Light- and electron microscopy of isolated vestibular hair cells from the guinea pig

Existence of nicotinic receptors in a subset of type I vestibular hair cells of guinea pigs

In mammals, vestibular hair cells (VHCs) are classified as type I and II according to morphological criteria. Acetylcholine (ACh) is identified as the primary efferent neurotransmitter. To date, cholinergic activities have been reported in mammalian type II VHCs, but similar activities in type I VHCs have not been pursued presumably because the body of type I VHCs were suggested to be totally surrounded by afferent nerve calyces. A few reports showed that part of type I VHCs were incompletely surrounded by calyces and received contact from the efferent nerve endings in the mammals studied. The possibility of the expression of cholinergic receptors, their subunit composition, and their function in mammals' type I VHCs are still unclear. In this study, nicotinic responses were investigated by the whole-cell patch clamp technique in isolated type I VHCs of guinea pigs. Of the cells, 7.3% were sensitive to cholinergic agonists and showed an excitatory current at -40 mV which was not sensitive to nifedipine, iberiotoxin (IBTX), and apamin. The main carriers of this current were Na⁺ and K⁺. The rank order of activation potency was nicotine > 1,1-dimethyl-4-phenyl-piperazinium (DMPP) > ACh. These nicotinic ACh receptors (nAChRs) were not blocked by strychnine and α-bungarotoxin (α-BTX), but sensitive to d-tubocurarine (dTC) and mecamylamine (Mec). The findings provide physiological evidence that some subtypes of nAChRs may be located in a subset of type I VHCs, which were different from α9α10 nAChRs.

Gentamicin blocks the ACh-induced BK current in guinea pig type II vestibular hair cells by competing with Ca²⁺ at the L-type calcium channel

Type II vestibular hair cells (VHCs II) contain big-conductance Ca²⁺-dependent K⁺ channels (BK) and l-type calcium channels. Our previous studies in guinea pig VHCs II indicated that acetylcholine (ACh) evoked the BK current by triggering the influx of Ca²⁺ ions through l-type Ca²⁺ channels, which was mediated by M2 muscarinic ACh receptor (mAChRs). Aminoglycoside antibiotics, such as gentamicin (GM), are known to have vestibulotoxicity, including damaging effects on the efferent nerve endings on VHCs II. This study used the whole-cell patch clamp technique to determine whether GM affects the vestibular efferent system at postsynaptic M2-mAChRs or the membrane ion channels. We found that GM could block the ACh-induced BK current and that inhibition was reversible, voltage-independent, and dose-dependent with an IC₅₀ value of 36.3 ± 7.8 μM. Increasing the ACh concentration had little influence on GM blocking effect, but increasing the extracellular Ca²⁺ concentration ([Ca²⁺]_o) could antagonize it. Moreover, 50 μM GM potently blocked Ca²⁺ currents activated by (−)-Bay-K8644, but did not block BK currents induced by NS1619. These observations indicate that GM most likely blocks the M2 mAChR-mediated response by competing with Ca²⁺ at the l-type calcium channel. These results provide insights into the vestibulotoxicity of aminoglycoside antibiotics on mammalian VHCs II.

Two distinct channels mediated by m2mAChR and α9nAChR co-exist in type II vestibular hair cells of guinea pig

Acetylcholine (ACh) is the principal vestibular efferent neurotransmitter among mammalians. Pharmacologic studies prove that ACh activates a small conductance Ca2+-activated K+ channels (KCa) current (SK2), mediated by α9-containing nicotinic ACh receptor (α9nAChR) in mammalian type II vestibular hair cells (VHCs II). However, our studies demonstrate that the m2 muscarinic ACh receptor (m2mAChR) mediates a big conductance KCa current (BK) in VHCs II. To better elucidate the correlation between these two distinct channels in VHCs II of guinea pig, this study was designed to verify whether these two channels and their corresponding AChR subtypes co-exist in the same VHCs II by whole-cell patch clamp recordings. We found that m2mAChR sensitive BK currents were activated in VHCs II isolated by collagenase IA, while α9nAChR sensitive SK2 currents were activated in VHCs II isolated by trypsin. Interestingly, after exposing the patched cells isolated by trypsin to collagenase IA for 3 min, the α9nAChR sensitive SK2 current was abolished, while m2mAChR-sensitive BK current was activated. Therefore, our findings provide evidence that the two distinct channels and their corresponding AChR subtypes may co-exist in the same VHCs II, and the alternative presence of these two ACh receptors-sensitive currents depended on isolating preparation with different enzymes.

