- Email: [email protected]
Calcium current in type I hair cells isolated from the semicircular canal crista ampullaris of the rat
Brain Research 994 (2003) 175 – 180 www.elsevier.com/locate/brainres
Research report
Calcium current in type I hair cells isolated from the semicircular canal crista ampullaris of the rat Ange´lica Almanza *, Rosario Vega, Enrique Soto Instituto de Fisiologı´a, Universidad Auto´noma de Puebla, Apartado Postal 406 Puebla, Pue, 72000 Mexico Accepted 16 September 2003
Abstract The low voltage gain in type I hair cells implies that neurotransmitter release at their afferent synapse should be mediated by low voltage activated calcium channels, or that some peculiar mechanism should be operating in this synapse. With the patch clamp technique, we studied the characteristics of the Ca2 + current in type I hair cells enzymatically dissociated from rat semicircular canal crista ampullaris. Calcium current in type I hair cells exhibited a slow inactivation (during 2-s depolarizing steps), was sensitive to nimodipine and was blocked by Cd2 + and Ni2 +. This current was activated at potentials above 60 mV, had a mean half maximal activation of 36 mV, and exhibited no steady-state inactivation at holding potentials between 100 and 60 mV. This data led us to conclude that hair cell Ca2 + current is most likely of the L type. Thus, other mechanisms participating in neurotransmitter release such as K+ accumulation in the synaptic cleft, modulation of K+ currents by nitric oxide, participation of a Na+ current and possible metabotropic cascades activated by depolarization should be considered. D 2003 Elsevier B.V. All rights reserved. Theme: Excitable membranes and synaptic transmission Topic: Presynaptic mechanisms Keywords: Calcium channels; Dihydropyridine; Inner ear; Afferent synapse; Calyx synapse; Transmitter release; Vestibular system
1. Introduction Ionic conductance of vestibular hair cells varies depending on the animal species, the region where the cells originate, the developmental stage of the animal, and the hair cell type. However, in types I and II hair cells studied up to date, potassium currents dominate the basolateral membrane response [8,17,18,29]. Type I hair cells have a very peculiar afferent synapse which forms a calyceal structure surrounding the basolateral surface of the cell; many authors have wondered what the possible function of this calyx might be [9,30,31,32]. Due to the difussional properties of the microspace formed between the calyx and the hair cell, it has been suggested that changes in this microambient may be relevant for synaptic transmission in type I hair cells. In contrast, type II hair cells have bouton afferent endings which have been thoroughly * Corresponding author. Tel.: +52-2-229-5500x7310; fax: +52-2-2334511. E-mail address: [email protected] (A. Almanza). 0006-8993/$ - see front matter D 2003 Elsevier B.V. All rights reserved. doi:10.1016/j.brainres.2003.09.033
studied [10]. They correspond to a typical chemical synapse, and transmitter release has been shown to be calcium dependent [13,20,33]. The most significant electrophysiological difference between types I and II vestibular hair cells is the expression of a low voltage-activated, non-inactivating outward K+ current ( gK,L) in type I hair cells [8,26,28]. gK,L is substantially activated at the resting potential, the half activation voltage (V1/2) value of gK,L can be as negative as 90 mV in physiological conditions, greatly reducing cell input resistance [28]. In fact, the low input resistance of type I hair cells determines that mechanoelectrical transducer currents seem not to be large enough to depolarize these cells to produce neurotransmitter release. Therefore, for these cells to activate the neurotransmitter release machinery, voltagedependent Ca2 + channels should have a very negative activating voltage, or there should be some other mechanisms contributing to further depolarize the cell [7,9,32]. In this study, we addressed the problem of determining the characteristics and particularly the half activation voltage of the Ca2 + current in type I hair cells.
176
A. Almanza et al. / Brain Research 994 (2003) 175–180
2. Materials and methods Experiments were performed in young Long-Evans rats (postnatal days 14 – 17) supplied by the bioterium ‘‘Claude Bernard’’ of the University of Puebla. All efforts were made to minimize animal suffering and to reduce the number of animals used, as outlined in the ‘‘Guide to the Care and Use of Laboratory Animals’’ issued by the National Academy of Science, USA. Hair cells were enzymatically dissociated from the semicircular canal crista ampullaris of the rat. Tissue pieces containing the vestibular sensory epithelia were incubated in Tyrode containing IA type colagenase (0.2 mg/ml) for 7 min at 35 jC, followed by cell incubation for 10 min at 35 jC in a Ca2 + and Mg2 +-free Tyrode solution (Table 1) with trypsin (1 mg/ml). Tissue was finally washed for 10 min at 4 jC in a Ca2 + and Mg2 +-free Tyrode solution containing serum bovine albumin (1 mg/ml). All enzymes were from Sigma (St Louis, MO). Isolated cells were placed in the recording chamber (400 Al) on an inverted microscope stage. Cells settled and adhered to the bottom of the recording chamber within 10 min. The recording chamber was continuously perfused with the corresponding Tyrode solution at a constant rate (0.5 ml/min) using a peristaltic pump (Microperpex, LKB, Sweden). Whole-cell currents were recorded at room temperature (22 –25 jC), according to the method described by Hamill et al. [11] and modified for isolated rat vestibular hair cells. In brief, the recordings were obtained with the patch clamp technique in the whole-cell configuration; patch pipettes had resistances of 3– 5 MV when filled with the corresponding intracellular solution (Table 1). Command pulse generation and data sampling ( c every 200 As) were controlled by pClamp 8.0 software (Axon Inst.) using a 12-bit analog to digital converter (Digidata 1200, Axon Inst.). Membrane capacitance and series resistance (80%) were electronically compensated. Voltages were corrected for a 7.9-mV liquid junction potential for the Ba2 + extracellular solution and Cs+ and TEA intracellular medium (solutions 3 and 4 in Table 1). Correction was 6.8 mV for the high Ca2 + extracellular solution and the intracellular Ca2 + solution (solutions 5 and 6 in Table 1). The junction potential was calculated with the generalized Henderson liquid junction potential equation [2] using pClamp 8.0 software (Axon
Fig. 1. Morphology of types I and II hair cells from the rat semicircular canal. In A, type I hair cell presenting a flask-shaped body with a highly refringent cuticular plate wider than the hair cell neck. In B, type II hair cell with a more cylindrical shape. The cuticular plate is hardly distinguishable under phase contrast optics. Calibration bar = 10 Am.
