Introduction

Top Summary Introduction Materials & Methods Results Discussion References

The CB is an arterial chemoreceptor origin of ventilatory reflexes directed to maintain blood levels of O₂, CO₂, and H⁺ under physiological limits. Chemoreceptor cells are the CB elements that sense blood PO₂ and PCO₂/[H⁺], being activated when PO₂ decreases and PCO₂/[H⁺] increases. Activated chemoreceptor cells release neurotransmitters in amounts that are proportional to the decrease in PO₂ and to the increase in PCO₂/H⁺; parallel increases in the action potential frequency of the sensory nerve of the CB and in ventilation follow (Gonzalez et al., 1994).

The coupling of the decrease in PO₂ to the exocytotic machinery responsible for the release of neurotransmitters in chemoreceptor cells (i.e., the chemotransduction process) is incompletely understood, but it is well documented that plasma membrane mechanisms are involved. The presence in rabbit chemoreceptor cells of O₂-sensitive K⁺ channels (López-Barneo et al., 1988; López-López et al., 1989), whose open probability decreases as a function of PO₂ (Ganfornina and López-Barneo, 1991), led to the proposal that hypoxia could control the excitability of the cells, causing cell depolarization, activation of Na⁺ and Ca²⁺ channels, an increase in [Ca²⁺]_i, and release of neurotransmitters (Gonzalez et al., 1992; Gonzalez et al., 1994). O₂-sensitive K⁺ channels also have been found in neonatal (Peers, 1990; Buckler, 1997) and adult (Hatton et al., 1997; López-López et al., 1997) rat chemoreceptor cells, and a similar transduction sequence has been proposed. The detailed characterization of the different CB chemoreceptor cell preparations showed some discrepancies that have been reported to be mainly species related [see López-López and Peers (1997) for a review]. In particular, O₂-sensitive K⁺ channels seem to be different between rabbit and rat chemoreceptor cells. Although hypoxia inhibits a transient voltage-dependent I_K in rabbit cells (López-Barneo et al., 1988), the O₂-sensitive currents in rats are both a ChTX-sensitive Ca²⁺-dependent I_K (Peers, 1990; Wyatt and Peers, 1995; López-López et al., 1997) and a leak I_K (Buckler, 1997).

In patients with chronic respiratory failure, acute exacerbations brought about by respiratory infections may further impair their blood gas levels. Treatment with central respiratory stimulants provides limited clinical success and many side effects; long term oxygen therapy, being more successful, is both expensive and hard for patients. A third possibility to improve oxygenation to the blood is to stimulate the CBs with a drug such as almitrine bismesylate, which improves ventilation without central nervous system disturbances. The effect of almitrine enhancing alveolar ventilation through stimulation of the CB has been reported in several studies (Laubie and Schmitt, 1980; Bisgard, 1981; McQueen et al., 1989; Lahiri et al., 1989), but its mechanism of action remains unknown. It was reported recently that almitrine produces a long-lasting increase in the release of catecholamines from chemoreceptor cells in resting normoxic conditions and potentiates low PO₂-induced catecholamine release (Almaraz et al., 1992). It also has been shown that almitrine inhibits I_K in neonatal rat cells (Peers and O'Donnell, 1990), although this effect has not been characterized. These results led to the suggestion that the O₂-sensitive K⁺-channels could be possible targets for almitrine in chemoreceptor cells (Almaraz et al., 1992).

The aim of the current work was to characterize the effects of almitrine on the ionic currents of rabbit and rat chemoreceptor cells. After previous suggestions, our main thrust was the description of the effects of almitrine on O₂-sensitive K⁺ channels, but possible effects on other ionic currents of chemoreceptor cells also were studied to uncover additional potential targets for the drug. Due to the well-documented species-related differences in the properties of ionic currents from CB chemoreceptor cells (López-López and Peers, 1997), the effects of almitrine on I_Na, I_K, and I_Ca were studied in freshly or acutely cultured cells isolated from adult rabbits or rats using the whole-cell configuration of the patch-clamp technique. Almitrine did not modify voltage-dependent I_Na, I_K, or I_Ca from rabbit or rat cells. However, almitrine inhibited the Ca²⁺-dependent component of the I_K recorded from rat cells in the whole-cell configuration. This selective effect was characterized further at the single-channel level in membrane patches excised from rat chemoreceptor cells.