M2 muscarinic ACh receptors sensitive BK channels mediate cholinergic inhibition of type II vestibular hair cells

There are two types of hair cells in the sensory epithelium of vestibular end organ. Type II vestibular hair cell (VHC II) is innervated by the efferent nerve endings, which employ a cholinergic inhibition mediated by SK channels through the activation of α9-containing nAChR. Our previous studies demonstrated that a BK-type cholinergic inhibition was present in guinea pig VHCs II, which may be mediated by an unknown mAChR. In this study, BK channel activities triggered by ACh were studied to determine the mAChR subtype and function. We found the BK channel was insensitive to α9-containing nAChR antagonists and m1, m3, m4 muscarinic antagonists, but potently inhibited by the m2 muscarinic antagonist. Muscarinic agonists could mimic the effect of ACh and be blocked by m2 antagonist. cAMP analog activated the BK current and adenyl cyclase (AC) inhibitor inhibited the ACh response. Inhibitor of Giα subunit failed to affect the BK current, but inhibitor of Giα and Giβγ subunits showed a potent inhibition to these currents. Our findings provide the physiological evidence that mAChRs may locate in guinea pig VHCs II, and m2 mAChRs may play a dominant role in BK-type cholinergic inhibition. The activation of m2 mAChRs may stimulate Giβγ-mediated excitation of AC/cAMP activities and lead to the phosphorylation of Ca(2+) channels, resulting in the influx of Ca(2+) and opening of the BK channel.

Developmental expression of Kcnq4 in vestibular neurons and neurosensory epithelia

Sensory signal transduction of the inner ear afferent neurons and hair cells (HCs) requires numerous ionic conductances. The KCNQ4 voltage-gated M-type potassium channel is thought to set the resting membrane potential in cochlear HCs. Here we describe the spatiotemporal expression patterns of Kcnq4 and the associated alternative splice forms in the HCs of vestibular labyrinth. Whole mount immunodetection, qualitative and quantitative RT-PCR were performed to characterize the expression patterns of Kcnq4 transcripts and proteins. A topographical expression and upregulation of Kcnq4 during development was observed and indicated that Kcnq4 is not restricted to either a specific vestibular structure or cell type, but is present in afferent calyxes, vestibular ganglion neurons, and both type I and type II HCs. Of the four alternative splice variants, Kcnq4_v1 transcripts were the predominant form in the HCs, while Kcnq4_v3 was the major variant in the vestibular neurons. Differential quantitative expression of Kcnq4_v1 and Kcnq4_v3 were respectively detected in the striolar and extra-striolar regions of the utricle and saccule. Analysis of gerbils and rats yielded results similar to those obtained in mice, suggesting that the spatiotemporal expression pattern of Kcnq4 in the vestibular system is conserved among rodents. Analyses of vestibular HCs of Bdnf conditional mutant mice, which are devoid of any innervation, demonstrate that regulation of Kcnq4 expression in vestibular HCs is independent of innervation.

Potassium depolarization of mammalian vestibular sensory cells increases [Ca2+]i through voltage-sensitive calcium channels

