochemical study of the endolymphatic hydrops due to secondary endolymphatic sac immune reaction]

Secondary endolymphatic sac (ES) immune response against keyhole limpet hemocyanin (KLH) were immunohistochemically investigated in guinea pigs by analysing the distribution of IgG and C3 complement in the inner ear over a 4-week period. After systemic sensitization with KLH and Freund's adjuvant, KLH was inoculated into the right ES by trans-cranial approach. KLH was found in the ES immediately after KLH challenge and then disappeared on the 7th day. On the 2nd and 3rd day moderate endolymphatic hydrops and a massive cell infiltrations within the ES were observed, and they gradually decreased their severity with time. Diffuse stainings with IgG and C3 were found in the ES, the epithelial and subepithelial regions of the vestibule. Similar findings were obtained in the spiral ligament and the hair cells after antigen challenge. IgG bearing cells were detected within the endolymphatic sac alone. Our results suggested that endolymphatic hydrops may be produced by the imbalance of in-out flow of the endolymph and the immunologically mediated inflammation through IgG-complement system.

[Pathohistological study on active endolymphatic hydrops in guinea pig--with stress stimulations]

Recently Meniere's disease is believed to show a pathological features of endolymphatic hydrops, but the etiology of this disease has not yet been ascertained. Many investigators tried to create animal models with endolymphatic hydrops by obliterating endolymphatic sac and duct. However, these methods have not been adequate to explain the mechanism of development of Meniere's disease, because it gives mechanical damage to destroy endolymphatic sac and duct. On the other hand, Meniere's disease is thought to develop even as a result of stress. The present paper reported the conduction of an acute experiment, creating active endolymphatic hydrops by stress load, to study the occurrence rate of endolymphatic hydrops and the pathological findings of inner ears by a light microscope. This was followed by a light and electron microscopic study on the changes in permeability of the cochlear lateral wall with horseradish peroxidase and a study of the development mechanism of endolymphatic hydrops. Results obtained were as follows: 1. Although it was impossible to cause endolymphatic hydrops with a single stimulation, four kinds of stress stimulations could cause endolymphatic hydrops with the rate of 37.8%. 2. Judging from the different pathological findings between the group with four kinds of stress stimulations and the untreated control group, vessel permeability in the stria vascularis was highly changeable, while that in the spiral ligament was not. 3. It was found out that increased permeability of the stria vascularis vessels was by increased pinocytotic vesicular transport and through tight junctions from vessel lumen to basal lamina and that it was presumably only by increased pinocytotic vesicular transport beyond basal lamina. 4. It was assumed that increased permeability of the stria vascularis vessels was one of the causes of endolymphatic hydrops which were believed to be pathological features of Meniere's disease.

[Clinical electrocochleography in Menière's disease]

Transtympanic electrocochleography (ECochG) was performed on 137 definite Menière ears, 29 normal ears and 383 ears with other diseases. The average SP/AP amplitude ratio (SP/AP) was significantly larger than that of normal subjects in the following four disease groups whose causes are thought to be endolymphatic hydrops: 1) definite Menière's disease, 2) suspected Menière's disease, 3) delayed endolymphatic hydrops and 4) labyrinthine syphilis. The SP amplitude could not provide any information for a differential diagnosis of endolymphatic hydrops from normal subjects nor from other sensorineural deafness . The upper 99% confidence limit of normal SP/AP using a 100dBSPL click was 0.37 for 29 subjects. In patients with Menière's disease the relationship between SP/AP and clinical data such as pure tone threshold, audiogram pattern, results of glycerol test and duration of illness from the first vertiginous attack to the test were investigated. In patients with severe hearing loss, the number of those showing SP/AP of greater than 0.37 was much higher than in patients with less severe hearing loss. There were no relationship, however, between SP/AP and duration of illness, or audiogram pattern, or glycerol test. ECochG was recorded before and after intravenous administration of 50 gr of glycerol (10% solution, 500ml), and a significant decrease of SP/AP was found in 21 Menière ears. However, there were no change of SP/AP after oral administration of 50 gr of glycerol (50% solution, 100ml) and isosorbide (70% solution, 2.0gr/kg of body weight). After endolymphatic sac surgery a decrease of SP/AP of 10% or more was found in 5 ears, an increase of SP/AP in 3 ears, and no change of SP/AP in 3 ears. In the patients showing no change of SP/AP after the surgery, numerical values of vertigo by AAOHNS were higher than in patients showing an increase or a decrease of SP/AP. The results demonstrate the usefulness of ECochG in diagnosis of endolymphatic hydrops and indicate a possibility of monitoring the state of hydrops after administration of glycerol and endolymphatic sac surgery.