Inst.). Student’s t-test was used to assess statistical significance, with P < 0.05 indicating a statistically significant difference. The results are given as mean F S.E.M. Current evoked between 100 and 60 mV was fitted with a linear function that was subtracted assuming a linear leak throughout the whole current– voltage relation [15]. Solutions containing drugs or blockers were dissolved in the corresponding Tyrode solution and applied by microperfusion. For this, a teflon microtube was drawn close to the cell and the desired solution was applied by pressure ejection controlled by a microsyringe (Baby Bee, BAS) with a flow rate of 20 Al/min. To distinguish between types I and II hair cells, their morphological and electrophysiological properties were correlated. Type I hair cells presented a flask-shaped body with a highly refringent cuticular plate wider than the hair cell neck (Fig. 1A), strongly negative membrane potential, low membrane resistance, and the whole-cell currents showed a large instantaneous outward component due to gK,L expression [28]. In contrast, type II hair cells exhibited more cylindrical shapes and their cuticular plate was hardly distinguishable under phase contrast optics (Fig. 1B). Electrophysiologically, they showed a less negative membrane potential, higher membrane resistance, and the whole-cell
Table 1 Solutions (in mM) KCl (1) (2) (3) (4) (5) (6)
Normal Tyrode 5.4 Normal intracellular 140 Extracellular with Ba2 + – Intracellular Cs+-TEA – High Ca2 + extracellular – – Intracellular Ca2 +
TEACl CsCl NaCl MgCl2 CaCl2
23 23 23 23
– 30 69.3 30 75
140 10 55 – 70 –
1.2 2 – – – –
BaCl2 HEPES NMDG+
3.6 – 0.134 – 0.1 20 0.1 – 10 – 0.1 –
10 10 10 10 10 10
– – – 80 – 80
EGTA GTPNa2 ATPMg Glucose 4-AP – 10 – 10 – 10
– 0.1 – 0.1 – 0.1
– 2 – 2 – 2
10 – 10 – 10 –
– – – – 5 –
A. Almanza et al. / Brain Research 994 (2003) 175–180 Table 2 Passive properties of hair cells from the rat
Type I Type II
Vm (mV)
Cm (pF)
Rm (MV)
sm = RmCm
n
75 F 1* 47 F 3*
6.7 F 1 6.1 F 1
45 F 10 * 592 F 62 *
302 As* 3.6 ms*
7 7
Resting membrane potential, Vm; membrane capacitance, Cm; membrane resistance, Rm. *P < 0.001.
current was dominated by outward rectifying currents without an instantaneous component [8].
3. Results Stable recordings in normal extra and intracellular solutions (Table 1) were readily obtained. Typically, type I hair cells displayed a gK,L current, thus having a small membrane resistance (Rm) value compared with type II
177
hair cells (Table 2). In current clamp with a membrane potential of about 73 F 4 mV, type I hair cells showed a low voltage gain with a near linear slope of 2.4 F 0.2 mV/100 pA (data not shown). A current injection of as much as 600 pA depolarized the cells up to 61 F 6 mV (n = 4). For Ca2 + current recording, extracellular Ba2 + was used as a current carrier and intracellular Cs+-TEA solutions were used to block K+ currents (Table 1). With these solutions, and since no gK,L was present, hair cells were differentiated on the basis of morphological criteria (see Materials and methods). Inward Ba2 + current showed no inactivation during 200-ms voltage pulses (Fig. 2A). Application of CdCl2 (30 and 300 AM; n = 3 each) reduced IBa in 74 F 4% and 90 F 6%, respectively; perfusion with 100 AM NiCl2 (n = 5) also reduced the current 48 F 9%. Both Cd2 + and Ni2 +did not modify the activation kinetics nor the voltage dependence of IBa (data not shown). The application of nimodipine (1, 3 and 10 AM, n = 3, 6 and 4,
Fig. 2. Current voltage relationship for IBa. In A, typical recording of IBa obtained for VH = 60- and 10-mV voltage pulses from 100 to + 50 mV. In B, current to voltage relationship for IBa recorded for holding voltages of 60 (., n = 13) and 100 mV (o, n = 11), and for ICa (z, n = 4) recorded with the same protocol and VH = 60 mV. The current – voltage relationship for ICa is about 50% IBa. Each point represents the mean F standard error. In C and D, typical records of the current recorded using Ca2 + as a current carrier. Slow inactivation can be observed during a 2-s pulse as shown in D (C and D are the average of three traces).
178
A. Almanza et al. / Brain Research 994 (2003) 175–180
Fig. 3. Nimodipine effects upon IBa current. In A, time course of 3 AM nimodipine to block IBa (bar shows application time). Drug effect reaches its maximum within the first 30 s. Washout of the drug has a slower time course, recovering to 98% of the control current after 2 min. In B, current recorded at 20 mV for VH = 60 mV, 3 AM nimodipine reduced IBa in 57% without affecting its time course or its voltage sensitivity. In C, bar graphs showing IBa percent block by 1, 3 and 10 AM nimodipine.
respectively) produced a reversible concentration-dependent reduction of the inward current. Nimodipine action began within the first 5 s after its perfusion reaching its maximum effect after 30 s of perfusion (Fig. 3A). It produced no apparent modification of IBa current kinetics nor modified its voltage dependence (Fig. 3B). Nimodipine (1, 3 and 10 AM) reduced IBa in 38%, 57% and 77%, respectively (Fig. 3C). The activation of IBa was studied at two holding levels (VH), 100 and 60 mV (n = 11 and 13, respectively). Current was produced by 10 mV, 200-ms pulses from 100 to 50 mV. IBa was activated at potentials above 60 mV and reached its peak at 18 mV. Current inversion was above + 40 mV (Fig. 2B). The mean amplitude of the current for VH = 60 mV was 91 F 8 pA (n = 13), whereas for VH = 100 mV, it was 78 F 11 pA (n = 11); differences
were not significant (Student’s t-test, p = 0.64). Conductances obtained at VH = 60 and 100 mV were normalized and approximated by a Boltzmann function:
Fig. 4. Activation curve for the Ca2 + current, using Ba2 + as current carrier. Data were approximated by a Boltzmann function (solid line) with V1/2 = 36 F 0.4 mV. Points represent the mean F standard error (n = 13). Dotted line shows the maximum depolarization reached in our experiments ( 61 F 6 mV, n = 4) in current clamp conditions, parting from a membrane voltage of 73 mV and using as much as 600-pA current injection.
