J Neurophysiol 96: 602– 612, 2006.
First published May 10, 2006; doi:10.1152/jn.01315.2005.
Single-Channel L-Type Ca2⫹ Currents in Chicken Embryo Semicircular Canal
Type I and Type II Hair Cells
Valeria Zampini, Paolo Valli, Giampiero Zucca, and Sergio Masetto
Dipartimento di Scienze Fisiologiche-Farmacologiche Cellulari-Molecolari, Sez. di Fisiologia Generale e Biofisica Cellulare, Università di
Pavia, Pavia, Italy
Submitted 14 December 2005; accepted in final form 27 April 2006
INTRODUCTION
Head rotations are initially signaled by semicircular canal
hair cells. Two types of hair cells are present in the sensory
epithelia of Amniotes, which are called type I and type II hair
cells— only type II hair cells are present in Anamniotes. Type
I and type II hair cells significantly differ in their morphology
and electrophysiology. Type I hair cells have a large apical
region bearing a short bundle of stereocilia and a very constricted neck compared with type II hair cells (Ricci et al.
1997). Moreover, type I hair cells basolateral region is enveloped by a single calyx afferent nervous terminal, whereas type
II hair cells are contacted by numerous bouton-like afferent and
efferent terminals. Finally, type I hair cells alone express a
low-voltage-activated outward rectifying K⫹ current, named
IK,L, which is responsible for the low cell membrane input
resistance in the voltage range close to the resting membrane
voltage found in vitro compared with type II hair cells (Rennie
and Correia 1994; Rüsch and Eatock 1996).
Address for reprint requests and other correspondence: S. Masetto, Dept. di
Scienze Fisiologiche-Farmacologiche Cellulari-Molecolari, Sez. di Fisiologia
Generale e Biofisica Cellulare, Università di Pavia, Via Forlanini 6, 27100
Pavia, Italy (E-mail: [email protected]).
602
It is commonly accepted that during head rotations the
stereocilia, which are located at the apex of the hair cells, are
bent as a consequence of the cupula deflection. Bending of the
stereocilia results in an increase of the open probability of
mechano-sensory ion channels located at the tip of the stereocilia (Hudspeth 1982). The resulting inflow of K⫹ ions would
depolarize the hair cell. The final shape of the receptor potential is also dependent on the specific array of basolateral ion
channels expressed; among these, voltage-dependent Ca channels are responsible for afferent transmitter release (Moser and
Beutner 2000; Parsons et al. 1994).
Whole cell recordings have shown that, in both type I and
type II hair cells, Ca2⫹ inflow occurs mainly through L-type Ca
channels (Almanza et al. 2003; Bao et al. 2003), which activate
close to – 60 mV, and show negligible inactivation when Ba2⫹
is the permeant ion. Aside from L-channels, the presence of
additional Ca channel types has been reported in type II hair
cells (Dou et al. 2004; Martini et al. 2000), whose properties,
however, have not yet been characterized.
As far as single Ca channel properties are concerned, to date
recordings have been performed in type II hair cells only. The
results, obtained by using the cell-attached configuration, have
shown the predominant expression of L-type Ca channels, plus
that of an “N-like” Ca channel type (Rodriguez-Contreras and
Yamoah 2001).
In the present paper we have investigated the single Ca
channel properties in both type I and type II hair cells from the
chick embryo semicircular canal. We found that the two hair
cell types express L-type Ca channels with similar properties.
Moreover, our results suggest that in physiological conditions
these channels might be already active at voltages that encompass the resting membrane potential of type I and type II hair
cells.
METHODS
Chicken embryo semicircular canal type I and type II hair cells
were studied at developmental stages ranging from embryonic day 13
(E13) to E21 (hatching).
Slice preparation
Detailed procedures for semicircular canal dissection and slice
preparation have been reported previously (Masetto et al. 2003).
Briefly, fertilized chicken eggs of the Cobb variety were obtained at
a local supplier and incubated at 38.3°C. Once removed from the eggs,
The costs of publication of this article were defrayed in part by the payment
of page charges. The article must therefore be hereby marked “advertisement”
in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
0022-3077/06 $8.00 Copyright © 2006 The American Physiological Society
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.32.246 on June 17, 2017
Zampini, Valeria, Paolo Valli, Giampiero Zucca, and Sergio
Masetto. Single-channel L-type Ca2⫹ currents in chicken embryo semicircular canal type I and type II hair cells. J Neurophysiol 96: 602– 612,
2006. First published May 10, 2006; doi:10.1152/jn.01315.2005. Few
data are available concerning single Ca channel properties in inner ear
hair cells and particularly none in vestibular type I hair cells. By using
the cell-attached configuration of the patch-clamp technique in combination with the semicircular canal crista slice preparation, we
determined the elementary properties of voltage-dependent Ca channels in chicken embryo type I and type II hair cells. The pipette
solutions included Bay K 8644. With 70 mM Ba2⫹ in the patch
pipette, Ca channel activity appeared as very brief openings at – 60
mV. Ca channel properties were found to be similar in type I and type
II hair cells; therefore data were pooled. The mean inward current
amplitude was ⫺1.3 ⫾ 0.1 (SD) pA at – 30 mV (n ⫽ 16). The average
slope conductance was 21 pS (n ⫽ 20). With 5 mM Ba2⫹ in the patch
pipette, very brief openings were already detectable at – 80 mV. The
mean inward current amplitude was ⫺0.7 ⫾ 0.2 pA at – 40 mV (n ⫽
9). The average slope conductance was 11 pS (n ⫽ 9). The mean open
time and the open probability increased significantly with depolarization. Ca channel activity was still present and unaffected when
␻-agatoxin IVA (2 ␮M) and ␻-conotoxin GVIA (3.2 ␮M) were added
to the pipette solution. Our results show that types I and II hair cells
express L-type Ca channels with similar properties. Moreover, they
suggest that in vivo Ca2⫹ influx might occur at membrane voltages
more negative than – 60 mV.
HAIR CELLS’ CA CHANNELS
Morphological criteria
Recordings were made from hair cells in selected regions, or zones,
of the neuroepithelium of the two vertical (posterior and anterior)
semicircular canals. The preparation consisted of a slice cut parallel to
the longitudinal axis of the crista and passing in between the two
eminentiae cruciatae. (see Fig. 1 in Masetto et al. 2003). This preparation allows recording from hair cells located at different distances
from the planum semilunatum. According to previous nomenclature
(Masetto et al. 2000, 2003), the two most peripheral regions of the
sensory epithelium in the slice are called zone 1, each contacting one
planum semilunatum (PS), which is a nonsensory epithelium; zone 3
is the most central region (which in the intact crista ampullaris would
be flanked by the two eminentiae cruciatae); zone 2 is a region
intermediate between zones 1 and 3. To increase our chances to record
from type I hair cells, for which no data on single Ca channels are
available in the literature, most of the recordings were made from
zone 2 because in birds, it contains the highest ratio of type I versus
type II hair cells (Kevetter et al. 2000; Masetto et al. 2000). Type I
hair cells were distinguished from type II hair cells by their amphora
shape, characterized by a very constricted region (neck) just below the
dense apical plate bearing the stereocilia (see for example Fig. 1D in
Masetto et al. 2003 and Fig. 1B in Masetto et al. 2005). The procedure
for sealing type I hair cells was rather complicated because calyx
remnants had often to be pulled out to reach the cell membrane. This
could require more than one attempt.
An additional feature distinguishing type I hair cells from type II
hair cells is the expression of IK,L, a low-voltage-activated outward
TABLE
rectifying K⫹ current (Rennie and Correia 1994; Rüsch and Eatock
1996). In a few hair cells, we could confirm the hair cell identity
electrophysiologically by achieving the ruptured whole cell configuration following cell-attached recording.
Electrical recordings
Patch pipettes were pulled from borosilicate glass pipettes (Hilgenberg GmbH, Malsfeld, Germany), tips were fire-polished, and partially coated with silicone elastomer (Sylgard; Dow Corning 184,
Midland, MI). For cell-attached experiments, micropipettes had a
resistance in the bath of 5–10 M⍀ when filled with Extra_Ba_70 or
Extra_Ba_5 (see Table 1). To control the patch transmembrane
potential, we zeroed the cell membrane voltage by perfusing the slice
with a high_K⫹ extracellular solution (Extra_high_K, see Table 1).
To verify that this was indeed occurring, in a few preliminary
experiments, we measured the hair cell membrane voltage in currentclamp mode in ruptured whole cell configuration before and during
perfusion with Extra_high_K; in these experiments, the pipette solution was a standard intracellular solution (Intra_K, Table 1), and
micropipettes had a resistance in the bath of 2–3 M⍀. We found that
after 3 min of perfusion, the cell membrane resting potential was
steadily shifted from –70, – 60 mV to 0 mV. Therefore cell-attached
recordings were started after ⱖ3 min of perfusion with the Extra_high_K. In a few cells, following cell-attached recording, we were
able to achieve the whole cell configuration. The value of the resting
membrane potential, read in current-clamp mode during the first
seconds in whole cell, i.e., before the cell content was washed out by
the pipette solution, confirmed that during cell-attached recording the
cell membrane potential was zeroed by the Extra_high_K solution.