The existence of voltage-sensitive Ca2+ channels in type I vestibular hair cells of mammals has not been conclusively proven. Furthermore, Ca2+ channels present in type II vestibular hair cells of mammals have not been pharmacologically identified. Fura-2 fluorescence was used to estimate, in both cell types, intracellular Ca2+ concentration ([Ca2+]i) variations induced by K+ depolarization and modified by specific Ca2+ channel agonists and antagonists. At rest, [Ca2+]i was 90 +/- 20 nM in both cell types. Microperifusion of high-K+ solution (50 mM) for 1 s increased [Ca2+]i to 290 +/- 50 nM in type I (n = 20) and to 440 +/- 50 nM in type II cells (n = 10). In Ca2+-free medium, K+ did not alter [Ca2+]i. The specific L-type Ca2+ channel agonist, Bay K, and antagonist, nitrendipine, modified in a dose-dependent manner the K+-induced [Ca2+]i increase in both cell types with maximum effect at 2 microM and 400 nM, respectively. Ni2+, a T-type Ca2+ channel blocker, reduced K+-evoked Ca2+ responses in a dose-dependent manner. For elevated Ni2+ concentrations, the response was differently affected by Ni2+ alone, or combined to nitrendipine (500 nM). In optimal conditions, nitrendipine and Ni2+ strongly depressed by 95% the [Ca2+]i increases. By contrast, neither omega-agatoxin IVA (1 microM), a specific P- and Q-type blocker, nor omega-conotoxin GVIA (1 microM), a specific N-type blocker, affected K+-evoked Ca2+i responses. These results provide the first direct evidence that L- and probably T-type channels control the K+-induced Ca2+ influx in both types of sensory cells.

The cellular localization of the neuropeptides substance P, neurokinin A, calcitonin gene-related peptide and neuropeptide Y in guinea-pig vestibular sensory organs: a high-resolution confocal microscopy study

Four neuropeptides, substance P, neurokinin A, calcitonin gene-related peptide and neuropeptide Y, were detected by radioimmunoassay in guinea-pig vestibular end-organs. High-resolution confocal microscopy visualization of immunofluorescence staining was used to determine the cellular localization of these peptides. Substance P- and neurokinin A-like immunoreactivities were found to co-exist in afferent fibers innervating the peripheral regions of both the utricular and ampullar sensory organs. The immunoreactivity was more concentrated in the distal ends of the calyceal-shaped nerve endings that innervate type I sensory cells. While in the guinea-pig, nerve calyces and type I cells are distributed in both the central and peripheral regions of the sensory epithelia, immunoreactive calyces were found only in the peripheral regions. Calcitonin gene-related peptide-like immunoreactivity was localized in small bouton endings situated at the level of the base of the hair cells. These boutons were in a position to make axosomatic contacts with type II sensory cells and axodendritic contacts with afferent nerve endings. Calcitonin gene-related peptide immunoreactivity co-existed with choline acetyltransferase immunoreactivity. The localization and shape of these boutons identified them as the axonal endings of efferent vestibular fibers. Neuropeptide Y-like immunoreactivity was not observed in the actual sensory epithelium but in the underlying connective tissue, where it was located in varicose fibers along blood vessels. The synaptic position of the tachykinins is clearly distinct from that of calcitonin gene-related peptide. This segregation distinguishes the vestibular end-organs from most peripheral tissues where these peptides are co-localized. The tachykinin-immunoreactive afferent fibers are postsynaptic to the hair cells. If, as in somatic sensory endings, these fibers can be triggered to release the neuropeptides by an axon reflex type of activation, then the tachykinins could interfere directly with the function of type I and type II vestibular hair cells. Calcitonin gene-related peptide co-exists with acetylcholine in the efferent axonal endings that are presynaptic to type II hair cells and to afferent fibers. Calcitonin gene-related peptide can thus interfere by direct synaptic action with type II hair cells only. It may also regulate the activity of the tachykinin-containing afferents.

Novel afferent terminal structure in the crista ampullaris of the goldfish, carassius auratus

Using transmission electron microscopy, we have identified a new type of afferent terminal structure in the crista ampullaris of the goldfish Carassius auratus. In addition to the bouton-type afferent terminals previously described in the ear of this species, the crista also contained enlarged afferent terminals that enveloped a portion of the basolateral hair cell membrane. The hair cell membrane was evaginated and protruded into the afferent terminal in a glove-and-finger configuration. The membranes of the two cells were regularly aligned in the protruded region of the contact and had a distinct symmetrical electron density. The electron-dense profiles of these contacts were easily identified and were present in every crista sampled. In some cases, efferent terminals synapsed onto the afferents at a point where the hair cell protruded into the terminal. The ultrastructural similarities of the goldfish crista afferents to calyx afferents found in amniotes (birds, reptiles, and mammals) are discussed. The results of the study support the hypothesis that structural variation in the vertebrate inner ear may have evolved much earlier in evolution than previously supposed.