4. Discussion
gion =gmax ¼ 1=f1 þ e½ðVm V1=2 Þs g where gion is the conductance, gmax is the maximal conductance, Vm is the membrane potential, V1/2 is the potential at which the current is half-maximally activated, and s is the exponential slope. The I Ba recorded for VH = 60 mV had a V1/2 = 36 F 0.4 mV and s = 5.7 F 0.3 mV (n = 13); the IBa recorded for VH = 100 mV had a V1/2 = 37 F 0.8 mV and s = 5 F 0.8 mV (n = 11, Fig. 4). There were no significant differences between them (Student’s t-test, p = 0.94). The gmax for IBa was 1.4 F 0.1 nS (n = 13). The inward current was also studied using Ca2 + as a current carrier in order to define whether any Ca2 +-dependent inactivation process was taking place (n = 6). For these experiments, extracellular high Ca2 + and intracellular Ca2 + solutions were used (Table 1). The current activated very rapidly with a s < 0.5 ms for pulses at 17 mV, and showed no inactivation during 200-ms current pulses. Peak current amplitude was 42 F 10 pA, 46% lower than IBa (Fig. 2B). The current having V1/2 = 35 F 0.6 mV and s = 6.1 F 0.5 was not significantly different to IBa. The gmax for ICa was 0.9 F 0.2 nS. No significant ICa inactivation was found with 200-ms pulses from 100 to + 50 in 10-mV steps and for a VH of 60 and 100 mV (Fig. 2C). However, when using 2-s pulses of the same magnitude, a slowly developing current inactivation with a s = 680 F 133 ms at 17 mV was found (Fig. 2D). Current characteristics for a VH of 60 and 100 mV were similar during the 2-s pulses.
Our results are consistent with the idea that a single Ca2 + channel subtype may account for the major part of the macroscopic Ca2 + current. We found a small inward current with a peak amplitude of about 50 pA at 17 mV. Current
A. Almanza et al. / Brain Research 994 (2003) 175–180
amplitude increased by using Ba2 + as a current carrier, was blocked by Cd2 + and Ni2 +, and was sensitive to nimodipine. This current, activated above 60 mV with V1/2 = 36 mV, showed a slowly developing inactivation while using Ca2 + as a current carrier. Variations in the holding potential ( 100 and 60 mV) did not significantly modify the current properties, indicating a homogenous current, most likely of the L type. This is in agreement and confirms recent results obtained for semicircular canal type I hair cells of the rat and chick, in which authors also reported a homogenous single Ca2 + current which activated above 60 mV [1,4]. In our recordings, V1/2 of the Ca2 + current in type I hair cells was 36 mV, that is, it was displaced 5 mV to the right when compared to the one reported by Bao et al. [1] which had a mean half activation voltage of 41.1 F 0.5 mV. Difference in the V1/2 of the Ca2 + current could be due to the patch configuration used in each case. Bao et al. [1] used the perforated patch configuration in which, there is an incomplete solution exchange between pipette and cell, where only the small permeant ions will be exchanged while the larger impermeant anions will not. Therefore, the potential difference will have a Donnan component due to impermeant anions and, previous to a complete equilibration, a diffusion potential dependent on the ion selectivity of the ionophore will be present. The Donnan potential across the membrane causes the pipette interior to differ in voltage from the cell interior by an amount depending upon the extent of ionic redistribution [23]. At the same time, perforated patch will allow the cell interior to retain various active molecules which may modulate Ca2 + channels. Compared to the results previously reported by Bao et al. [1], our records also showed a higher efficacy of nimodipine. Since a dihydropyridine block is enhanced by depolarization, this difference could be due to a more depolarized VH and longer test pulses used in our records. It is also worth to note that we obtained hair cells from animals with a much narrower age range (P14 –17 as compared to P4 –20 in Ref. [1]), thus assuring in our experiments that hair cell phenotype and innervation were completely established [29]. Previous studies have reported that in guinea pig type I hair cells, there is a transient Ca2 + inward current, probably T type [25]. By using immunohistochemical techniques, it has also been reported that chinchilla semicircular canal hair cells expressed a1B, a1C and a1D subunits [14], thus suggesting that various Ca2 + channel subtypes may be functionally assembled in the system. Pharmacological studies have also suggested that L, T and R type Ca2 + channels remain the most feasible to be expressed in hair cells [5]. In type II vestibular hair cells, it has been shown that calcium current related to neurotransmitter release from hair cells is a high voltage activating current, probably L type [1,16,20,21,22]. However, in frog semicircular canals and sacculus, it seems that non-L type Ca2 + channels may
179
also be involved in hair cell response and transmitter release [16,27]. In our experimental conditions, the macroscopic Ca2 + current did not show any component suggesting the presence of an inactivating Ca2 + current (R or T type) in type I hair cells. Furthermore, our results suggest that Ca2 + current activation in types I and II hair cells appears to be similar, as reported in the chick [4]. Our results leave open the question about the role of Ca2 + current in afferent synaptic transmission in hair cells. In the turtle crista, it has been reported that less than 10% of type I hair cells do not have a gK,L activated at VH> 70 mV [6]. Also, gK,L half activation in the turtle crista ( 55 to 60 mV) has been reported to be shifted to the right as compared to previous reports in mammals and birds [8,18,28]. This implies that type I hair cells very probably do not constitute a homogenous population, and more probably that the activation of the dominant gK,L current spans a whole range and is highly sensitive to cyclic GMP modulation [3,7]. In the case of type II hair cells, at least two sub-types have been reported: fast and slow. They have significantly different input resistances and voltage gains, although both are capable of generating receptor potentials of an amplitude enough to activate the Ca2 + current [6,32]. The question is whether in some type I hair cells, the Ca2 + activation threshold could be reached without implying other mechanisms. At least in those type I hair cells in which gK,L shows the more negative half activation value (> 70 mV), some additional mechanism is needed to boost synaptic transmission. Cells displaying no such hyperpolarized gK,L activation may have receptor potentials within the range of Ca2 + current activation. Among the proposed mechanisms participating in type I hair cell afferent synapse, K+ accumulation in the synaptic cleft that may depolarize both hair cell and afferent terminal, increasing transmitter release, and facilitating the afferent neuron spike discharge seems the most feasible [9,32]. This hypothesis has been reinforced by recent results showing that afferent calyceal synapses in type I hair cells also express a low voltage activated K+ current [12,24]. Another mechanism operating in this synapse may involve the modulation of gK,L by cyclic GMP [3,7]. In addition, in the chick, a Na+ current could reinforce, at least in a subpopulation of hair cells, the membrane depolarization which will boost synaptic output [4,19].
Acknowledgements ´ lvarez for The authors wish to thank Marcela Sa´nchez A proof reading the English manuscript. This material is based upon work supported by the Consejo Nacional de Ciencia y Tecnologı´a, Me´xico (CONACyT) grant 40672 to RV and by Secretaria de Educacio´n Superior, DGES-VIEP grant II-84G02. AA was supported by a CONACyT fellowship.