Furthermore, recording in whole cell voltage-clamp mode allowed us
to confirm hair cell type I versus type II identity on the basis of the
total ionic current expressed. Unfortunately, a good seal and cell
access were only occasionally achieved during the passage from
cell-attached to whole cell because of the small pipette tip; on the
other hand these small pipette tips produced much better seals (⬎20
G⍀) and lower noise than larger pipette tips. As a routine, the crista
slice was changed after each recording, since perfusion with Extra_high_K accelerated deterioration of the preparation, likely as a
consequence of the prolonged cells’ depolarization. All recordings
were made at room temperature (22–24°C).
In cell-attached experiments, BaCl2 was used instead of CaCl2 to
emphasize currents through Ca channels and to block K channels.
Pipette solutions included TEACl, 4-aminopyridine, and CsCl to
block K channels (see Table 1), niflumic acid (Sigma) 50 ␮M to block
chloride channels, and (⫾)-Bay K 8644 (5 ␮M) (Sigma) to better
resolve L-channel openings, otherwise hardly detected (Hess et al.
1984). In a few experiments, Extra_Ba_5 included the P/Q, and N
type Ca channels blockers ␻-agatoxin IVA (2 ␮M) and ␻-conotoxin
GVIA (3.2 ␮M) (Alomone Laboratories).
1. Pipette and bath solutions
NaCl
Slicing solution
Extra_std.
Extra_Ba_70
Extra_Ba_5
Intra_K
Extra_high_K
145
145
KCl
3
3
134
135
HEPES
CaCl2
MgCl2
15
15
5
5
10
15
0.1
2
7.5
0.6
1
2
5
EGTA
CsCl
BaCl2
TEACl
4-AP
1
1
70
5
40
152
5
5
11
2
Quantities are expressed in mM. pH was brought to 7.4 with NaOH (Slicing solution and Extra_std.), KOH (Intra_K; Extra_high_K), or HCl (Extra_Ba_70
and Extra_Ba_5). Extra_Ba_70 and Extra_Ba_5 also contained niflumic acid (50 ␮M), (⫾)-Bay K 8644 (5 ␮M). The High_K solution also contained L-glutamine
(4 mM) and glucose (5 mM). Stock solutions of niflumic acid and Bay K 8644 were dissolved in DMSO 100% and kept in plastic vials ⬍0 °C. Each day of
experiment the content of a vial was allowed to melt and dissolved in the pipette solution. The final concentration of DMSO in the pipette solution was ⬍0.02%.
4-AP, 4-aminopyridine. Osmolarity was 305 mosM for extracellular solutions (Slicing sol., Extra_std., Extra_Ba_70, Extra_Ba_5, Extra_high_K), and 290 mosM
for intracellular solution (Intra_K).
J Neurophysiol • VOL
96 • AUGUST 2006 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.32.246 on June 17, 2017
embryos were decapitated and semicircular canals were dissected out.
The ampullae were incubated in D-MEM (No. 31600 – 026, GIBCO
BRL-Life Technologies), supplemented with 1.5% newborn calf serum (No. N-4637, Sigma, St. Louis, MO), NaHCO3 24 mM, piperazine- N,N⬘-bis (2-ethanesulfonic acid) 15 mM (Sigma), titrated to pH
7.4 with NaOH, and carboxygenated (95% O2-5% CO2) in a humidity-saturated chamber at 37°C. Osmolality was adjusted to ⬃320
mOsm. After an incubation period of 2– 6 h, the organ was removed
from the culture medium and embedded in 4% agar wt/vol (Sigma) in
a slicing solution (see Table 1).
An agar block containing the ampulla and a small portion of the
semicircular duct, was immersed in the slicing solution (partially
frozen) and cut using a vibratome (Dosaka). Slice thickness varied
between 150 and 250 ␮m. The slices were then transferred to a dish
with a glass coverslip bottom and immobilized using a weighted nylon
mesh. The tissue and microelectrode were viewed using differential
interference contrast optics employing an upright microscope (Zeiss
Axioskop) equipped with a ⫻63 water-immersion objective. The dish
contained a standard extracellular solution (Extra_std.; see Table 1).
Test solutions were delivered to the slice by way of a multibarreled
pipette, gravity fed.
603
604
V. ZAMPINI, P. VALLI, G. ZUCCA, AND S. MASETTO
Analysis
Data analysis was performed using Clampfit (pClamp version 9.2,
Axon Instrument), Microcal Origin (Version 6.0, Microcal Software,
Northampton, MA), Microsoft Excel V. 5.0c (Microsoft, Redmond,
WA), GraphPad Prism version 2.01 (GraphPad Software, San Diego,
CA). For statistical analysis, F-test and Student’s t-test were used
when comparing two groups of hair cells. Data are presented as
means ⫾ SE or ⫾ SD as specified in the text; n ⫽ number of cases.
Fast capacitive transients were minimized on-line by the patch-clamp
analogue compensation. Uncompensated capacitive currents were
corrected by averaging sweeps with no channel activity (nulls) and
subtracting them from the active sweeps. Event detection was performed with the 50% threshold detection method with each transition
visually inspected before being accepted. This allowed us to reconstruct idealized traces for calculating single-channel amplitude distribution, open and closed time histograms, and open probability (Po)
diagram. Po was calculated as the ratio: (total open time)/(total open
time ⫹ total closed time). Analyses were performed from the beginning to the end of the sweeps because, consistent with previous whole
cell measurements (Masetto et al. 2000, 2005), we found no evidence
of Ca channel inactivation during the sweep.
The total number of Ca channels (N) was initially estimated as the
largest number of simultaneously open channels seen in the record at
most depolarized voltages, at which potentials the probability of
channel opening was highest. Because, however, maximum Po was
ⱕ0.5, the following algorithm was applied to estimate the likelihood
of single-channel activity in those patches without superimposed
openings (Plummer et al. 1989): P2(T) ⫽ 1 ⫺ (1 ⫺ P2o)T/t, where
P2(T) is the cumulative probability of observing superimposed openings due to the activity of two identical channels over the total
observation time T, P2o is the overall probability of finding two
simultaneous openings, and t is twice the mean open time. At most
depolarized voltages P2(T) was ⬎0.999 despite the absence of superimposed openings during the observation time T.
Histograms for amplitude distribution at each voltage were fitted
with a single or second-order Gaussian distribution by using the
Levenberg-Marquardt algorithm to obtain the mean amplitude and
SD. Average Po values at each potential (Po-V diagrams; Fig. 4) were
calculated by averaging the mean values obtained from each patch,
excluding null sweeps, fitted with a Boltzmann function (Eq. 1)
J Neurophysiol • VOL
P o ⫽ 共Pmin ⫺ Pmax兲/兵1 ⫹ exp关共V ⫺ V1/2兲/k兴其 ⫹ Pmax
where Po represents the sweep, mean open probability Pmin and Pmax
the minimum and maximum open probability, V is voltage, and V1/2
is the voltage of half-maximum Po, k ⫽ slope factor.
Null sweeps were excluded from Po and open/closed times analysis
except when indicated. The rationale for excluding null sweeps from
Po and dwell times distributions is that, as already observed in
previous works concerning L-type Ca channels (Carabelli et al. 2001;
Hess et al. 1984; Nowycky et al. 1985), it can be useful to separately
analyze sweeps where the channel is actively gating (i.e., channel
“mode” 1” or “2”), from null sweeps, where the channel is unavailable
for opening (channel mode “0”). On the other hand, characterization
of null sweeps distribution as well is important by itself because it can
be affected by drugs, second messengers, cell metabolic state, hormones, etc. Nonetheless, to better compare present results with those
reported in the literature, values for Po including null sweeps are also
given in the text.
To calculate the open and closed time at each potential two
different approaches were used: 1) mean open (to) and closed (tc)
times obtained from each experiment were averaged; this method
gives an estimate of the two parameters independently from the
number of exponentials used for fitting the data in each experiment
and was used because in many experiments, only few data could be
obtained due to rare openings; 2) data from the analysis of all
experiments were pooled together to build a distribution of the open
and closed times on a log binned scale (20 bins/decade). The interpolation of the histograms with two or three exponentials with
maximum likelihood method (Sigworth and Sine 1987) provided the
time constant values for the open (␶o) and closed (␶c) states (Eq. 2)
冘
n
f(t) ⫽
Pi e关ln共t兲⫺ln共␶i兲兴e
ln共t兲⫺ln共␶i兲
i⫽1
where Pi and ␶i are the relative area and time constant of the ith
component of the distribution.