[Results of trapezoid rotation test in guinea pigs with unilateral endolymphatic hydrops]

Obliteration of the right endolymphatic sac was performed by Kimura's method in 57 guinea pigs with normal hearing and vestibular function. However, 43 animals exhibited postoperative cerebellopontile disturbance and labyrinthitis detected by gait test, ABR test and histological examination. The remaining 14 guinea pigs were evaluated periodically by the trapezoid rotation test. Two weeks after the operation, 11 of these 14 guinea pigs exhibited prolonged right beating nystagmus based on the labyrinthine preponderance of impaired side (Lpi). In addition, endolymphatic hydrops was histological detected in the labyrinth on the operation side. A positive relation was observed between the degree of endolymphatic hydrops and the degree of Lpi. Lpi at 4 weeks after the operation was markedly lower than that at 2 weeks (P = 0.05). Therefore, the degree of Lpi was thought to be influenced by the rate of hydrops development. These results corresponded with the clinical findings that the degree of Lpi increases prior to vertiginous episodes in Meniere's patients.

[Experimental endolymphatic hydrops and its related morphological and functional changes in guinea pigs]

The endolymphatic hydrops was induced successfully in 13 guinea pigs by blocking the endolymphatic duct and sac, and the morphological and functional alterations (especially on semicircular canal function) after hydrops were studied by means of ENG, EcochG and serial section techniques. Hydrops in the cochlear duct was evident in the apical and/or near apical turns, especially in earlier stages. Dilation of the saccule was seen in all samples, but was barely noticeable in the utricle and semicircular canal's endolymphatic space. Degeneration in the corti's apparatus, stria vasculi and spiral ganglion occurred late. No evident changes of the sensory components in the saccule, utricle and semicircular canals were seen under the light microscope. The hearing loss in earlier stages of hydrops was evident at lower frequencies: later stages showed effects at all frequencies. The cochlear recruitment and fluctuation phenomena were observed in a few of the operated animals. The value of -SP/CAP in a few animals was significantly after operation. The functional changes of semicircular canals were demonstrated by the semicircular paralysis (CP), directional preponderance (DP), vestibular recruitment (VR) and elevated threshold of nystagmus to angular acceleration stimuli. There was no significant correlation between the degrees of hearing loss and impaired semicircular function. The criteria of ENG elicited by the quantitative angular acceleration stimuli for evaluating the semicircular canal function in guinea pigs is proposed for the first time in our experiment.

Size variations in the lateral intercellular spaces of the endolymphatic sac induced by dietary factors

Since much evidence suggests that the endolymphatic sac is responsible for endolymph resorption, and that the endolymphatic sac lateral intercellular spaces which are lined by the energy-dependent transport complex NA+,K(+)-ATPase are important in this process, we sought to evaluate the effects of dietary salt and a food extract that inhibits the activity of Na+,K(+)-ATPase on lateral intercellular space size. Animals fed this food factor and a high-sodium diet had significantly smaller endolymphatic sac lateral intercellular spaces than those animals fed only a high-sodium diet (analysis of variance with Scheffe's multiple comparison test, P less than 0.001). Animals fed a high-sodium diet had significantly larger endolymphatic sac lateral intercellular spaces than those animals fed a control diet only (analysis of variance with Scheffe's multiple comparison test, P less than 0.001). Results of this study suggest that dietary sodium affects endolymphatic sac fluid dynamics and that other food factors may regulate sodium metabolism, and therefore endolymphatic sac function.

[An experimental study on the inhibitory response of the frog semicircular canal nerve]