180
A. Almanza et al. / Brain Research 994 (2003) 175–180
References [1] H. Bao, W.H. Wong, J.M. Goldberg, R.A. Eatock, Voltage-gated calcium channel currents in type I and type II hair cells isolated from the rat crista, J. Neurophysiol. 90 (2003) 155 – 164. [2] P.H. Barry, JPCalc, a software package for calculating liquid junction potential corrections in patch-clamp, intracellular, epithelial and bilayer measurements and for correcting junction potential measurements, J. Neurosci. Methods 51 (1994) 107 – 116. [3] O. Behrend, C. Schwark, T. Kunihiro, M. Strupp, Cyclic GMP inhibits and shifts the activation curve of the delayed-rectifier (IK1) of type I mammalian vestibular hair cells, NeuroReport 8 (1997) 2687 – 2690. [4] M. Bosica, G. Zucca, P. Valli, S. Masetto, Na+ and Ca2 + currents in chick embryo developing type I and type II hair cells, 3rd Federation of European Physiological Societies, Nice, France (2003) 19 – 20. [5] C. Boyer, J. Lehouelleur, A. Sans, Potassium depolarization of mammalian vestibular sensory cells increases [Ca2 +] through voltage-sensitive calcium channels, Eur. J. Neurosci. 10 (1998) 971 – 975. [6] A.M. Brichta, A. Aubert, R.A. Eatock, J.M. Goldberg, Regional analysis of whole cell currents from hair cells of the turtle posterior crista, J. Neurophysiol. 88 (2002) 3259 – 3278. [7] J.W.Y. Chen, R.A. Eatock, Major potassium conductance in type I hair cells from rat semicircular canals: characterization and modulation by nitric oxide, J. Neurophysiol. 84 (2000) 139 – 151. [8] M.J. Correia, D.G. Lang, An electrophysiological comparison of solitary type I and type II vestibular hair cells, Neurosci. Lett. 116 (1990) 106 – 111. [9] J.M. Goldberg, Theoretical analysis of intracellular communication between the vestibular type I hair cell and its calyx ending, J. Neurophysiol. 76 (1996) 1942 – 1957. [10] P.S. Guth, P. Perin, C.H. Norris, P. Valli, The vestibular hair cells: post-transductional signal processing, Prog. Neurobiol. 54 (1998) 193 – 247. [11] O.P. Hamill, A. Marty, E. Neher, B. Sakman, F.J. Sigworth, Improved patch clamp technique for high resolution current recording from cell and cell free membrane patches, Pflu¨gers Arch. 391 (1981) 85 – 100. [12] K.M. Hurley, R.A. Eatock, Conductances in the post-synaptic calyx ending on the type I hair cell, Assoc. Res. Otolaryngol. Abstr. 26 (2003) 113. [13] D. Lenzi, W. Roberts, Calcium signalling in hair cells: multiple roles in a compact cell, Curr. Opin. Neurobiol. 4 (1994) 496 – 502. [14] I. Lo´pez, G. Ishiyama, A. Ishiyama, J.C. Jen, F. Liu, R.W. Baloh, Differential subcellular immunolocalization of voltage-gated calcium channel alpha1 subunits in the chinchilla cristae ampullaris, Neurosci. 92 (1999) 773 – 782. [15] C. Martinez-Dunst, R.L. Michaelis, P.A. Fuchs, Release sites and calcium channels in hair cells of the chick’s cochlea, J. Neurosci. 17 (1997) 9133 – 9144. [16] M. Martini, M.L. Rossi, G. Rubbini, G. Rispoli, Calcium currents in hair cells isolated from semicircular canals of the frog, Biophys. J. 78 (2000) 1240 – 1254.
[17] S. Masetto, G. Russo, I. Prigioni, Differential expression of potassium currents by hair cells in thin slices of frog crista ampullaris, J. Neurophysiol. 72 (1994) 443 – 455. [18] S. Masetto, P. Perin, A. Malusa, G. Zucca, P. Valli, Membrane properties of chick semicircular canal hair cells in situ during embryonic development, J. Neurophysiol. 83 (2000) 2740 – 2756. [19] S. Masetto, M. Bosica, M.J. Correia, O.P. Ottersen, G. Zucca, P. Perin, P. Valli, Na+ currents in vestibular type I and type II hair cells of the embryo and adult chicken, J. Neurophysiol. 90 (2003) 1266 – 1278. [20] P. Perin, E. Soto, R. Vega, L. Botta, S. Masetto, G. Zucca, P. Valli, Calcium channels functional roles in the frog semicircular canal, NeuroReport 11 (2000) 417 – 420. [21] P. Perin, S. Masetto, M. Martini, M.L. Rossi, G. Rubbini, G. Rispoli, P.S. Guth, G. Zucca, P. Valli, Regional distribution of calcium currents in frog semicircular canal hair cells, Hear. Res. 152 (2001) 67 – 76. [22] I. Prigioni, S. Masetto, G. Russo, V. Taglietti, Calcium currents in solitary hair cells isolated from frog crista ampullaris, J. Vest. Res. 2 (1992) 31 – 39. [23] J. Rae, K. Cooper, P. Gates, M. Watsky, Low access resistance perforated patch recording using amphotericin B, J. Neurosci. Methods 37 (1991) 15 – 26. [24] K.J. Rennie, Voltage-dependent currents in calyx terminals isolated from gerbil semicircular canals, Assoc. Res. Otolaryngol. Abstr. 26 (2003) 271. [25] K.J. Rennie, J.F. Ashmore, Ionic currents in isolated vestibular hair cells from the guinea-pig crista ampullaris, Hear. Res. 51 (1991) 279 – 292. [26] K.J. Rennie, M.J. Correia, Potassium currents in mammalian and avian isolated type I semicircular canal hair cells, J. Neurophysiol. 71 (1994) 317 – 329. [27] A. Rodriguez-Contreras, E. Yamoah, Direct measurement of singlechannel Ca2 + currents in bullfrog hair cells reveals two distinct channel subtypes, J. Physiol. 534 (2001) 669 – 689. [28] A. Ru¨sch, R.A. Eatock, A delayed conductance in type I hair cells of the mouse utricle, J. Neurophysiol. 76 (1996) 995 – 1004. [29] A. Ru¨sch, A. Lysakowski, R.A. Eatock, Postnatal development of type I and type II hair cells in the mouse utricle: acquisition of voltage-gated conductances and differentiated morphology, J. Neurosci. (1998) 7487 – 7501. [30] E. Scarfone, D. Demeˆmes, R. Jahn, P. De Camilli, A. Sans, Secretory function of the vestibular nerve calyx suggested by presence of vesicles, synapsin I, and synaptophysin, J. Neurosci. 8 (1988) 4640 – 4645. [31] D.A. Schessel, R. Ginzberg, S.M. Highstein, Morphophysiology of synaptic transmission between type I hair cells and vestibular primary afferents. An intracellular study employing horseradish peroxidase in the lizard, Calotes versicolor, Brain Res. 544 (1991) 1 – 16. [32] E. Soto, R. Vega, R. Budelli, The receptor potential in type I and type II vestibular system hair cells: a model analysis, Hear. Res. 165 (2002) 35 – 47. [33] M. Yamashita, H. Ohmori, Synaptic responses to mechanical stimulation in calyceal and bouton type vestibular afferents studied in an isolated preparation of semicircular canal ampullae of chicken, Exp. Brain Res. 80 (1990) 475 – 488.