Calcium homeostasis in guinea pig type-I vestibular hair cell: possible involvement of an Na(+)-Ca2+ exchanger

In type-I vestibular hair cells (VHCs), the mechanisms involved in intracellular calcium homeostasis have not yet been established. In order to investigate the involvement of an Na(+)-dependent ionic exchanger in the regulation of cytosolic free calcium concentration, we analyzed the effect of the removal of external sodium on the cytosolic concentration of calcium ions ([Ca2+]i), sodium ions ([Na+]i), and protons (pHi). These concentrations were measured in type-I VHCs isolated from guinea pig labyrinth, using Fura-2, sodium benzofuran isophtalate (SBFI), and 1,4 diacetoxy-2,3 dicyanobenzol (ADB) respectively. Complete replacement of Na+ in the superfusion solution with N-methyl-D-glucamine (NMDG+), reversibly increased [Ca2+]i by 276 +/- 89% (n = 46) and decreased [Na+]i by 23 +/- 6% (n = 14). Both responses were prevented by removing external Ca2+ or chelating internal Ca2+. This suggests the presence of coupled Ca2+ and Na+ transport. The [Ca2+]i increase evoked by Na(+)-free solution was reduced by about 55% with the application of amiloride derivatives and was totally abolished in the presence of high [Mg2+]o. No pHi variation was detected during [Na+]o reduction. In the absence of external K+, the Na(+)-free solution failed to induce [Ca2+]i increase; the readmission of external K+ restored the [Ca2+]i response. These results are consistent with a Na(+)-Ca2+ exchanger operating in reverse mode. An K+ dependence of this exchange is also suggested.

Uptake of amikacin by hair cells of the guinea pig cochlea and vestibule and ototoxicity: comparison with gentamicin

The distribution of amikacin (AK), an exclusive cochleo-toxic aminoglycosidic antibiotic (AA), and of gentamicin (GM), which is both cochleo- and vestibulo-toxic, has been studied in cochlear and vestibular hair cells. Guinea pigs were treated during six days with one daily injection of AK (450 mg/kg/day) or GM (60 mg/kg/day). AAs were detected, using immunocytochemical technique with scanning laser confocal microscopy, in isolated cells from guinea pigs sacrificed from 2 to 30 days after the end of the treatments. Results demonstrate a rapid uptake (as soon as after 2-day treatment) of both AAs by cochlear and vestibular hair cells and a very slow clearance. Particularly GM and AK are detected in type I and type II hair cells of the utricles and cristae ampullaris. The presence of these two molecules with different toxic potentialities towards cochlear and vestibular hair cells indicates that the selective ototoxicity of aminoglycosides cannot be explained simply on the basis of particular uptake and accumulation in the different sensory hair cells.

Neurofilament proteins form an annular superstructure in guinea-pig type I vestibular hair cells

Neurofilaments, the neuron-specific intermediate filaments, are composed of three immunochemically distinct subunits: NF-L, NF-M and NF-H that can be either phosphorylated or unphosphorylated. In mammals, the distribution of these subunits has been described in vestibular ganglion neurons, but there are no reports on the presence of neurofilaments in vestibular hair cells. We investigated, by immunocytochemistry, neurofilaments in vestibular hair cells from rat and guinea-pig using antibodies against the three subunits and to dephosphorylated NF-H (clone SMI 32, recognizes also NF-M on immunoblots), on Vibratome sections of the vestibular end-organs and on isolated hair cells. Various immunostaining protocols were used, as appropriate for the method of observation: laser scanning confocal microscopy (immunofluorescence) and transmission electron microscopy (immunoperoxidase, pre-embedding technique). In rat and guinea-pig cristae and utricles, neurofilament immunoreactivity was observed in axons inside and below the sensory epithelia. In guinea-pig, in addition to this staining, intensely immunoreactive annular structures were found in the basal regions of hair cells. These rings were detected with anti-NF-L, -NF-M and -dephosphorylated NF-H/M antibodies, but not with anti-phosphorylation-independent NF-H. Ring-containing hair cells were present in all regions of the sensory epithelia but were more abundant in the peripheral areas. All levels of observation (Vibratome and thin sections, and isolated hair cells) showed that only the guinea-pig type I hair cells contained a neurofilament ring. High-resolution observations showed that the ring was located below the nucleus, often close to smooth endoplasmic reticulum and the cell membrane.