	Materials and Methods

Top Summary Introduction Materials & Methods Results Discussion References

Cell isolation and culture. Experiments were performed on cultured rat and rabbit CB chemoreceptor cells. Adult Wistar rats (3-4 months old) or adult New Zealand White rabbits (1.5-2 kg) were anesthetized with pentobarbital sodium (100 mg/kg administered intraperitoneally to the rats or 40 mg/kg administered through the lateral vein of the ear to the rabbits). After tracheostomy, the carotid artery bifurcations were removed, and the animals were killed by an intracardiac bolus injection of pentobarbital sodium. The CBs were cleaned of surrounding connective tissue and enzymatically dispersed as described previously (Pérez-García et al., 1992; López-López et al., 1997). Dispersed cells were plated onto small poly-L-lysine-coated coverslips and maintained in culture for up to 36 hr.

Electrophysiological recording. Ionic currents were recorded at room temperature (20-25°) using the whole-cell and inside-out modes of the patch-clamp techniques (Hamill et al., 1981). Whole-cell current recordings and data acquisition were made as described previously (López-López et al., 1997). Patch pipettes used for single-channel recordings were made from borosilicate glass (0.8 mm; World Precision Instruments, New Haven, CT) and double-pulled (Narishige PP-83) and heat-polished (Narishige MF-83) to resistances of 12-20 M when filled with the internal solution.

Analysis. Analysis of the data was performed with the CLAMPFIT and FETCHAN subroutines of the pCLAMP software. Single-channel amplitudes and open probabilities were measured from amplitude histograms generated with FETCHAN. The amplitude histograms consisted of 256 bins with each bin containing the number of sample points falling within the bin width. The amplitude of the single-channel currents was taken as the difference between the peaks for opened and closed currents levels. Because most of the patches had multiple channels, open probabilities were expressed as NPo, where N represents the number of single channels present in the patch, and Po represents the open probability of a single channel. NPo was calculated using the following expression (Kajioka et al., 1991): where A₀ is the area under the curve of the amplitude histogram corresponding to current in the closed state, and A₁ ... A_n represents the histogram areas reflecting the different open-state current levels for 1 to n channels present in the patch. Histogram parameters were obtained from multiple least-squares gaussian fits of the data using ORIGIN 4.0 software (MicroCal, Northampton, MA).

Solutions. The compositions of the bathing and pipette solutions for all recording conditions are given in Table 1. Gigaseals were formed in the standard extracellular solution (standard_e). Whole-cell I_K were recorded in this same extracellular solution with 125K_i in the pipette. When we studied I_Na or I_Ca, the 0K_i solution was used in the pipette; for I_Ca, the external solution was switched to 10 Ca_e (or 10 Ba_e), and in some experiments TTX was added to block I_Na.

Chemicals and drugs. All chemicals used in pipette and bath solutions were obtained from Sigma Chemical (St. Louis, MO). ChTX (Alomone Labs, Jerusalem, Israel) was used as described previously (López-López et al., 1997). Almitrine bismesylate [1-(4,6-diallylamino-2-triazinyl)-4-(bis-4, 4-fluorobenzydryl)piperazyne bismethane sulfonate; Vectarion, Servier International, Paris, France] was prepared in a 4.5 mM stock, with a solvent of a solution of 45 mM malic acid. At the highest concentration used (0.1 mM), malic acid alone had no effect on whole-cell ionic currents from chemoreceptor cells (data not shown)