Research report
Calcium current in type I hair cells isolated from the semicircular canal crista ampullaris of the rat Ange´lica Almanza *, Rosario Vega, Enrique Soto Instituto de Fisiologı´a, Universidad Auto´noma de Puebla, Apartado Postal 406 Puebla, Pue, 72000 Mexico Accepted 16 September 2003
Abstract The low voltage gain in type I hair cells implies that neurotransmitter release at their afferent synapse should be mediated by low voltage activated calcium channels, or that some peculiar mechanism should be operating in this synapse. With the patch clamp technique, we studied the characteristics of the Ca2 + current in type I hair cells enzymatically dissociated from rat semicircular canal crista ampullaris. Calcium current in type I hair cells exhibited a slow inactivation (during 2-s depolarizing steps), was sensitive to nimodipine and was blocked by Cd2 + and Ni2 +. This current was activated at potentials above 60 mV, had a mean half maximal activation of 36 mV, and exhibited no steady-state inactivation at holding potentials between 100 and 60 mV. This data led us to conclude that hair cell Ca2 + current is most likely of the L type. Thus, other mechanisms participating in neurotransmitter release such as K+ accumulation in the synaptic cleft, modulation of K+ currents by nitric oxide, participation of a Na+ current and possible metabotropic cascades activated by depolarization should be considered. D 2003 Elsevier B.V. All rights reserved. Theme: Excitable membranes and synaptic transmission Topic: Presynaptic mechanisms Keywords: Calcium channels; Dihydropyridine; Inner ear; Afferent synapse; Calyx synapse; Transmitter release; Vestibular system
1. Introduction Ionic conductance of vestibular hair cells varies depending on the animal species, the region where the cells originate, the developmental stage of the animal, and the hair cell type. However, in types I and II hair cells studied up to date, potassium currents dominate the basolateral membrane response [8,17,18,29]. Type I hair cells have a very peculiar afferent synapse which forms a calyceal structure surrounding the basolateral surface of the cell; many authors have wondered what the possible function of this calyx might be [9,30,31,32]. Due to the difussional properties of the microspace formed between the calyx and the hair cell, it has been suggested that changes in this microambient may be relevant for synaptic transmission in type I hair cells. In contrast, type II hair cells have bouton afferent endings which have been thoroughly * Corresponding author. Tel.: +52-2-229-5500x7310; fax: +52-2-2334511. E-mail address: [email protected] (A. Almanza). 0006-8993/$ - see front matter D 2003 Elsevier B.V. All rights reserved. doi:10.1016/j.brainres.2003.09.033
studied [10]. They correspond to a typical chemical synapse, and transmitter release has been shown to be calcium dependent [13,20,33]. The most significant electrophysiological difference between types I and II vestibular hair cells is the expression of a low voltage-activated, non-inactivating outward K+ current ( gK,L) in type I hair cells [8,26,28]. gK,L is substantially activated at the resting potential, the half activation voltage (V1/2) value of gK,L can be as negative as 90 mV in physiological conditions, greatly reducing cell input resistance [28]. In fact, the low input resistance of type I hair cells determines that mechanoelectrical transducer currents seem not to be large enough to depolarize these cells to produce neurotransmitter release. Therefore, for these cells to activate the neurotransmitter release machinery, voltagedependent Ca2 + channels should have a very negative activating voltage, or there should be some other mechanisms contributing to further depolarize the cell [7,9,32]. In this study, we addressed the problem of determining the characteristics and particularly the half activation voltage of the Ca2 + current in type I hair cells.
176
A. Almanza et al. / Brain Research 994 (2003) 175–180
2. Materials and methods Experiments were performed in young Long-Evans rats (postnatal days 14 – 17) supplied by the bioterium ‘‘Claude Bernard’’ of the University of Puebla. All efforts were made to minimize animal suffering and to reduce the number of animals used, as outlined in the ‘‘Guide to the Care and Use of Laboratory Animals’’ issued by the National Academy of Science, USA. Hair cells were enzymatically dissociated from the semicircular canal crista ampullaris of the rat. Tissue pieces containing the vestibular sensory epithelia were incubated in Tyrode containing IA type colagenase (0.2 mg/ml) for 7 min at 35 jC, followed by cell incubation for 10 min at 35 jC in a Ca2 + and Mg2 +-free Tyrode solution (Table 1) with trypsin (1 mg/ml). Tissue was finally washed for 10 min at 4 jC in a Ca2 + and Mg2 +-free Tyrode solution containing serum bovine albumin (1 mg/ml). All enzymes were from Sigma (St Louis, MO). Isolated cells were placed in the recording chamber (400 Al) on an inverted microscope stage. Cells settled and adhered to the bottom of the recording chamber within 10 min. The recording chamber was continuously perfused with the corresponding Tyrode solution at a constant rate (0.5 ml/min) using a peristaltic pump (Microperpex, LKB, Sweden). Whole-cell currents were recorded at room temperature (22 –25 jC), according to the method described by Hamill et al. [11] and modified for isolated rat vestibular hair cells. In brief, the recordings were obtained with the patch clamp technique in the whole-cell configuration; patch pipettes had resistances of 3– 5 MV when filled with the corresponding intracellular solution (Table 1). Command pulse generation and data sampling ( c every 200 As) were controlled by pClamp 8.0 software (Axon Inst.) using a 12-bit analog to digital converter (Digidata 1200, Axon Inst.). Membrane capacitance and series resistance (80%) were electronically compensated. Voltages were corrected for a 7.9-mV liquid junction potential for the Ba2 + extracellular solution and Cs+ and TEA intracellular medium (solutions 3 and 4 in Table 1). Correction was 6.8 mV for the high Ca2 + extracellular solution and the intracellular Ca2 + solution (solutions 5 and 6 in Table 1). The junction potential was calculated with the generalized Henderson liquid junction potential equation [2] using pClamp 8.0 software (Axon
Fig. 1. Morphology of types I and II hair cells from the rat semicircular canal. In A, type I hair cell presenting a flask-shaped body with a highly refringent cuticular plate wider than the hair cell neck. In B, type II hair cell with a more cylindrical shape. The cuticular plate is hardly distinguishable under phase contrast optics. Calibration bar = 10 Am.