Potassium currents in type II vestibular hair cells isolated from the guinea-pig's crista ampullaris

Type II vestibular hair cells were isolated from cristae ampullares of guinea-pig and maintained in vitro for 2-3 h. Outward membrane currents were studied under whole-cell voltage-clamp conditions. Type II hair cells had resting potentials of about -45 mV. Depolarizing voltage steps from a holding potential of -80 or -90 mV induced time- and voltage-dependent outward currents which slowly decayed to a sustained level. Tail currents reversed at about -70 mV, indicating that the outward currents were mainly carried by potassium ions. The currents had an activation threshold around -50 mV. The transient component was completely removed by a depolarizing pre-pulse positive to -10 mV. While bath application of 4-aminopyridine (5 mM) reduced both components, extracellular tetraethylammonium (10 mM) or zero calcium preferentially diminished the sustained current. We conclude that at least two potassium conductances are present, a delayed rectifier with a relatively fast inactivation and a calcium-dependent potassium current. Depolarizing current injections induced an electrical resonance in the voltage responses, with a frequency of 25-100 Hz, larger currents causing higher frequencies.

Glycoconjugates in the vestibular organs as revealed by the silver methenamine method

The glycoconjugates in the vestibular organs of the guinea pig were studied after staining by the silver methenamine method and by the periodic acid-Schiff (PAS) reaction. The organic matrix of otoconia, otolithic membranes and cupulae were stained to the same degree by the PAS reaction. In contrast, the mineralizing and non-mineralizing matrices were clearly distinguished by the silver methenamine method. The otoconia were surrounded by an intensely stained organic matrix, while the otolithic membranes and cupulae were moderately stained. This histochemical difference suggests that the positively stained organic matrix of otoconia is not identical to the otolithic membranes and cupulae in terms of its biochemical composition. The strongly stained material may play an important role in turnover of calcium in otoconia. The contact areas between type I hair cell and nerve calyx were contained silver methenamine-positive material which is probably involved in adhesion of these cell membranes.

Isolated brush cells of the rat stomach retain their structural polarity

The brush cells (BC) are highly polarized elements occurring in epithelia of endodermal origin. They have a preferential topographical distribution in the organs in which they reside. In the stomach of the rat, BC prevail near the transitional zone separating the forestomach from the glandular stomach. Thus, a method was developed to isolate and recover BC from this organ with the aim of investigating the changes they may undergo after dissociation. Strips of the rat stomach were severed from the very proximal border of the glandular region and incubated in Hanks' balanced salt solution containing pronase. After sedimentation of the dissociated cells (crude sediment containing all stomach epithelial cell types) two successive cell fractions were prepared on performed Percoll gradient in an attempt to enrich BC in a defined layer. BC were recovered in a fraction at a density close to 1.03 g/ml where they represented about 2% of all cells. The isolated BC changed their form from columnar to pear-shaped; however, they maintained their structural polarity over 2 h as demonstrated by light microscopy, transmission-and scanning-electron microscopy. The fine structure of BC was always satisfactorily preserved. Maintenance of the structural polarity of isolated BC is contrary to the general rule according to which all conventional epithelial cells examined to date lose their polarity after isolation. This result is discussed in relation to morphological findings in isolated sensory cells (hair cells, photoreceptor cells) leading to the suggestion that BC are more similar to these than to conventional epithelial cells.

Anaesthetics may change the shape of isolated type I hair cells

Type I hair cells isolated from animals anaesthetised with barbiturates or ether were found to be shorter and to lack a prominent 'neck' region when compared to cells isolated from non-anaesthetised animals. Ketamine did not have this effect. The changes observed could have important implications for the physiology of inner ear receptors. These findings infer that care should be taken in the choice of anaesthetics used in studies on cells from the inner ear.