The half-amplitude threshold analysis method hitherto employed
failed to separate the sublevel events from the main levels. This was
an expected consequence of 1) the small elementary current amplitude, meaning a low signal-to-noise ratio (half-amplitude analysis
method requiring this ratio to be ⬎7) (Colquhoun and Sigworth 1983),
and 2) the low occurrence of the sublevels. Therefore to provide
quantitative measurements of the subconductance levels, we used the
mean-variance analysis method, which allows to exclude from the
analysis those sample points at the transitions between current levels
(Patlak 1988, 1993). By this analysis, the different current levels are
defined as periods of time in which the variance of the current is as
low as when no current is flowing (the background noise at the closed
state). The analysis consists of sliding a “window” of width N sample
points over the sampled data, meanwhile calculating the mean current,
m, and sample variance, s2, for the N points in the window. The
window is then advanced one point at a time, and the m-s2 paired
values are calculated each time. Each bin, m-s2, can then be plotted
versus its frequency of occurrence (number of events) in threedimensional histograms like the one showed in Fig. 6B. Histograms
were generated with the mvMachine software program (version
4.0.0.1, kindly provided by Dr. JB Patlak website www.physiology.
med.uvm.edu/patlak/, University of Vermont, Burlington, VT), converted through Histofilter (version 1.0.0.1, TODO), and plotted and
analyzed by Microcal OriginPro (Version 7.5, Microcal Software). In
the histogram, the closed level, the main open level, and the two
subconductance levels were obvious as peak volumes in the lowvariance region. We summed all the conductive events for each m
value the s2 of which was less than the background noise (Gollasch et
al. 1992; Patlak et al. 1993), and we fit the resulting three-peaks
distribution with the sum of three Gaussians. This provided the
numerical values for the mean current amplitude (⫾SD) and area for
each conductive level.
96 • AUGUST 2006 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.32.246 on June 17, 2017
The patch-clamp amplifier was an Axopatch 200B (Axon Instruments, Foster City, CA). The amplifier’s filter bandwidth was set at 2
kHz. Current and voltage were measured and controlled through a
DigiData 1322 interface (AD/DA converter; Axon Instruments) connected to a personal computer running pClamp software (V 9.2, Axon
Instruments). Data were digitized at 10 times the filter rate (i.e., 20
kHz), and analysis was limited to events lasting ⬎360 ␮s, i.e., longer
than twice the dead time (166 ␮s) (see Colquhoun and Hawkes 1995).
Openings were long enough and well resolved. Single-channel activity was recorded by applying 500-ms voltage pulses in the range from
⫺80 to 20 mV, from –70 mV holding potential (Vh). Consecutive
voltage steps were applied every 550 ms; each recording session
consisted of 50 or 100 pulses.
To minimize voltage errors due to liquid junction potentials between the pipette solution and the bath solution, prior to each experiment session, the offset potential was electronically zeroed with the
patch pipette tip immersed in the recording chamber containing the
same solution as the pipette. The bath chamber was then rinsed and
filled with standard extracellular solution (See Extra_std in Table 1).
The resulting offset between the pipette solution and the bath solution
was not adjusted because it would disappear following cell-attached
establishment (Neher 1992). Usually the offset potential did not
change by ⬎1–2 mV when measured at the end of the experiment,
indicating the absence of significant drifts of the circuit voltage
offsets.
HAIR CELLS’ CA CHANNELS
RESULTS
Cell-attached recordings
active channel, and a few patches two channels; data from the
latter patches were included in the single-channel currentvoltage relationship but excluded from the analysis of the mean
open probability and open and closed time.
Figure 1C shows the amplitude distribution for single-channel current at –30 mV; data were obtained with Extra_Ba_70
(top) or Extra_Ba_5 (bottom) as the pipette solution. Mean
amplitude was –1.3 ⫾ 0.04 (SE) pA with Extra_Ba_70 and
⫺0.5 ⫾ 0.06 pA with Extra_Ba_5. Data from type I and type
II hair cells were pooled because ANOVA (8 type I and 12 type
II hair cells) showed no significant differences in the unitary
current amplitude, slope conductance or voltage activation
threshold (see Table 2). Present data are consistent with type I
and type II hair cells expressing a similar population of Ca
channels as already observed for whole cell Ca currents (Bao
et al. 2003; Masetto et al. 2005).
The amplitude distributions were well fitted by a single
Gaussian function (Fig. 1C). Intermediate current levels were
rarely observed during channel transitions, presumably reflecting subconductance levels (see Fig. 6).
Figure 1D shows average current calculated from the same
patch over 500 consecutive sweeps at –20 mV. No timedependent inactivation of Ca channels was found, as singlechannel activity did not change significantly during 500-ms
depolarizing pulses.
Single-channel activity was unaffected by inclusion of
␻-agatoxin IVA (2 ␮M) and ␻-conotoxin GVIA (3.2 ␮M) in
the pipette solutions (n ⫽ 2; see legend for Fig. 1A).
Kinetics of single-channel currents
Figure 2 shows the average single-channel current-voltage
(I-V) plot for the inward current recorded with Extra_Ba_70
(F) or Extra_Ba_5 (■) as the pipette solution. The mean inward
current was always larger with Ba2⫹ 70 mM relative to Ba2⫹
5 mM in the pipette. A 20-mV shift in the voltage activation
threshold appeared because the channel activity was first deFIG. 1. Unitary L-type Ca channels currents.
A and B: representative traces from type I (n ⫽ 6)
and type II (n ⫽ 7) hair cells. Recordings were
obtained with Extra_Ba_70 (top) or Extra_Ba_5
(bottom) as the pipette solutions. Membrane
patches were held at –70 mV for 550 ms and
stepped for 500 ms to the potentials specified.
Arrows over the upper traces indicate the beginning and the end of the voltage steps; 8-ms traces
before and after the step are shown. Arrow heads
to the right of a few traces indicate the increase in
unitary currents amplitude upon repolarization to
–70 mV. The thin continuous lines indicate the
closed state. Records of A (all voltages in Extra_Ba_5) and B (–40, –30, and –20 mV in
Extra_Ba_5) were obtained with ␻-agatoxin
IVA (2 ␮M) and ␻-conotoxin GVIA (3.2 ␮M) in
the pipette solution (n ⫽ 2). C: amplitude distributions of events with duration ⬎0.36 ms at –30
mV. The open state amplitude distributions were
well fitted by a single Gaussian function (top:
n ⫽ 16 type I and type II hair cells, mean ⫽
⫺1.34 ⫾ 0.067 pA; bottom: n ⫽ 8 type I and
type II hair cells, mean ⫽ ⫺0.54 ⫾ 0.19 pA). D:
ensemble-averaged current obtained from 500
consecutive traces at –20 mV recorded in a type
II hair cell with Extra_Ba_70 as the pipette solution.
J Neurophysiol • VOL
96 • AUGUST 2006 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.32.246 on June 17, 2017
Patch pipettes were sealed onto the hair cells basolateral
surface, where Ca channels have been found either dispersed or
clustered at the active (presynaptic) zones in both otolith and
acoustic organs (Brandt et al. 2005; Issa and Hudspeth 1994;
Roberts et al. 1990; Rodriguez-Contreras and Yamoah 2001).
However, the majority of patches here investigated did not
show any single-channel activity, and those showing detectable
activity mostly expressed only one Ca channel. It is possible
that, by randomly patching the basolateral region, we never
encountered an active zone. Other alternative explanations may
exist: for example, Ca channel densities at the synaptic zones
of semicircular canal hair cells might be lower, or most Ca
channels are normally inhibited (or were inhibited in our
experimental conditions).
An outward single-channel current was sometimes detected.
Such patches were excluded from the analysis.
Figure 1, A and B, shows representative unitary currents
recorded from type I and type II hair cells respectively, with
Extra_Ba_70 (top) or Extra_Ba_5 (bottom) in the pipette (see
Table 1 for solutions’ composition). The bath solution had a
high content in K⫹ to set hair cells’ resting membrane potential
at 0 mV. Most recordings were from zone 2, which contains the
higher proportion of type I hair cells (see METHODS). Type I and
type II hair cells were classified based on their morphology
(Masetto et al. 2000); moreover, in a few cases, we were able
to confirm their identity electrophysiologically by shifting to
the whole cell configuration (see Whole cell recordings).
With 70 mM Ba2⫹ in the pipette, single-channel inward
currents first appeared at – 60 mV as very brief and rare
openings. Conversely, with 5 mM Ba2⫹ in the pipette, singlechannel inward currents were already detectable at – 80 mV.