Inst.). Student’s t-test was used to assess statistical significance, with P < 0.05 indicating a statistically significant difference. The results are given as mean F S.E.M. Current evoked between 100 and 60 mV was fitted with a linear function that was subtracted assuming a linear leak throughout the whole current– voltage relation [15]. Solutions containing drugs or blockers were dissolved in the corresponding Tyrode solution and applied by microperfusion. For this, a teflon microtube was drawn close to the cell and the desired solution was applied by pressure ejection controlled by a microsyringe (Baby Bee, BAS) with a flow rate of 20 Al/min. To distinguish between types I and II hair cells, their morphological and electrophysiological properties were correlated. Type I hair cells presented a flask-shaped body with a highly refringent cuticular plate wider than the hair cell neck (Fig. 1A), strongly negative membrane potential, low membrane resistance, and the whole-cell currents showed a large instantaneous outward component due to gK,L expression [28]. In contrast, type II hair cells exhibited more cylindrical shapes and their cuticular plate was hardly distinguishable under phase contrast optics (Fig. 1B). Electrophysiologically, they showed a less negative membrane potential, higher membrane resistance, and the whole-cell
Table 1 Solutions (in mM) KCl (1) (2) (3) (4) (5) (6)
Normal Tyrode 5.4 Normal intracellular 140 Extracellular with Ba2 + – Intracellular Cs+-TEA – High Ca2 + extracellular – – Intracellular Ca2 +
TEACl CsCl NaCl MgCl2 CaCl2
23 23 23 23
– 30 69.3 30 75
140 10 55 – 70 –
1.2 2 – – – –
BaCl2 HEPES NMDG+
3.6 – 0.134 – 0.1 20 0.1 – 10 – 0.1 –
10 10 10 10 10 10
– – – 80 – 80
EGTA GTPNa2 ATPMg Glucose 4-AP – 10 – 10 – 10
– 0.1 – 0.1 – 0.1
– 2 – 2 – 2
10 – 10 – 10 –
– – – – 5 –
A. Almanza et al. / Brain Research 994 (2003) 175–180 Table 2 Passive properties of hair cells from the rat
Type I Type II
Vm (mV)
Cm (pF)
Rm (MV)
sm = RmCm
n
75 F 1* 47 F 3*
6.7 F 1 6.1 F 1
45 F 10 * 592 F 62 *
302 As* 3.6 ms*
7 7
Resting membrane potential, Vm; membrane capacitance, Cm; membrane resistance, Rm. *P < 0.001.
current was dominated by outward rectifying currents without an instantaneous component [8].
3. Results Stable recordings in normal extra and intracellular solutions (Table 1) were readily obtained. Typically, type I hair cells displayed a gK,L current, thus having a small membrane resistance (Rm) value compared with type II
177
hair cells (Table 2). In current clamp with a membrane potential of about 73 F 4 mV, type I hair cells showed a low voltage gain with a near linear slope of 2.4 F 0.2 mV/100 pA (data not shown). A current injection of as much as 600 pA depolarized the cells up to 61 F 6 mV (n = 4). For Ca2 + current recording, extracellular Ba2 + was used as a current carrier and intracellular Cs+-TEA solutions were used to block K+ currents (Table 1). With these solutions, and since no gK,L was present, hair cells were differentiated on the basis of morphological criteria (see Materials and methods). Inward Ba2 + current showed no inactivation during 200-ms voltage pulses (Fig. 2A). Application of CdCl2 (30 and 300 AM; n = 3 each) reduced IBa in 74 F 4% and 90 F 6%, respectively; perfusion with 100 AM NiCl2 (n = 5) also reduced the current 48 F 9%. Both Cd2 + and Ni2 +did not modify the activation kinetics nor the voltage dependence of IBa (data not shown). The application of nimodipine (1, 3 and 10 AM, n = 3, 6 and 4,
Fig. 2. Current voltage relationship for IBa. In A, typical recording of IBa obtained for VH = 60- and 10-mV voltage pulses from 100 to + 50 mV. In B, current to voltage relationship for IBa recorded for holding voltages of 60 (., n = 13) and 100 mV (o, n = 11), and for ICa (z, n = 4) recorded with the same protocol and VH = 60 mV. The current – voltage relationship for ICa is about 50% IBa. Each point represents the mean F standard error. In C and D, typical records of the current recorded using Ca2 + as a current carrier. Slow inactivation can be observed during a 2-s pulse as shown in D (C and D are the average of three traces).
178
A. Almanza et al. / Brain Research 994 (2003) 175–180
Fig. 3. Nimodipine effects upon IBa current. In A, time course of 3 AM nimodipine to block IBa (bar shows application time). Drug effect reaches its maximum within the first 30 s. Washout of the drug has a slower time course, recovering to 98% of the control current after 2 min. In B, current recorded at 20 mV for VH = 60 mV, 3 AM nimodipine reduced IBa in 57% without affecting its time course or its voltage sensitivity. In C, bar graphs showing IBa percent block by 1, 3 and 10 AM nimodipine.
respectively) produced a reversible concentration-dependent reduction of the inward current. Nimodipine action began within the first 5 s after its perfusion reaching its maximum effect after 30 s of perfusion (Fig. 3A). It produced no apparent modification of IBa current kinetics nor modified its voltage dependence (Fig. 3B). Nimodipine (1, 3 and 10 AM) reduced IBa in 38%, 57% and 77%, respectively (Fig. 3C). The activation of IBa was studied at two holding levels (VH), 100 and 60 mV (n = 11 and 13, respectively). Current was produced by 10 mV, 200-ms pulses from 100 to 50 mV. IBa was activated at potentials above 60 mV and reached its peak at 18 mV. Current inversion was above + 40 mV (Fig. 2B). The mean amplitude of the current for VH = 60 mV was 91 F 8 pA (n = 13), whereas for VH = 100 mV, it was 78 F 11 pA (n = 11); differences
were not significant (Student’s t-test, p = 0.64). Conductances obtained at VH = 60 and 100 mV were normalized and approximated by a Boltzmann function:
Fig. 4. Activation curve for the Ca2 + current, using Ba2 + as current carrier. Data were approximated by a Boltzmann function (solid line) with V1/2 = 36 F 0.4 mV. Points represent the mean F standard error (n = 13). Dotted line shows the maximum depolarization reached in our experiments ( 61 F 6 mV, n = 4) in current clamp conditions, parting from a membrane voltage of 73 mV and using as much as 600-pA current injection.
4. Discussion
gion =gmax ¼ 1=f1 þ e½ðVm V1=2 Þs g where gion is the conductance, gmax is the maximal conductance, Vm is the membrane potential, V1/2 is the potential at which the current is half-maximally activated, and s is the exponential slope. The I Ba recorded for VH = 60 mV had a V1/2 = 36 F 0.4 mV and s = 5.7 F 0.3 mV (n = 13); the IBa recorded for VH = 100 mV had a V1/2 = 37 F 0.8 mV and s = 5 F 0.8 mV (n = 11, Fig. 4). There were no significant differences between them (Student’s t-test, p = 0.94). The gmax for IBa was 1.4 F 0.1 nS (n = 13). The inward current was also studied using Ca2 + as a current carrier in order to define whether any Ca2 +-dependent inactivation process was taking place (n = 6). For these experiments, extracellular high Ca2 + and intracellular Ca2 + solutions were used (Table 1). The current activated very rapidly with a s < 0.5 ms for pulses at 17 mV, and showed no inactivation during 200-ms current pulses. Peak current amplitude was 42 F 10 pA, 46% lower than IBa (Fig. 2B). The current having V1/2 = 35 F 0.6 mV and s = 6.1 F 0.5 was not significantly different to IBa. The gmax for ICa was 0.9 F 0.2 nS. No significant ICa inactivation was found with 200-ms pulses from 100 to + 50 in 10-mV steps and for a VH of 60 and 100 mV (Fig. 2C). However, when using 2-s pulses of the same magnitude, a slowly developing current inactivation with a s = 680 F 133 ms at 17 mV was found (Fig. 2D). Current characteristics for a VH of 60 and 100 mV were similar during the 2-s pulses.