The inhibitory response of the posterior semicircular canal (p.s.c.) ampullary nerve was studied in comparison with the excitatory response. The isolated p.s.c. of the bull frog was stimulated by the mechanical endolymphatic flow toward either the ampullofugal (AF, excitatory) or the ampullopetal (AP, inhibitory) direction. Three experiments were performed. 1) Experiment I: The influence of the endolymphatic K+ concentration. In both the high (50mM) and the low (2.5mM) K+ environment, the inhibitory response was smaller than the excitatory response. This was more pronounced in the low K+ environment. 2) Experiment II: The influence of the initial position of the cupula. The magnitudes of the responses when the cupula was moved from various positions were compared. When the same amount of stimulus was given in the same direction, the responses were equal regardless of the initial position of the cupula. The inhibitory response yielded a time constant approximately 2-3 times as long as the excitatory response. 3) Experiment III: The interaction of the AF and the AP stimulus. The influence of the inhibitory response upon the following excitatory response and vice versa, were studied by changing the timing of the stimulus onset. Both the action potential and the DC potential were recorded for comparison. The AP stimulus which immediately followed the AF stimulus suppressed the action potential and the negative DC potential was observed. On the other hand the AF stimulus which immediately followed the AP stimulus gave rise to the distinct positive DC potential but the action potential could hardly be evoked. That AF stimulus resulted in different responses between the action potential and the DC potential indicates that the DC potential recorded in this study is possibly a generator potential.

Calcium gradients in inner ear endolymph

Recent studies suggest that endolymphatic hydrops resulting from the ablation of the endolymphatic duct and sac in guinea pigs may be caused by a disturbance of endolymph calcium homeostasis. A similar disturbance of calcium homeostasis could represent the underlying cause of Ménière's disease. In this study, we mapped the calcium concentrations and electrical potentials throughout the endolymphatic system in normal guinea pigs. Large concentration differences exist between different compartments, including a more than twofold increase along the length of the cochlea. The electrochemical potential for calcium (the force driving passive longitudinal calcium movement) was calculated for all the endolymphatic compartments. The results show that endolymph is extremely inhomogenous with respect to calcium potentials. On the basis of these potentials, it appears that calcium is transported into endolymph in the cochlea and out of endolymph in the saccule and utricle. The possibility that endolymphatic hydrops arises from a disturbance in longitudinal flow of calcium, rather than in longitudinal volume flow, is considered.

Effects of amiloride on the endolymphatic sac

The effect of amiloride on the murine endolymphatic sac was investigated. The amiloride caused collapse of the lateral intercellular spaces in the endolymphatic sac epithelium and a subsequent mild endolymphatic hydrops. These changes indicated a decreased absorption of endolymph in the endolymphatic sac. Amiloride is known to inhibit the transcellular fluid transport without inducing any changes in the paracellular fluid transport. It is therefore suggested that amiloride specially inhibits the fluid and ion exchange in the apical portion of the epithelial cells resulting a decrease in transcellular fluid transport across the endolymphatic sac epithelium. The transcellular fluid transport seems to be one of the main mechanisms in the endolymphatic sac fluid exchange system.

An electron microscopic study of the function of the root cells in the external spiral sulcus of the cochlea

The ultrastructure of the area of the external spiral sulcus of the cochlea was studied in 15 hamsters. Intravital fixation with glutaraldehyde was followed by osmium citrate fixation and embedding in araldite. The endolymphatic surface of the external spiral sulcus is nearly completely covered by Claudius cells under which the root cells are situated. These epithelial cells are composed of bundles (roots), which protrude into the connective tissue of the spiral ligament in a tree-like fashion, forming primary, secondary, and tertiary branches. The basement membrane is a reliable landmark in this complex region, serving as a differentiating zone between the root cells and the mesenchymal tissue. Two different types of root cells are found. There are organelle rich "dark" cells and "light" cells which contain only a few cell organelles. All types of transitional forms between these two cell types are found, which can be interpreted as an expression of different metabolic activity of the same cell type. Between the root cells are seen extracellular clefts into which the "dark" cells, with their high synthetic and secretory activity, expel their secretions. These clefts lead to the endolymphatic space. A continuous production and secretion of part of the endolymph by the root cells is very likely. The anchoring of fibrils at the basement membrane of the root cell bundles, and their continuity with the basilar membrane leads to the assumption that this whole system may act as a mechanical pump. At high sound pressure levels, the strong vibration of the basilar membrane may lead to a locally and temporarily enhanced secretion of substances from the root cells into the cochlear duct.

Alterations in oxygenation of cochlear endolymph during loud sound exposure

The oxygen tension (pO2) of endolymph of the guinea pig cochlea was measured during exposure to loud sound (12 kHz or high-pass noise; 110 dB SPL up to 1 h duration). A small, but significant, steady decline in mean pO2 was observed after both pure tone and high-pass noise exposure. The extent of the change in pO2 varied from 0-50% in individual animals, compared with unexposed control animals. All exposed animals had an extensive loss of compound action potential (CAP) thresholds at frequencies of 8-30 kHz. However, there was no relationship between the extent of the change in pO2 of endolymph and CAP threshold loss.