The amplitude of the elementary current decreased with depolarization; inward currents were clearly detectable ⱕ20 mV
(data not shown). The majority of patches contained only one
605
606
V. ZAMPINI, P. VALLI, G. ZUCCA, AND S. MASETTO
2. Comparison between type I and type II hair cells
Ca channels
TABLE
Type I
Type II
Voltage activation threshold, mV
Extra_Ba 5 mM
⫺72.5 ⫾ 5.0 SD (4) ⫺74.0 ⫾ 5.5 SD (5)
Extra_Ba 70 mM
⫺54.0 ⫾ 5.5 SD (5) ⫺53.3 ⫾ 5.2 SD (6)
␣, pS
Extra_Ba 5 mM
10.8 (n ⫽ 4)
10.2 (n ⫽ 5)
Extra_Ba 70 mM
20.1 (n ⫽ 8)
22.1 (n ⫽ 12)
Open time constant, ms
␶o1
1.0
1.3
␶o2
8.6
5.2
␶o3
—
12.4
Closed time constant, ms
␶c1
0.4
0.4
␶c2
10.7
8.8
␶c3
148.3
73.2
Open and closed time constants (␶) calculated at ⫺20 mV with Extra_Ba_70. Voltage activation threshold was calculated as the arithmetic mean
of the voltages at which openings were first detected for each patch. Interpolation of open and closed times distributions was performed with the sum of
three exponential functions, except for type I hair cells open time distribution,
which was interpolated with the sum of two exponential functions because of
the paucity of long openings. These, however, were comparable with the long
openings observed in type II hair cells (range: 10 –70 ms). F-test and t-test
were used to compare voltage activation threshold and slope conductance of
type I and type II hair cells (P ⬎ 0.05). Number of cells in parentheses.
J Neurophysiol • VOL
FIG. 2. Current-voltage plot for Ca channels. Average single-channel current-voltage relationship (⫾SE) obtained with Extra_Ba_70 (F; data from 8
type I and 12 type II hair cells) and Extra_Ba_5 (■; data from 4 type I and 5
type II hair cells) as the pipette solution. The average values for the current
amplitude were obtained from amplitude histograms as in Fig. 1C. Data points
were fitted by linear equations with slope conductance of 21 pS with Extra_Ba_70 (4 ⱕ n ⱕ 18 patches for each voltage) and 11 pS with Extra_Ba_5
(2 ⱕ n ⱕ 9 patches for each voltage).
Values obtained by weighted means of the time constants
were in close agreement with to and tc. ␶o and ␶c had similar
values for type I and type II hair cells (see Table 2). L channels
may exhibit an increased activity following repolarization from
depolarized voltages (Jones 1998). We therefore checked for
any effect of predepolarization on single-channel activity on
repolarization to –70 mV with Extra_Ba_5 as the pipette
solution. We found no significant effect on the mean open time
or the mean open probability (Po); however, we found that for
openings that persisted on repolarization (single-channel tail
current, see for example arrow heads in Fig. 1, A and B), tail
current duration increased significantly with predepolarization
in the range from – 60 to 30 mV (Fig. 3D).
Po was not significantly different between type I and type II
hair cells, (P ⬎ 0.05 at all voltages with Extra_Ba_5 and
Extra_Ba_70, except P ⬎ 0.01 at –20 and – 40 mV with
Extra_Ba_70). Po was strictly voltage dependent, increasing
with depolarization (Fig. 4). As shown in Fig. 4A, with Extra_Ba_70 in the pipette, Po was 0.03 at –50 ⫾ 0.01 (SE) mV
(n ⫽ 6) and 0.28 at –20 ⫾ 0.15 mV (n ⫽ 12) with halfmaximum Po (V1/2) at –22.4 ⫾ 2.6 (SD) mV; the maximum Po
(Pomax) was 0.42 ⫾ 0.26 (SE) mV (n ⫽ 10) at 0 mV; k ⫽ 4.7.
With Extra_Ba_5 in the pipette, Po was 0.15 at – 60 ⫾ 0.12 mV
(n ⫽ 9) and 0.47 at –30 ⫾ 0.29 mV (n ⫽ 7), with V1/2 at
⫺45.7 ⫾ 6.15 (SD) mV; Pomax was 0.54 ⫾ 0.22 (SE) mV (n ⫽
3) at –20 mV (Fig. 4B); k ⫽ 6.9. Thus V1/2 shifted by
approximately ⫺23 mV in Ba2⫹ 5 mM relative to Ba2⫹ 70
mM. Figure 4B also shows the Po/V relationships obtained
after nulls inclusion (F). Po with nulls was 0.05 at – 60 mV and
0.16 at –30 mV. Po appeared to asymptote to a nonzero value
when null sweeps were excluded from the analysis, whereas it
did show a tendency to zero for voltages more negative than
– 80 mV when null sweeps were included. This might suggest
that the channel has a low intrinsic (i.e., voltage-independent)
probability of being open when it is outside the quiescent
mode, which would then increase up to Pomax following depolarization.
96 • AUGUST 2006 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.32.246 on June 17, 2017
tectable at – 60 mV in Ba2⫹ 70 mM and at – 80 mV in Ba2⫹ 5
mM. This leftward shift in 5 mM Ba2⫹ versus 70 mM Ba2⫹
was also consistent with the shift in the mean open probability
versus voltage curves (see Fig. 4, A and B). The voltage shift
was likely due to the different patch-membrane surface charge
screening effect by Ba2⫹ 70 mM versus Ba2⫹ 5 mM (see
DISCUSSION). In both experimental conditions, data were well
fitted with straight lines, indicating that the channel has a linear
conductance in the voltage range investigated. The average
slope conductance was 21 pS (n ⫽ 20) with Extra_Ba_70 and
11 pS (n ⫽ 9) with Extra_Ba_5 as the pipette solution.
Figure 3A shows the voltage dependence of the mean open
times (to) and mean closed times (tc) as from the arithmetic
means of open times (n ⫽ 20) and closed times (n ⫽ 13) of all
patches (Extra_Ba_70). to increased with voltage, whereas tc
decreased. Figure 3B shows the open times distribution at –20
mV (Extra_Ba_70). The distribution was well-fitted by the sum
of three exponential functions (Eq. 2) with open time constants
(ms): ␶o1 ⫽ 2.4 (29.2%), ␶o2 ⫽ 6.8 (68.4%), ␶o3 ⫽ 24.2 (2.4%).
The presence of time constants differing by one order of
magnitude suggests the presence of at least two gating modes,
“mode 1” (short openings, consistent with ␶o1 and ␶o2) and
“mode 2” (long openings, consistent with ␶o3) – see e.g., Fig.
5A, inset. In our recordings, the gating “mode 2” was likely
enhanced by the agonist Bay K 8644 (Hess et al. 1984;
Nowycky et al. 1985).
Figure 3C shows the closed times distribution at –20 mV.
Data were well fitted by the sum of three exponential functions
(Eq. 2). The time constants were (ms): ␶c1 ⫽ 0.5 (41.8%),
␶c2 ⫽ 8.8 (47.8%), ␶c3 ⫽ 76.6 (10.37%). These results indicate
that the channel can spend long periods in the closed state (see
null sweep probability in Fig. 4B). The significant differences
in closed time constants would reflect the presence of at least
two closed states.
HAIR CELLS’ CA CHANNELS
607
A high percentage of null sweeps (bars in Fig. 4, A and B)
was found at all voltages in both Extra_Ba_70 (mean value ⫽
0.51 ⫾ 0.08 SD) and Extra_Ba_5 (mean value ⫽ 0.63 ⫾ 0.01
SD). Because of the high variability among patches, it is
difficult to judge if null sweeps percentage decreased with
depolarization. Moreover, null sweeps percentage appeared to
decrease by increasing Ba2⫹ concentration. One possible explanation is that the dwell time in the closed state is both ion
and concentration dependent as suggested by Rodriguez-Contreras and Yamoah (2003), who also reported, in frog sacculus
hair cells, an increased propensity for L-type Ca channels to
enter into quiescent modes when Ba2⫹ or Ca2⫹ concentrations
were decreased.
Figure 5 shows a plot of Po versus recording time. Cachannel activity persisted unaltered for the whole recording
period (260 s) without evidence of channel run-down. Similar
results were found for recordings lasting 10 –20 min (data not
shown). The recordings’ sweeps with null or low activity
(Po ⬍ 0.02) appeared clustered. During these sweeps, Ca
channel openings were very brief (see e.g., Fig. 5, inset, b).
During sweeps’ clusters with higher Po, Ca channels could
both open in mode 1 (e.g., a) or mode 2 (e.g., c). However, we
did not distinguish clusters of channel activity in mode 1 or
mode 2 (i.e., sweeps dominated by mode 1 vs. sweeps dominate by mode 2).
Transitions from the open state to sub-conductance levels
were sometimes detectable at voltages between – 40 and –10
mV (see e.g., Fig. 6A). Following careful inspection of all
traces, these events appeared to occur very seldom. In those
sweeps where sub-conductance levels were found, we performed the mean-variance (MV) analysis (Patlak 1993) to
generate MV histograms like the one shown in Fig. 6B. At least
two sub-conductance levels were clearly detectable, showing a
mean amplitude at –30 mV of – 0.5 ⫾ 0.2 and – 0.9 ⫾ 0.1 (SD)
pA. The contribution of the sub-conductance levels to the
conductive state was low even for selected sweeps where
sublevel events were most frequent. For recordings at –20 and
–30 mV for example, the sum of the two sublevel events
accounted for little ⬍1% of the conductive events, that is the
channel spent almost 99% of the conductive state in the main
open level. Multiple conductance levels of L-type Ca channels
were previously reported (Kunze and Ritchie 1990).
