|Home | About | Journals | Submit | Contact Us | Français|
Motility of outer hair cells underlies the cochlear amplifier, which is critical for the ear’s sensitivity and fine tuning. Of the two motile mechanisms present in these cells, electromotility at the lateral wall depends on the receptor potential and thus depends on currents through the cell body. We found that, in the guinea pig cochlea, basal turn outer hair cells have a fast-activating ion current (τ < 0.3ms at 23°C), which is absent in apical turn cells. This finding is consistent with our previous theoretical analysis that a fast-activating potassium current is required only in the basal turn to counteract the capacitive current and thereby to enhance the effectiveness of electromotility. Thus our finding is consistent with the functional significance of electromotility. We conjecture therefore that the current reduces the capacitance of the outer hair cell in order to increase hearing bandwidth.
High sensitivity and sharp tuning of the ear requires that outer hair cells (OHCs) generate a force that counteracts viscous damping in the cochlea . Such force must be generated cycle-by-cycle at auditory frequencies. There are two elements in the OHC that are capable of high frequency force generation. One is located in the hair bundle [2, 3] and the other at the lateral membrane of the cylindrical cell body [4-6]. Force generation by the hair bundle is common in frogs  and turtles . It can be fast enough to operate cycle-by-cycle at high frequencies  and does not directly depend on the receptor potential. Force generation in the lateral membrane is based on a piezoelectric membrane motor  involving prestin . Although this motile mechanism, which is referred to as electromotility, is intrinsically fast, it is driven by the receptor potential. Since the receptor potential is strongly attenuated by capacitive current, so must be electromotility. Thus the effectiveness of electromotility could be questioned (the RC time constant problem). However, the poor hearing of prestin-knockout animals inidicates the sifnificance of electromotility . Additionally there is evidence for electromotility operating as a high-frequency amplifier rather than as a low-frequency feedback for optimizing the operating point of the hair bundle motor .
A key to the problem could be ion currents because fast voltage-gated channels could function as inductance  and could improve their high frequency performance. Previous studies showed, however, that currents in OHCs activate [14-16] much more slowly than do those in inner hair cells  (τ >10 ms versus 2ms at room temperature). These reports also show that OHCs’ conductance does not activate sharply near to the cells’ resting membrane potential. This difference between the two kinds of hair cell conductance is paradoxical if OHCs must function as feedback motors driven by the receptor potential  to enhance the sensitivity and frequency selectivity of the ear . Why are OHCs devoid of fast potassium currents?
To address this issue, we previously compared viscous drag force to the force generated by electromotility of the lateral cell membrane. We found that viscous drag can effectively be counteracted by receptor potential driven electromotility for frequencies below about 10 kHz . This analysis suggested that OHCs in the basal turn of the guinea pig cochlea, with characteristic frequencies between 7 and 40 kHz , may have fast potassium currents.
Here we report fast K+ currents in basal turn OHCs. Because these cells are very difficult to maintain in our experimental chamber, the data obtained remain somewhat preliminary.
Bullas were obtained after decapitation of anesthetized guinea pigs in accordance with animal protocol 1061-02 approved by NINDS/NIDCD. Isolated short OHCs were then obtained by dissociation of the organ of Corti. The extracellular medium contained 135 mM NaCl, 4 mM KCl, 2 mM MgCl2,1.5 mM CaCl2, 5 mM HEPES. Its osmolarity was adjusted with glucose to 290-300 mOsm/kg and pH set to 7.4. While not used for voltage-clamping, low Ca2+ external solution was used to assist in cell dissociation. This was similar to the above but instead contained 4.6 mM CaCl2 and 5 mM EGTA, leading to a free calcium concentration of approximately 1 μM. Some of the partial cell dissociations used neutral protease (dispase, Worthington Biochemical). The calculated potassium reversal potential of -90 mV was used throughout.
Whole-cell voltage-clamp experiments on short OHCs were performed at 23°C. Patch pipettes were made with a pipette puller (Model 81, Sutter Instruments, Novato, CA). Access resistances of the pipettes were between 4 to 8 MΩ when filled with internal medium. Internal solution was 15 mM HEPES, 140 mM KCl, 1 mM MgCl2, 4 mM CaCl2, and 10 mM EGTA,leading to a calculated free calcium concentration between 50-100 nM. A Patch amplifier (Axopatch 200B, Axon Instruments, Union City, CA) was used in whole-cell voltage-clamp mode and was sent a train of voltage pulse commands through an ITC-16 interface (Instrutech, Great Neck, NY). It was driven by an Igor Pro program (Wave Metrics, Lake Oswego, OR) with a data acquisition module created by R. J. Bookman’s laboratory at the University of Miami (http://chroma.med.miami.edu/cap/).
To characterize the OHC’s conductance we fit our cell conductance to the sum of an Ohmic conductance and a voltage-dependent conductance that gated according to a first-order Boltzmann function g0 + g/(1 + exp[-(Vm - V1/2)/s]), with Ohmic conductance g0, gated conductance g, membrane potential Vm, half-activation potential V1/2, and voltage sensitivity s.