Our results are consistent with the idea that a single Ca2 + channel subtype may account for the major part of the macroscopic Ca2 + current. We found a small inward current with a peak amplitude of about 50 pA at 17 mV. Current
A. Almanza et al. / Brain Research 994 (2003) 175–180
amplitude increased by using Ba2 + as a current carrier, was blocked by Cd2 + and Ni2 +, and was sensitive to nimodipine. This current, activated above 60 mV with V1/2 = 36 mV, showed a slowly developing inactivation while using Ca2 + as a current carrier. Variations in the holding potential ( 100 and 60 mV) did not significantly modify the current properties, indicating a homogenous current, most likely of the L type. This is in agreement and confirms recent results obtained for semicircular canal type I hair cells of the rat and chick, in which authors also reported a homogenous single Ca2 + current which activated above 60 mV [1,4]. In our recordings, V1/2 of the Ca2 + current in type I hair cells was 36 mV, that is, it was displaced 5 mV to the right when compared to the one reported by Bao et al. [1] which had a mean half activation voltage of 41.1 F 0.5 mV. Difference in the V1/2 of the Ca2 + current could be due to the patch configuration used in each case. Bao et al. [1] used the perforated patch configuration in which, there is an incomplete solution exchange between pipette and cell, where only the small permeant ions will be exchanged while the larger impermeant anions will not. Therefore, the potential difference will have a Donnan component due to impermeant anions and, previous to a complete equilibration, a diffusion potential dependent on the ion selectivity of the ionophore will be present. The Donnan potential across the membrane causes the pipette interior to differ in voltage from the cell interior by an amount depending upon the extent of ionic redistribution [23]. At the same time, perforated patch will allow the cell interior to retain various active molecules which may modulate Ca2 + channels. Compared to the results previously reported by Bao et al. [1], our records also showed a higher efficacy of nimodipine. Since a dihydropyridine block is enhanced by depolarization, this difference could be due to a more depolarized VH and longer test pulses used in our records. It is also worth to note that we obtained hair cells from animals with a much narrower age range (P14 –17 as compared to P4 –20 in Ref. [1]), thus assuring in our experiments that hair cell phenotype and innervation were completely established [29]. Previous studies have reported that in guinea pig type I hair cells, there is a transient Ca2 + inward current, probably T type [25]. By using immunohistochemical techniques, it has also been reported that chinchilla semicircular canal hair cells expressed a1B, a1C and a1D subunits [14], thus suggesting that various Ca2 + channel subtypes may be functionally assembled in the system. Pharmacological studies have also suggested that L, T and R type Ca2 + channels remain the most feasible to be expressed in hair cells [5]. In type II vestibular hair cells, it has been shown that calcium current related to neurotransmitter release from hair cells is a high voltage activating current, probably L type [1,16,20,21,22]. However, in frog semicircular canals and sacculus, it seems that non-L type Ca2 + channels may
179
also be involved in hair cell response and transmitter release [16,27]. In our experimental conditions, the macroscopic Ca2 + current did not show any component suggesting the presence of an inactivating Ca2 + current (R or T type) in type I hair cells. Furthermore, our results suggest that Ca2 + current activation in types I and II hair cells appears to be similar, as reported in the chick [4]. Our results leave open the question about the role of Ca2 + current in afferent synaptic transmission in hair cells. In the turtle crista, it has been reported that less than 10% of type I hair cells do not have a gK,L activated at VH> 70 mV [6]. Also, gK,L half activation in the turtle crista ( 55 to 60 mV) has been reported to be shifted to the right as compared to previous reports in mammals and birds [8,18,28]. This implies that type I hair cells very probably do not constitute a homogenous population, and more probably that the activation of the dominant gK,L current spans a whole range and is highly sensitive to cyclic GMP modulation [3,7]. In the case of type II hair cells, at least two sub-types have been reported: fast and slow. They have significantly different input resistances and voltage gains, although both are capable of generating receptor potentials of an amplitude enough to activate the Ca2 + current [6,32]. The question is whether in some type I hair cells, the Ca2 + activation threshold could be reached without implying other mechanisms. At least in those type I hair cells in which gK,L shows the more negative half activation value (> 70 mV), some additional mechanism is needed to boost synaptic transmission. Cells displaying no such hyperpolarized gK,L activation may have receptor potentials within the range of Ca2 + current activation. Among the proposed mechanisms participating in type I hair cell afferent synapse, K+ accumulation in the synaptic cleft that may depolarize both hair cell and afferent terminal, increasing transmitter release, and facilitating the afferent neuron spike discharge seems the most feasible [9,32]. This hypothesis has been reinforced by recent results showing that afferent calyceal synapses in type I hair cells also express a low voltage activated K+ current [12,24]. Another mechanism operating in this synapse may involve the modulation of gK,L by cyclic GMP [3,7]. In addition, in the chick, a Na+ current could reinforce, at least in a subpopulation of hair cells, the membrane depolarization which will boost synaptic output [4,19].
Acknowledgements ´ lvarez for The authors wish to thank Marcela Sa´nchez A proof reading the English manuscript. This material is based upon work supported by the Consejo Nacional de Ciencia y Tecnologı´a, Me´xico (CONACyT) grant 40672 to RV and by Secretaria de Educacio´n Superior, DGES-VIEP grant II-84G02. AA was supported by a CONACyT fellowship.