2⫹
FIG. 4. Mean open probability and null-sweeps probability with Ba
70 mM or Ba2⫹ 5 mM as charge carrier. A: mean ⫾ SE open probability (Po) as a function of patch
membrane voltage with Extra_Ba_70 in the pipette; data were pooled from 13 type I and type II hair cells (5 ⱕ n ⱕ 12 patches at each voltage). Null sweeps were excluded.
B: mean ⫾ SE open probability (Po) as a function of patch membrane voltage with Extra_Ba_5 in the pipette; data pooled from 9 type I and type II hair cells (3 ⬍ n ⬍ 9 patches
at each voltage). Null sweeps were excluded (■) or included (F). In both sets of experiments, membrane patches were held at –70 mV for 550 ms and stepped for 500 ms to
the potentials specified. Average data for Po excluding null sweeps (■) were fitted with a Boltzmann function (Eq. 1). 䊐, mean percentage of null sweeps at each voltage. Error
bars are shown in only 1 direction for better clarity.
J Neurophysiol • VOL
96 • AUGUST 2006 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.32.246 on June 17, 2017
FIG. 3. Open and closed times. A: mean open time
(⫾SE; F) and mean closed time (⫾SE; 䊐) plotted as a
function of patch transmembrane voltage; data from
type I and type II hair cells (n ⫽ 20) with Extra_Ba_70
as the pipette solution. Data points represent the average of the mean open and closed times of each patch
(4 ⱕ n ⱕ 18 patches for each voltage). B and C: open
and closed time distributions at –20 mV (Extra_Ba_70).
Data were collected from type I and type II hair cells
(n ⫽ 18) and plotted in a square root-log binned
histogram. The distributions were fitted with the sum of
3 exponential functions. D: mean open time of singlechannel tail currents at –70 mV as a function of patch
membrane voltage conditioning step with Extra_Ba_5
in the pipette. Straight line through points indicates
linear regression; r2 ⫽ 0.55; P ⫽ 0.036.
608
V. ZAMPINI, P. VALLI, G. ZUCCA, AND S. MASETTO
FIG. 5. Open probability vs. time. Membrane
patch was held at –70 mV and repeatedly stepped
to –20 mV for 500 ms. (intersweeps interval ⫽ 550
ms). Pipette solution: Extra_Ba_70. Inset: portions
of traces characterized by short (“mode 1”; top)
and long (“mode 2”; bottom) openings, and a low
Po sweep with fast openings (middle). Arrows
above top trace indicate the beginning and the end
of the voltage steps; thin continuous lines indicate
the closed state. a– c identify the corresponding
position of the traces in the Po distribution. Recording is from a type II hair cell.
Occasionally, following cell-attached recording, we were
able to achieve the ruptured whole cell configuration without
losing the seal. Figure 7, A and B, show such two examples for
a type I and a type II hair cell, respectively. A large inward
current was present in both hair cells at the holding voltage
(– 60 mV). This is consistent with the high-monovalent cations’ concentration in the bath solution (135 mM K⫹, see
Extra_high_K in Table 1) versus the internal solution (1 mM
Cs⫹, see Extra_Ba_70). However, hyperpolarizing steps elicited inward currents the kinetics of which was significantly
different in the two cell types. In the type I hair cell, a large
instantaneous inward current was present on hyperpolarization,
which progressively deactivated. At the end of the –120 mV
pulse, the steady-state inward current was almost zero, indicating that almost all ion channels were closed at this membrane voltage. This behavior is consistent with IK,L expression
by this cell (Masetto et al. 2000; Rennie and Correia 1994;
Rüsch and Eatock 1996).
Conversely, in the type II hair cells, the inward current
showed both an instantaneous and a time-dependent compo-
nent at the beginning of the hyperpolarizing step. Moreover,
hyperpolarizing steps elicited only partial inactivation. This
behavior appears consistent with the expression of an anomalous inward rectifying K⫹ current (IK1), but not IK,L, by this
cell. Concerning the relatively large fraction of IK1 that is
active at – 60 mV, it must be considered that IK1 activation
range depends on the extracellular K⫹ concentration, shifting
rightward in the voltage axis by increasing extracellular K⫹
concentration (Hagiwara et al. 1976).
In both hair cells, little or no outward current was present at
the test potential (40 mV) due to the small monovalent cations
concentration in the internal solution.
DISCUSSION
This paper describes the elementary properties of voltagegated Ca channels in chick embryo semicircular canal hair cells
in a slice preparation. A novel finding of the present study
concerns the hyperpolarized voltage activation threshold of
single L-type Ca channels. Moreover, this is the first characterization of single Ca channel properties in type I hair cells.
FIG. 6. Sub-conductance levels of unitary Ca channel currents. A: single-channel data from a type II hair cell (E14) recorded at –30 mV with Extra_Ba_70
as the pipette solution. The traces shown were selected for the particularly high sublevels frequency of occurrence and duration (see 1). - - -, sub-conductance
levels; —, closed level. B: mean-variance relationship of currents shown in A. Window width ⫽ 50 sample points. The 3-dimensional histogram illustrate number
of observations for any single combination of mean and variance (see METHODS). The large peak on the right represents number of events in the closed state (mean
current ⫽ 0 pA). The variance at the base of the peak spans the background noise. The other peaks in the same low-variance region (i.e., variance ⬍0.1) reflected
conductive states (main open and sublevels). Events in the arch region (variance ⬎0.1) reflect the transitions among the different current levels. These points
were excluded from the analysis. Note that, consistent with visual inspection of the traces in A, transitions into and out the sublevels appear to arch from the
open level.
J Neurophysiol • VOL
96 • AUGUST 2006 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.32.246 on June 17, 2017
Whole cell recordings
HAIR CELLS’ CA CHANNELS
Ca channel classification
Ca channel activity was still present and not affected when
P/Q, and N type Ca channels blockers were added to the pipette
solution. Because R-type Ca channels have a significantly
lower slope conductance (13 pS in Ba2⫹ 110 mM) than found
here and because T-type Ca channels show rapid (␶ ⬃ 20 –50
ms) (Hille 2001) and complete inactivation, we identified type
I and type II hair cells Ca channels as L-type.
L-type Ca channels are multisubunit complexes constituted
by different isoforms of the pore-forming ␣1 subunit, named
␣1S (CaV1.1), ␣1C (CaV1.2), ␣1D (CaV1.3), and ␣1F (CaV1.4).
Most literature agrees that hair cell L-type Ca channels activate
close to – 60 mV. This feature would distinguish them from
classical neuronal L-type Ca channels, which activate 20 –30
mV less negative (Ertel et al. 2000). However, more recently it
has been reported that the CaV1.3 (␣1D) subunit of L-type Ca
channels activates at significantly more negative voltages than
the other L-type CaV1 subunits (Koschak et al. 2001; Mangoni
J Neurophysiol • VOL
et al. 2003; Xu and Lipscombe 2001; Zhang et al. 2002 –
reviewed in Lipscombe et al. 2004). The expression of the
CaV1.3 subunit has been reported in cochlear hair cells (Green
et al. 1996; Kollmar et al. 1997; Platzer et al. 2000) as well as
in mammalian vestibular epithelia (Bao et al. 2003). Also
consistent are the observations that chick cochlear hair cell
L-type Ca channels (Zidanic and Fuchs 1995) and the CaV1.3
subunit (Xu and Lipscombe 2001) are incompletely inhibited
by dihydropyridines. Finally, genetic ablation of the CaV1.3
subunit resulted in cochlear hair cell exocytosis dysfunction
(Brandt et al. 2003) and congenital deafness (Dou et al. 2004),
although vestibular function did not appear compromised,
which might be explained by the expression of other Ca
channel types in vestibular hair cells (see following text).
Given the hyperpolarized activation range found here, the most
reasonable hypothesis is that the single Ca channels activity we
recorded involved the CaV1.3 subunit.
On the other hand, we did not find evidence for other Ca
channel types. The expression of Ca channels involving different ␣(1) subunits was reported in frog vestibular hair cells
(Martini et al. 2000; Rodriguez-Contreras and Yamoah 2001).
Moreover, recent works on CaV1.3-(␣1D)-knock-out mice
showed the expression of a persisting Ca channel population in
acoustic and vestibular hair cells (Brandt et al. 2003; Dou et al.
2004; Platzer et al. 2000). Because we limited our analysis to
the basolateral region of hair cells, and these were mostly from
crista zone 2, it is possible that other Ca channel types are
expressed in different cell membrane regions, for example
close to the apical surface and/or different crista zones.