	Results

Top Summary Introduction Materials & Methods Results Discussion References

Effects of almitrine on ionic currents from rat chemoreceptor cells. The effect of almitrine (10 µM) on whole-cell I_K from rat CB chemoreceptor cells is shown in Fig. 1. After establishment of the whole-cell configuration, I-V relationships for IK were obtained every minute, with the application of groups of 200-msec depolarizing steps from 60 to +100 mV in 10-mV steps. The holding potential was 60 mV. After several minutes under control conditions, almitrine (10 µM) was applied for 5 min, and the recording continued for 6 min after removal of the drug. Peak current values obtained at +20 and +90 mV in one representative cell are plotted against time in Fig. 1a. The slope of the usual decay of I_K amplitude at +20 mV (), due to washout of I_Ca (see López-López et al., 1997), sharply increased after the application of almitrine 10 µM (Fig. 1, hatched rectangle). The effect of almitrine on the current elicited at +90 mV was much less pronounced. In the 6-min period after the removal of the drug from the bathing solution, there was no recovery from inhibition. Fig. 1b shows sample records and the whole I-V relationships obtained before (time 2 min) and after (time 12 min) application of the drug. It is evident that the prominent hump in the I-V curve, which is due to the activation of Ca²⁺-dependent K⁺ channels (Peers, 1990; López-López et al. 1997), disappeared in the presence of almitrine. Results obtained in 11 cells with the same protocol were averaged and presented in Fig. 1c as percentage of inhibition produced by the application of 10 µM almitrine. In seven cells (), the inhibition was clearly voltage dependent, being maximal between the range of potentials of activation of Ca²⁺ channels (0-20 mV; p < 0.001 at 0 mV). In four cells (), there was almost no effect of almitrine in the entire range of tested voltages; two of these four cells did not exhibit a clear hump in their I-V relationships, but we do not have any explanation for the lack of effect of almitrine in the other two cells with a clear Ca²⁺-activated component. The voltage dependence of the inhibition strongly suggested that almitrine effectively inhibited IK_Ca in rat cells; indeed, the effect of the abolishment by almitrine of the hump of the I-V relationship is comparable to that observed with the IK_Ca blocker ChTX (Peers, 1990; López-López et al., 1997). To confirm the specificity of almitrine on IK_Ca, we studied the effect of the drug on I_K recorded in the presence of 20 nM ChTX. Fig. 2 shows the I_peak at +20 mV elicited by voltage ramps (0.023 mV/ms) from 60 to +100 mV applied every 30 sec. This ramp protocol was used instead of step depolarizations to enhance the Ca²⁺-dependent component of I_K (López-López et al., 1997). The application of ChTX produces a marked decrease of the current amplitude that is maximal at voltages between +10 and +20 mV (Fig. 2, inset). In the presence of ChTX, almitrine did not modify I_K, suggesting a common target (IK_Ca) for the inhibitory effect of the two drugs. The same lack of effect of almitrine on ChTX treatment was observed in an additional three cells. However, because IK_Ca decayed slowly along the experiments (Fig. 1a, initial 4 min and ; see also López-López et al., 1997) and because the washout of almitrine seemed to be very slow, it was very difficult to quantify the inhibition due to the drug; the two effects (current decay and current inhibition) combine in an apparently irreversible fashion. Moreover, the effect of almitrine can be due to a direct inhibition of Ca²⁺-activated K⁺ channels, an inhibition of the entry of Ca²⁺ through the Ca²⁺ channels, or both.

View larger version (29K): [in this window] [in a new window]   Fig. 1.   Effect of almitrine (10 µM) on whole-cell IK from rat chemoreceptor cells. Families of IK at different potentials were obtained every minute, and (a) peak currents measured at +20 and +90 mV are plotted against the time when the first pulse of the corresponding family was applied. Hatched area, application of almitrine. The averaged normalized decay of the current amplitude at +20 mV due to the washing-out of ICa for four control cells also is plotted in the figure. b, Current families and I-V relationships corresponding to the families recorded at times 2 and 12 min. c, Percentage of inhibition at potentials over 10 mV was computed according to the expression: Inhibition (%) = 100·(Icontrol  IAlmitrine)/Icontrol. A total of 11 cells were studied and grouped according the shape of the voltage dependence of the inhibition (, seven cells; , four cells). The effect of washing-out was not corrected. Solutions: standarde, 125Ki.