180
A. Almanza et al. / Brain Research 994 (2003) 175–180
References [1] H. Bao, W.H. Wong, J.M. Goldberg, R.A. Eatock, Voltage-gated calcium channel currents in type I and type II hair cells isolated from the rat crista, J. Neurophysiol. 90 (2003) 155 – 164. [2] P.H. Barry, JPCalc, a software package for calculating liquid junction potential corrections in patch-clamp, intracellular, epithelial and bilayer measurements and for correcting junction potential measurements, J. Neurosci. Methods 51 (1994) 107 – 116. [3] O. Behrend, C. Schwark, T. Kunihiro, M. Strupp, Cyclic GMP inhibits and shifts the activation curve of the delayed-rectifier (IK1) of type I mammalian vestibular hair cells, NeuroReport 8 (1997) 2687 – 2690. [4] M. Bosica, G. Zucca, P. Valli, S. Masetto, Na+ and Ca2 + currents in chick embryo developing type I and type II hair cells, 3rd Federation of European Physiological Societies, Nice, France (2003) 19 – 20. [5] C. Boyer, J. Lehouelleur, A. Sans, Potassium depolarization of mammalian vestibular sensory cells increases [Ca2 +] through voltage-sensitive calcium channels, Eur. J. Neurosci. 10 (1998) 971 – 975. [6] A.M. Brichta, A. Aubert, R.A. Eatock, J.M. Goldberg, Regional analysis of whole cell currents from hair cells of the turtle posterior crista, J. Neurophysiol. 88 (2002) 3259 – 3278. [7] J.W.Y. Chen, R.A. Eatock, Major potassium conductance in type I hair cells from rat semicircular canals: characterization and modulation by nitric oxide, J. Neurophysiol. 84 (2000) 139 – 151. [8] M.J. Correia, D.G. Lang, An electrophysiological comparison of solitary type I and type II vestibular hair cells, Neurosci. Lett. 116 (1990) 106 – 111. [9] J.M. Goldberg, Theoretical analysis of intracellular communication between the vestibular type I hair cell and its calyx ending, J. Neurophysiol. 76 (1996) 1942 – 1957. [10] P.S. Guth, P. Perin, C.H. Norris, P. Valli, The vestibular hair cells: post-transductional signal processing, Prog. Neurobiol. 54 (1998) 193 – 247. [11] O.P. Hamill, A. Marty, E. Neher, B. Sakman, F.J. Sigworth, Improved patch clamp technique for high resolution current recording from cell and cell free membrane patches, Pflu¨gers Arch. 391 (1981) 85 – 100. [12] K.M. Hurley, R.A. Eatock, Conductances in the post-synaptic calyx ending on the type I hair cell, Assoc. Res. Otolaryngol. Abstr. 26 (2003) 113. [13] D. Lenzi, W. Roberts, Calcium signalling in hair cells: multiple roles in a compact cell, Curr. Opin. Neurobiol. 4 (1994) 496 – 502. [14] I. Lo´pez, G. Ishiyama, A. Ishiyama, J.C. Jen, F. Liu, R.W. Baloh, Differential subcellular immunolocalization of voltage-gated calcium channel alpha1 subunits in the chinchilla cristae ampullaris, Neurosci. 92 (1999) 773 – 782. [15] C. Martinez-Dunst, R.L. Michaelis, P.A. Fuchs, Release sites and calcium channels in hair cells of the chick’s cochlea, J. Neurosci. 17 (1997) 9133 – 9144. [16] M. Martini, M.L. Rossi, G. Rubbini, G. Rispoli, Calcium currents in hair cells isolated from semicircular canals of the frog, Biophys. J. 78 (2000) 1240 – 1254.
[17] S. Masetto, G. Russo, I. Prigioni, Differential expression of potassium currents by hair cells in thin slices of frog crista ampullaris, J. Neurophysiol. 72 (1994) 443 – 455. [18] S. Masetto, P. Perin, A. Malusa, G. Zucca, P. Valli, Membrane properties of chick semicircular canal hair cells in situ during embryonic development, J. Neurophysiol. 83 (2000) 2740 – 2756. [19] S. Masetto, M. Bosica, M.J. Correia, O.P. Ottersen, G. Zucca, P. Perin, P. Valli, Na+ currents in vestibular type I and type II hair cells of the embryo and adult chicken, J. Neurophysiol. 90 (2003) 1266 – 1278. [20] P. Perin, E. Soto, R. Vega, L. Botta, S. Masetto, G. Zucca, P. Valli, Calcium channels functional roles in the frog semicircular canal, NeuroReport 11 (2000) 417 – 420. [21] P. Perin, S. Masetto, M. Martini, M.L. Rossi, G. Rubbini, G. Rispoli, P.S. Guth, G. Zucca, P. Valli, Regional distribution of calcium currents in frog semicircular canal hair cells, Hear. Res. 152 (2001) 67 – 76. [22] I. Prigioni, S. Masetto, G. Russo, V. Taglietti, Calcium currents in solitary hair cells isolated from frog crista ampullaris, J. Vest. Res. 2 (1992) 31 – 39. [23] J. Rae, K. Cooper, P. Gates, M. Watsky, Low access resistance perforated patch recording using amphotericin B, J. Neurosci. Methods 37 (1991) 15 – 26. [24] K.J. Rennie, Voltage-dependent currents in calyx terminals isolated from gerbil semicircular canals, Assoc. Res. Otolaryngol. Abstr. 26 (2003) 271. [25] K.J. Rennie, J.F. Ashmore, Ionic currents in isolated vestibular hair cells from the guinea-pig crista ampullaris, Hear. Res. 51 (1991) 279 – 292. [26] K.J. Rennie, M.J. Correia, Potassium currents in mammalian and avian isolated type I semicircular canal hair cells, J. Neurophysiol. 71 (1994) 317 – 329. [27] A. Rodriguez-Contreras, E. Yamoah, Direct measurement of singlechannel Ca2 + currents in bullfrog hair cells reveals two distinct channel subtypes, J. Physiol. 534 (2001) 669 – 689. [28] A. Ru¨sch, R.A. Eatock, A delayed conductance in type I hair cells of the mouse utricle, J. Neurophysiol. 76 (1996) 995 – 1004. [29] A. Ru¨sch, A. Lysakowski, R.A. Eatock, Postnatal development of type I and type II hair cells in the mouse utricle: acquisition of voltage-gated conductances and differentiated morphology, J. Neurosci. (1998) 7487 – 7501. [30] E. Scarfone, D. Demeˆmes, R. Jahn, P. De Camilli, A. Sans, Secretory function of the vestibular nerve calyx suggested by presence of vesicles, synapsin I, and synaptophysin, J. Neurosci. 8 (1988) 4640 – 4645. [31] D.A. Schessel, R. Ginzberg, S.M. Highstein, Morphophysiology of synaptic transmission between type I hair cells and vestibular primary afferents. An intracellular study employing horseradish peroxidase in the lizard, Calotes versicolor, Brain Res. 544 (1991) 1 – 16. [32] E. Soto, R. Vega, R. Budelli, The receptor potential in type I and type II vestibular system hair cells: a model analysis, Hear. Res. 165 (2002) 35 – 47. [33] M. Yamashita, H. Ohmori, Synaptic responses to mechanical stimulation in calyceal and bouton type vestibular afferents studied in an isolated preparation of semicircular canal ampullae of chicken, Exp. Brain Res. 80 (1990) 475 – 488.