Ca channel biophysical properties
No data are available in the literature concerning the elementary properties of the CaV1.3␣1-subunit-containing Ca
channel. However, present data can be placed in the context of
the more general literature on L-type Ca channels. For the
following discussion, it is worth recalling that all excitable
cells, except skeletal muscle and perhaps retina, express the
CaV1.2 and the CaV1.3 gene (see Lipscombe et al. 2004 for a
review). The CaV1.3 gene is coexpressed in many cells with
CaV1.2; in neurons, CaV1.2 and CaV1.3 are also often found in
the same membrane compartments; in the heart, conversely,
CaV1.3 is present in atrial tissue, but not in ventricular muscle
that expresses CaV1.2.
As far as single-channel gating properties are concerned, the
two different mean open and closed times found here resemble
brief and long openings of L channels in cardiac ventricular
(Hess et al. 1984) and neuronal (Nowycky et al. 1985) cells,
called “mode 1” for brief and “mode 2” for long lifetimes,
respectively. However, our analysis of the recorded traces did
not reveal a clear segregation of mode 1 or mode 2 as in Hess
et al. (1984). The frequent occurrence of null sweeps (mode 0)
as well is reminiscent of L-type Ca channels described in
cardiac (Hess et al. 1984), neuronal (Nowycky et al. 1985), and
chromaffin cells (Carabelli et al. 2001). Interestingly, Ca channel opening probability appears to result from voltage dependence of Po between a minimum (intrinsic) and maximum level
during active sweeps and voltage dependence of null sweeps
frequency.
Finally, the finding that single-channel tail current duration
increased with predepolarization is consistent with increased
96 • AUGUST 2006 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.32.246 on June 17, 2017
FIG. 7. Whole cell recordings. A: whole cell ionic currents recorded from a
hair cell morphologically identified as type I following single Ca channel
recording in cell-attached configuration. Pipette solution was Extra_Ba_70.
The voltage protocol is shown above and applies to B as well. At – 60 mV, a
large inward current is present (- - -, 0-current level). Hyperpolarizing steps
elicit an increase of the instantaneous inward current amplitude, followed by a
progressive decrease. The steady-state inward current amplitude diminished by
increasing hyperpolarization, being almost zero at the end of the –120 mV
pulse. No outward current was present. This is consistent with K⫹ currents
reversal potential ⬎123 mV, as estimated from Nernst equilibrium potential
for external K⫹ (135 mM) vs. internal Cs⫹ (1 mM) and considering that Cs⫹
is generally less permeant than K⫹ through K channels. B: whole cell ionic
currents recorded from a hair cell morphologically identified as type II,
following single Ca channel recording in cell-attached configuration. Pipette
solution was Extra_Ba_5. At – 60 mV a large inward current is present (- - -,
0-current level). The inward current instantaneous amplitude increases with
hyperpolarization and then inactivates only partially so that a large inward
current is still present at the end of all hyperpolarizing steps. A small outward
current is present at 40 mV, which we believe consists of “leakage” current. In
fact, we would expect no outward currents through outward rectifier K
channels because of the very positive reversal potential estimated (see preceding text) and of the presence of TEA, 4 AP, and Bd2⫹ in the pipette, which
should block outward rectifier K channels. Current traces (0.2 to 0.5 ms) were
blanked in correspondence of capacitive transients. Calibration bars apply to A
and B.
609
610
V. ZAMPINI, P. VALLI, G. ZUCCA, AND S. MASETTO
J Neurophysiol • VOL
channel gating has a much higher affinity for Ca2⫹ than for
Ba2⫹.
However, a caveat concerns the use of Bay K 8644, which
has been reported to exert part of its agonistic effect on L-type
calcium channels by shifting the voltage-dependence of channel gating toward more negative membrane potentials. However, this would be obtained by Bay K 8644 increasing the
chances that the channel stays open longer (Cena et al. 1989;
Hess et al. 1984; Markwardt and Nilius 1988; Nowycky et al.
1985; Sanguinetti et al. 1986). Also, in pancreatic b-cells,
which express the CaV1.3 subunit of the L-type Ca channel
(Ertel et al. 2000; Scholze et al. 2001), no effects of Bay K
8644, tested ⱕ1 mM, were found on the Ca channel voltage
dependence (Smith et al. 1993).
Functional implications of single Ca channel properties
Depending on the cell system of expression, L-type Ca
channels may have quite different roles. In neurons, L-type
(CaV1.2 and CaV1.3) Ca channels are particularly important in
translating synaptic activity into alterations in gene expression
and cellular function (e.g., Bito et al. 1996; Deisseroth et al.
2003; Dolmetsch et al. 2001; Graef et al. 1999; Weick et al.
2003). On the other hand, L-type Ca channels do not appear
involved in synaptic transmission between neurons, which
conversely depends on presynaptic Ca2⫹ influx through N-type
(CaV2.2) and P/Q-type (CaV2.1) Ca channels (Dunlap et al.
1995). L-type Ca channels are expressed in vertebrate cardiac,
skeletal, and smooth muscle and are prominent in several
endocrine cells, in rod photoreceptors, and in certain nonspiking synaptic terminals that secrete continuously. Their most
obvious function is to mediate Ca2⫹ entry in cells that contract
or secrete in response to long or steady depolarizations (Hille
2001).
A similar role appears convenient for L-type calcium channels in hair cells, which sustain a tonic release of afferent
transmitter. Consonant with this, several pieces of evidence
exist that hair cell afferent transmission involves L-type Ca
channels (Brandt et al. 2003; Perin et al. 2000; Spassova et al.
2001). Furthermore, the hyperpolarized threshold of the
CaV1.3 subunit would suit the voltage operating range of hair
cells, the resting membrane potential of which has generally
been reported to be close to, or more negative than, – 60 mV.
The present results, together with previous whole cell data
showing that Ca channel properties are very similar in 5 mM
Ba2⫹ or 2 mM Ca2⫹ (Masetto et al. 2005), suggest that
vestibular hair cells Ca channels might be already active at – 80
and –70 mV in physiological conditions (i.e., in 2 mM Ca2⫹).
If this hypothesis was true, hair cell bidirectional operating
range would result extended toward the negative voltage region. Moreover, it might at least in part account for the type I
hair cell “paradox.” In vitro in fact, type I hair cells show a
resting membrane potential more negative than –70 mV,
which, together with a low membrane input resistance (Masetto et al. 2000; Rennie and Correia 1994; Rüsch and Eatock
1996), would preclude Ca-dependent neurotransmitter release
if the Ca current activated at, or less negative than, – 60 mV.
This would not be consistent with in vivo recordings from
rodent calyx afferents, showing both background and evoked
afferent discharge (Baird et al. 1988; Goldberg et al. 1990;
Rennie and Streeter 2006) and spontaneous excitatory postsyn-
96 • AUGUST 2006 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.32.246 on June 17, 2017
activity of L-type Ca channels following repolarization from
depolarized voltages (Jones 1998).
Present experiments revealed that type I and type II hair cells
express Ca channels with similar properties (see also Table 1):
voltage activation threshold at – 80/– 60 mV with 5/70 mM
Ba2⫹, negligible inactivation, and a unitary conductance of
11/21 pS with 5/70 mM Ba2⫹. A similar Ca channel unitary
conductance was found in hair cells of the chick cochlea (24 pS
in 110 mM Ba2⫹) (Kimitsuki et al. 1994) and frog saccule (21
pS in 65 mM Ba2⫹) (Rodriguez-Contreras and Yamoah 2001).
The maximum Po in our recordings was 0.54 (with BayK 8644
and excluding null sweeps), analogous to that found in frog
sacculus hair cells (0.4 with BayK 8644) (Rodriguez-Contreras
and Yamoah 2001), but significantly less than reported in inner
hair cells (0.82 without BayK 8644; Brandt et al. 2005). So far
dwell times distribution have been characterized only in frog
sacculus hair cells (Rodriguez-Contreras and Yamoah 2001),
where L channels showed three mean open times (␶o ⫽ 0.3,
9.7, and 22.7 ms), and three mean closed times (␶c ⫽ 1.5, 10.3,
and 56.6 ms) rather similar to those found here (␶o ⫽ 2.4, 6.8,
and 24.2 ms; ␶c ⫽ 0.5, 8.8, and 76.6 ms).
An original finding of the present paper concerns the hyperpolarized voltages at which single L-type Ca channel activity
could be detected. With 5 mM Ba2⫹ in the patch pipette, brief
unitary Ca channel openings were detected at voltages as
negative as – 80 mV. Single events were rather short and
infrequent at – 80 and –70 mV; therefore they should produce
a small total inward current, possibly hard to detect in whole
cell configuration. Indeed, in this same preparation we reported
whole cell inward currents through Ca channels activating
close to – 60 mV (Masetto et al. 2005). However, in rat crista
type I hair cells, Bao et al. (2003), by interpolating the I-V
relationship, estimated that 1% of the peak total Ca current
recorded with 1.3 mM Ca2⫹ was available at –72 mV, thus
providing a first hint that L-type Ca channels in hair cells might
be active at voltages more negative than previously thought.