The following are the specific experimental conditions for the 6 cells shown in the figures: Cell in Fig. 1a held at -75 mV, capacitance 41 pF uncompensated, access resistance of 3.7 MΩ, voltage-insensitive conductance 2.8 nS, no protease used for dissociation. Cell in Fig. 1b held at -72 mV by -0.1 nA, capacitance 20 pF uncompensated, access resistance 5 MΩ, voltage-insensitive conductance 30 nS, no protease used for dissociation. Cell in Fig. 1c held at -70 mV by +0.4 nA, capacitance 9 pF 95% compensated, access resistance 6 MΩ, voltage-insensitive conductance 30 nS, partially dissociated by .1 mg/ml dispase in 1 μM free calcium saline for 5 minutes, 85 mM of the external Na+ replaced by n-methyl-gluconate (NMDG). Cell in Fig. 2a held at -67 mV by -0.7 nA, capacitance 30 pF uncompensated, access resistance 8 MΩ, voltage-insensitive conductance 63 nS, no protease used for dissociation. Cell in Fig. 2c was held at -69 mV by -1.7 nA, 19 pF of uncompensated capacitance, access resistance 6 MΩ, voltage-insensitive conductance 44 nS, no protease used for dissociation. Cell in Fig. 2d was held at -60 mV by a -1.5 nA, access resistance 7 MΩ, 12 pF of capacitance 95% compensated, voltage-insensitive conductance 41 nS, partially dissociated by 0.1 mg/ml dispase in saline containing 20 nM free calcium for 6 minutes, 85 mM of the external Na+ replaced by NMDG.
Currents elicited from OHCs by voltage pulses are highly dependent on the turn of the cochlea from which the cell was isolated. A slow potassium current, called IK, with a half-activation voltage near -25 mV is known to be more prominent in apical turn cells, while a second slow potassium current IK,n with a half-activation near -90 mV is known to prevail in the mid to basal turns [14-16]. Fig. 1a shows current from an 80 μm-long apical cell with a slow-activating current consistent with its being IK, and having an activation time constant of approximately 15 ms. The instantaneous current in the cell is linear (Fig. 1d,e). A 30 μm long OHC from the apical end of basal turn also shows a slow-activating current together with a fast-activating component (Fig. 1b). Very short, 25 μm long basal coil cells show a prominent fast outward current (Figs. (Figs.1c,1c, 2a, 2d) that is not seen in longer OHCs, in addition to a slow current similar to IK,n. The activation range of this fast current is 40-50 mV more negative than that for IK and is in the vicinity of the cell’s resting potential.
The gating of this current was too fast for our whole-cell voltage clamp. We instead determined an upper bound for its activation time constant. One major problem is the existence of voltage-dependent membrane capacitance [21-23] due to the OHC membrane motor. This limits our electronic capacitance compensation to within a narrow voltage range. As a result, the current due to any fast conductance becomes contaminated with a capacitive transient. Therefore we obtained current-voltage plots typically 1-2 and 3-4 ms after the voltage step. Voltage clamp time resolution is determined by the product of a 5 MΩ access resistance and a cell capacitance typically around 15 pF, giving an RC time constant of 0.075 ms. Thus a reasonable upper bound on the current’s rise time would be several time constants, or about 0.3 ms.
This nonlinear current in short OHCs is quite labile. It fades in 20 minutes after the cell’s isolation. It is particularly susceptible to trypsin. This susceptibility likely explains its absence from an earlier report . We found that partial dissociation by neutral protease (dispase, Worthington Biochemical) in a low calcium external medium retains the current. The current recorded from those cells is consistent with that obtained from cells prepared without any protease (Fig. 2a, 2c). Blockage by TEA (Fig. 2c) demonstrates that it is a K+ current. On several occasions we observed that short OHCs had both fast inward and outward rectifying currents (Fig. 2d), usually with the inward currents having the weaker rectification (Fig. 1d) and similar to the current seen in an situ report .
The nonlinearity of the current elicited by voltage pulses shows an e-fold conductance increase for a voltage change (s value) less than about 10 mV. A nonlinear K+ current with an s value less than 25 mV cannot be due to an electrolyte asymmetry across an open pore (Goldman-Hodgkin-Katz nonlinearity) . Thus the current we observe cannot be as a result of an open pore, but is due either to activation or unblocking.
What is the role of this fast current? It has long been recognized that a voltage-activated K+ channel can act as an inductance within a narrow frequency band near its activation rate . Activation of such a current would lag membrane potential by 90°, and thus oppose capacitive current, which is advanced by 90°. Since the major part of transducer current from the hair bundle is spent in alternately charging and discharging cell capacitance, such a fast-activating current would reduce this loss and enhance the receptor potential. To be effective in an OHC operating at 10 kHz, the current would require a gating time constant of about 30 μs at body temperature . Since at 37°C many ion channels gate about 3 to 4 fold faster , we would then expect a time constant of about 0.1 ms at room temperature. Our experimental time resolution shows that the current gates in less than 0.3 ms.
The attenuation of the receptor potential by membrane capacitance is not limited to OHCs in the basal turn. However, nonlinear currents in apical cells are too slow to reduce their capacitive current. Why are fast currents absent in apical cells? Assuming that OHCs supply energy to a local mechanical resonance in the cochlea, we previously calculated that even without a fast current OHCs would nevertheless be able to counteract viscous drag up to about 10 kHz, near the low-frequency end of the basal turn . Thus only cells in the basal turn would require a fast channel.
We observe that there is a fast potassium current localized exclusively to the cochlea’s basal turn and that it has a sharply nonlinear voltage response near to the cell’s resting potential. These observations are consistent with our conjecture that the current contributes to the OHC’s receptor potential, supporting the cell’s piezoelectric motor at high frequencies. However, the effectiveness of this current depends on the rise time and detailed temporal structure. Thus, the demonstration of this functional role awaits further studies.
Finally, it is likely that the presence of this conductance in the basal turn cells makes these cells labile. We speculate therefore that the presence of this conductance is a factor that contributes to high frequency hearing loss, associated with the loss of outer hair cells [26, 27].
This research was supported by the Intramural Research Program of the NIDCD, NIH.