The 20-mV more-negative activation range found here with
Ba2⫹ 5 mM, relative to Ba2⫹ 70 mM, accompanied with a ⫺23
mV shift in Ca channel Po. These phenomena should reflect the
local increase in negative surface charge screening effect
exerted by divalent cations, which shifts the activation curve of
voltage-dependent Ca channels toward more depolarized voltages (Frankenhaeuser and Hodgkin 1957; Hille 2001; Kostyuk
et al. 1982). As a result, stronger depolarization would be
required to open the same number of channels following
increase of divalent cations concentration.
We speculate that in physiological conditions (i.e., with 2
mM Ca2⫹ in the perilymph) Ca channels might activate at such
negative voltages as found here in 5 mM Ba2⫹ (i.e., at – 80
mV). The rationale for this hypothesis is that we found that the
activation kinetics of the inward current at several potentials
was similar in 5 mM Ba2⫹ relative to 2 mM Ca2⫹ (Masetto et
al. 2005), and it has been reported that the position of the
CaV1.3␣1 current-voltage curve obtained with 2 mM Ca2⫹
overlaps with that recorded in 5 mM Ba2⫹ (Xu and Lipscombe
2001), as also reported for the activation curve of L-type Ca
channels in rat neocortical neurons (Lorenzon and Foehring
1995). These results are consistent with previous studies showing that Ca2⫹ is more effective than Ba2⫹ at shifting the gating
of voltage-dependent Na (Hille et al. 1975) and Ca (Smith et al.
1993) channels; that is, the surface charge associated with
HAIR CELLS’ CA CHANNELS
aptic currents (Rennie and Streeter 2006). To reconcile these
results, it has been proposed that type I hair cells are depolarized by K⫹ accumulation in the synaptic cleft (Goldberg 1996;
Soto et al. 2002) or by IK,L inhibition by cGMP (Behrend et al.
1997) or NO (Chen and Eatock 2000). Here we add the
possibility that Ca2⫹ inflow resulting from the short and
infrequent openings at voltages between – 80 and – 60 mV be
sufficient to trigger afferent transmitter release in these cells.
Noteworthy, the driving force for Ca2⫹ inflow is high at such
negative voltages; moreover, Brandt et al. (2005) have recently
suggested that transmitter release in inner hair cells is under the
control of an intracellular Ca2⫹ nanodomain shaped by the
gating of one or few Ca channels.
ACKNOWLEDGMENTS
GRANTS
This work was supported by the Ministero dell’Università e della Ricerca
Scientifica e Tecnologica, Rome.
REFERENCES
Almanza A, Vega R, and Soto E. Calcium current in type I hair cells isolated
from the semicircular canal crista ampullaris of the rat. Brain Res 994:
175–180, 2003.
Baird RA, Desmadryl G, Fernandez C, and Goldberg J. The vestibular
nerve of the chinchilla. II. Relation between afferent response properties and
peripheral innervation pattern in the semicircular canals. J Neurophysiol 60:
182–203, 1988.
Bao H, Wong WH, Goldberg JM, and Eatock RA. Voltage-gated calcium
channel currents in type I and type II hair cells isolated from the rat crista.
J Neurophysiol 90: 155–164, 2003.
Behrend O, Schwark C, Kunihiro T, and Strupp M. Cyclic GMP inhibits
and shifts the activation curve of the delayed-rectifier (I[K1]) of type I
mammalian vestibular hair cells. Neuroreport 8: 2687–2690, 1997.
Bito H, Deisseroth K, and Tsien RW. CREB phosphorylation and dephosphorylation: a Ca(2⫹)- and stimulus duration-dependent switch for hippocampal gene expression. Cell 87: 1203–1214, 1996.
Brandt A, Khimich D, and Moser T. Few CaV1.3 channels regulate the
exocytosis of a synaptic vesicle at the hair cell ribbon synapse. J Neurosci
25: 11577–11585, 2005.
Brandt A, Striessnig J, and Moser T. CaV1.3 channels are essential for
development and presynaptic activity of cochlear inner hair cells. J Neurosci
23: 10832–10840, 2003.
Carabelli V, Hernandez-Guijo JM, Baldelli P, and Carbone E. Direct
autocrine inhibition and cAMP-dependent potentiation of single L-type
Ca2⫹ channels in bovine chromaffin cells. J Physiol 532: 73–90, 2001.
Cena V, Stutzin A, and Rojas E. Effects of calcium and Bay K-8644 on
calcium currents in adrenal medullary chromaffin cells. J Membr Biol 112:
255–265, 1989.
Chen JW and Eatock RA. Major potassium conductance in type I hair cells
from rat semicircular canals: characterization and modulation by nitric
oxide. J Neurophysiol 84: 139 –151, 2000.
Colquhoun D and Hawkes AG. The principles of the stochastic interpretation
of ion channel mechanisms. In: Single Channel Recording, edited Sackmann
B and Neher E. New York: Plenum, p. 397– 479, 1995.
Colquhoun D and Sigworth FJ. Fitting and statistical analysis of singlechannel records. In: Single-Channel Recording,. Edited by Sakmann B and
Neher E. New York: Plenum, p. 191–263, 1983.
Deisseroth K, Mermelstein PG, Xia H, and Tsien RW. Signaling from
synapse to nucleus: the logic behind the mechanisms. Curr Opin Neurobiol
13: 354 –365, 2003.
Dolmetsch RE, Pajvani U, Fife K, Spotts JM, and Greenberg ME.
Signaling to the nucleus by an L-type calcium channel-calmodulin complex
through the MAP kinase pathway. Science 294: 333–339, 2001.
Dou H, Vazquez AE, Namkung Y, Chu H, Cardell EL, Nie L, Parson S,
Shin HS, and Yamoah EN. Null mutation of alpha1D Ca2⫹ channel gene
results in deafness but no vestibular defect in mice. J Assoc Res Otolaryngol
5: 215–226, 2004.
J Neurophysiol • VOL
Dunlap K, Luebke JI, and Turner TJ. Exocytotic Ca2⫹ channels in mammalian central neurons. Trends Neurosci 18: 89 –98, 1995.
Ertel EA, Campbell KP, Harpold MM, Hofmann F, Mori Y, Perez-Reyes
E, Schwartz A, Snutch TP, Tanabe T, Birnbaumer L, Tsien RW, and
Catterall WA. Nomenclature of voltage-gated calcium channels. Neuron
25: 533–535, 2000.
Frankenhaeuser B and Hodgkin AL. The action of calcium on the electrical
properties of squid axons. J Physiol 137: 218 –244, 1957.
Goldberg JM. Theoretical analysis of intercellular communication between
the vestibular type I hair cell and its calyx ending. J Neurophysiol 75:
1942–1957, 1996.
Goldberg JM, Desmadryl G, Baird RA, and Fernandez C. The vestibular
nerve of the chinchilla. V. Relation between afferent discharge properties
and peripheral innervation patterns in the utricular macula. J Neurophysiol
63: 791– 804, 1990.
Gollasch M, Hescheler J, Quayle JM, Patlak JB, and Nelson MT. Single
calcium channel currents of arterial smooth muscle at physiological calcium
concentrations. Am J Physiol Cell Physiol 263: C948 –952, 1992.
Graef IA, Mermelstein PG, Stankunas K, Neilson JR, Deisseroth K, Tsien
RW, and Crabtree GR. L-type calcium channels and GSK-3 regulate the
activity of NF-ATc4 in hippocampal neurons. Nature 401: 703–708, 1999.
Green GE, Khan KM, Beisel DW, Drescher MJ, Hatfield JS, and Drescher
DG. Calcium channel subunits in the mouse cochlea. J Neurochem 67:
37– 45, 1996.
Hagiwara S, Miyazaki S, and Rosenthal NP. Potassium current and the
effect of cesium on this current during anomalous rectification of the egg
cell membrane of a starfish. J Gen Physiol 67: 621– 638, 1976.
Hess P, Lansman JB, and Tsien RW. Different modes of Ca channel gating
behavior favoured by dihydropyridine Ca agonists and antagonists. Nature
311: 538 –544, 1984.
Hille B. Ion Channels of Excitable Membrane (3rd Ed.). Sunderland, MA:
Sinauer, 2001.
Hille B, Woodhull AM, and Shapiro BI. Negative surface charge near
sodium channels of nerve: divalent ions, monovalent ions, and pH. Philos
Trans R Soc Lond B Biol Sci 270: 301–318, 1975.
Hudspeth AJ. Extracellular current flow and the site of transduction by
vertebrate hair cells. J Neurosci 2: 1–10, 1982.
Issa NP and Hudspeth AJ. Clustering of Ca2⫹ channels and Ca(2⫹)-activated
K⫹ channels at fluorescently labeled presynaptic active zones of hair cells.
Proc Natl Acad Sci USA 91: 7578 –7582, 1994.
Jones SW. Overview of voltage-dependent Ca channels. J Bioenerg Biomembr
30: 299 –312, 1998.
Kevetter GA, Blumberg KR, and Correia MJ. Hair cell and supporting cell
density and distribution in the normal and regenerating posterior crista
ampullaris of the pigeon. Int J Dev Neurosci 18: 855– 867, 2000.
Kimitsuki T, Nakagawa T, Hisashi K, Komune S, and Komiyama S. Single
channel recordings of calcium currents in chick cochlear hair cells. Acta
Otolaryngol 114: 144 –148, 1994.
Kollmar R, Montgomery LG, Fak J, Henry LJ, and Hudspeth AJ.
Predominance of the alpha1D subunit in L-type voltage-gated Ca2⫹ channels of hair cells in the chicken’s cochlea. Proc Natl Acad Sci USA 94:
14883–14888, 1997.
Koschak A, Reimer D, Huber I, Grabner M, Glossmann H, Engel J, and
Striessnig J. Alpha 1D (Cav1.3) subunits can form l-type Ca2⫹ channels
activating at negative voltages. J Biol Chem 276: 22100 –22106, 2001.
Kostyuk PG, Mironov SL, Doroshenko PA, and Ponomarev VN. Surface
charges on the outer side of mollusk neuron membrane. J Membr Biol 70:
171–179, 1982.
Kunze DL and Ritchie AK. Multiple conductance levels of the dihydropyridine-sensitive calcium channel in GH3 cells. J Membr Biol 118: 171–178,
1990.
Lipscombe D, Helton TD, and Xu W. L-type calcium channels: the low
down. J Neurophysiol 92: 2633–2641, 2004.
Lorenzon NM and Foehring RC. Characterization of pharmacologically
identified voltage-gated calcium channel currents in acutely isolated rat
neocortical neurons. I. Adult neurons. J Neurophysiol 73: 1430 –1442, 1995.
Mangoni ME, Couette B, Bourinet E, Platzer J, Reimer D, Striessnig J,
and Nargeot J. Functional role of L-type Cav1.3 Ca2⫹ channels in cardiac
pacemaker activity. Proc Natl Acad Sci USA 100: 5543–5548, 2003.
Markwardt F and Nilius B. Modulation of calcium channel currents in
guinea-pig single ventricular heart cells by the dihydropyridine Bay K 8644.
J Physiol 399: 559 –575, 1988.
96 • AUGUST 2006 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.32.246 on June 17, 2017
We thank Drs. Valentina Carabelli and Manning J. Correia for helpful
comments on the manuscript.
611
612
V. ZAMPINI, P. VALLI, G. ZUCCA, AND S. MASETTO
J Neurophysiol • VOL
Roberts WM, Jacobs RA, and Hudspeth AJ. Colocalization of ion channels
involved in frequency selectivity and synaptic transmission at presynaptic
active zones of hair cells. J Neurosci 10: 3664 –3684, 1990.
Rodriguez-Contreras A and Yamoah EN. Direct measurement of singlechannel Ca2⫹ currents in bullfrog hair cells reveals two distinct channel
subtypes. J Physiol 534: 669 – 689, 2001.
Rodriguez-Contreras A and Yamoah EN. Effects of permeant ion concentrations on the gating of L-type Ca2⫹ channels in hair cells. Biophys J 84:
3457–3469, 2003.
Rüsch A, and Eatock RA. A delayed rectifier conductance in type I hair cells
of the mouse utricle. J Neurophysiol 76: 994 –1004, 1996.
Sanguinetti MC, Krafte DS, and Kass RS. Voltage-dependent modulation of
Ca channel current in heart cells by Bay K8644. J Gen Physiol 88: 369 –392,
1986.
Scholze A, Plant TD, Dolphin AC, and Nurnberg B. Functional expression
and characterization of a voltage-gated CaV1.3 (alpha1D) calcium channel
subunit from an insulin-secreting cell line. Mol Endocrinol 15: 1211–1221,
2001.
Sigworth FJ and Sine SM. Data transformations for improved display and
fitting of single-channel dwell time histograms. Biophys J 52: 1047–1054,
1987.
Smith PA, Aschroft FM, Frances M, and Fewtrell CM. Permeation and
gating properties of the L-type calcium channel in mouse pancreatic beta
cells. J Gen Physiol 101: 767–797, 1993.
Soto E, Vega R, and Budelli R. The receptor potential in type I and type II
vestibular system hair cells: a model analysis. Hear Res 165: 35– 47, 2002.
Spassova M, Eisen MD, Saunders JC, and Parsons TD. Chick cochlear hair
cell exocytosis mediated by dihydropyridine-sensitive calcium channels.
J Physiol 535: 689 – 696, 2001.
Weick JP, Groth RD, Isaksen AL, and Mermelstein PG. Interactions with
PDZ proteins are required for L-type calcium channels to activate cAMP
response element-binding protein-dependent gene expression. J Neurosci
23: 3446 –3456, 2003.
Xu W and Lipscombe D. Neuronal Ca(V)1.3alpha(1) L-type channels activate at relatively hyperpolarized membrane potentials and are incompletely
inhibited by dihydropyridines. J Neurosci 21: 5944 –5951, 2001.
Zhang Z, Xu Y, Song H, Rodriguez J, Tuteja D, Namkung Y, Shin HS, and
Chiamvimonvat N. Functional Roles of Ca(v)1.3 (alpha(1D)) calcium
channel in sinoatrial nodes: insight gained using gene-targeted null mutant
mice. Circ Res 90: 981–987, 2002.
Zidanic M and Fuchs PA. Kinetic analysis of barium currents in chick
cochlear hair cells. Biophys J 68: 1323–1336, 1995.
96 • AUGUST 2006 •
www.jn.org
Downloaded from http://jn.physiology.org/ by 10.220.32.246 on June 17, 2017
Martini M, Rossi ML, Rubbini G, and Rispoli G. Calcium currents in hair
cells isolated from semicircular canals of the frog. Biophys J 78: 1240 –
1254, 2000.
Masetto S, Bosica M, Correia MJ, Ottersen OP, Zucca G, Perin P, and
Valli P. Na⫹ currents in vestibular type I and type II hair cells of the embryo
and adult chicken. J Neurophysiol 90: 1266 –1278, 2003.
Masetto S, Perin P, Malusà A, Zucca G, and Valli P. Membrane properties
of chick semicircular canal hair cells in situ during embryonic development.
J Neurophysiol 83: 2740 –2756, 2000.
Masetto S, Zampini V, Zucca G, and Valli P. Ca2⫹ currents and voltage
responses in type I and type II hair cells of the chick embryo semicircular
canal. Pflugers Arch 451: 395– 408, 2005.
Moser T and Beutner D. Kinetics of exocytosis and endocytosis at the
cochlear inner hair cell afferent synapse of the mouse. Proc Natl Acad Sci
USA 97: 883– 888, 2000.
Neher E. Correction for liquid junction potentials in patch clamp experiments.
Methods Enzymol 207: 123–131, 1992.
Nowycky MC, Fox AP, and Tsien RW. Long-opening mode of gating of
neuronal calcium channels and its promotion by the dihydropyridine calcium agonist Bay K 8644. Proc Natl Acad Sci USA 82: 2178 –2182, 1985.
Parsons TD, Lenzi D, Almers W, and Roberts WM. Calcium-triggered
exocytosis and endocytosis in an isolated presynaptic cell: capacitance
measurements in saccular hair cells. Neuron 13: 875– 883, 1994.
Patlak JB. Sodium channel subconductance levels measured with a new
variance-mean analysis. J Gen Physiol 92: 413– 430, 1988.
Patlak JB. Measuring kinetics of complex single ion channel data using
mean-variance histograms. Biophys J 65: 29 – 42, 1993.
Perin P, Soto E, Vega R, Botta L, Masetto S, Zucca G, and Valli P. Calcium
channels functional roles in the frog semicircular canal. Neuroreport 11:
417– 420, 2000.
Platzer J, Engel J, Schrott-Fischer A, Stephan K, Bova S, Chen H, Zheng
H, and Striessnig J. Congenital deafness and sinoatrial node dysfunction in
mice lacking class D L-type Ca2⫹ channels. Cell 102: 89 –97, 2000.
Plummer MR, Logothetis DE, and Hess P. Elementary properties and
pharmacological sensitivities of calcium channels in mammalian peripheral
neurons. Neuron 2: 1453–1463, 1989.
Rennie KJ and Correia MJ. Potassium currents in mammalian and avian
isolated type I semicircular canal hair cells. J Neurophysiol 71: 317–329,
1994.
Rennie KJ and Streeter MA. Voltage-dependent currents in isolated vestibular afferent calyx terminals. J Neurophysiol 95: 26 –32, 2006.
Ricci AJ, Rennie KJ, Cochran SL, Kevetter GA, and Correia MJ. Vestibular type I and type II hair cells. I. Morphometric identification in the
pigeon and gerbil. Vestib Res 7: 393– 406, 1997.