View larger version (15K): [in this window] [in a new window]   Fig. 2.   Effect of almitrine (10 µM) on whole-cell IK from rat chemoreceptor cells after ChTX application. Peak current amplitude at +20 mV obtained from 7-sec depolarizing ramps from 60 to +100 mV is plotted against time. Solid bars, periods of time in which the indicated drugs (ChTX or almitrine) were present in the bath solution. Inset, I-V relationships for the ramps taken at the times indicated (arrows 1, control; 2, 20 nm ChTX; 3, 20 nM ChTX plus 10 µM almitrine). Currents were not leak-subtracted. Solutions: standarde, 125Ki.

View larger version (29K): [in this window] [in a new window]   Fig. 3.   Effect of 10 µM almitrine on whole-cell ICa from rat chemoreceptor cells. a, Families of ICa at different potentials were obtained every 2 min, and peak currents measured at 20, 0, and +40 mV were plotted against the time when the first pulse of the corresponding family was applied. Hatched area, application of almitrine or Cd2+ 100 µM. Whole-cell currents were obtained with pulses to 40, 20, 0, +20, and +40 mV during the families 1-3. b, I-V relationships obtained during the application of almitrine or Cd2+ and normalized against the average of those obtained in control and recovery conditions. Values are mean ± standard error from five cells. Solutions: 10Bae, 0Ki.

Effects of almitrine on rat Ca²⁺-activated K⁺ channels. The inhibition of the Ca²⁺-dependent component of IK and the lack of effect of almitrine on I_Ca strongly suggested that Ca²⁺-activated K⁺ channels were in fact the targets for the action of the drug. This possibility was tested by studying the effect of almitrine on the activity of single Ca²⁺-activated K⁺ channels present in excised chemoreceptor cell membrane patches and recorded in the inside-out configuration of the patch-clamp technique. It is well known that Ca²⁺-activated K⁺ channels present in rat cells are mainly ChTX-sensitive BKs (Peers and Buckler, 1995; Wyatt and Peers, 1995; López-López et al., 1997). BK channels in the isolated patches were identified in this study on the basis of their voltage dependence, large conductance, and Ca²⁺ sensitivity (Fig. 4). In asymmetrical solutions (4.7 mM K⁺ at internal membrane face, 125 mM K⁺ at external membrane face), the I-V relationship of the isolated channels showed some inward rectification (Fig. 4a, ), which is well described by the Goldman-Hodgkin-Katz current equation. The extrapolated reversal potential was close to the K⁺ equilibrium potential, which was +84 mV in these experiments. When the bath solution was changed and single-channel currents were recorded under symmetrical conditions (125 mM K⁺ at both sides of the membrane), the rectification disappeared and the reversal potential shifted to 0 mV, which is as expected for a K⁺-selective channel (Fig. 4a, ). Fig. 4a also shows the averaged I-V relationships obtained with symmetrical high K⁺ conditions from different patches (). The average slope conductance under these conditions was 152 ± 13 pS (five cells).

View larger version (26K): [in this window] [in a new window]   Fig. 4.   High-conductance K+ channels in inside-out excised patches. Ca2+ dependence. a, Current amplitudes were measured in a single patch at different potentials in 125Ko/4.7Ki [solutions: standarde (bath)/0Ca (pipette)] and in symmetrical high K+ [solutions: 1 µM Ca (bath)/0Ca (pipette)]. Top, sample recordings at three different potentials in the two conditions. Solid line, closed level. Downward deflections, inward current. Upward deflections, outward current. Bottom, full I-V relationships under the two conditions. Dotted line, I-V relationship predicted by the Goldman-Hodgkin-Katz equation. Mean ± standard error values obtained in four different patches in symmetrical K+ also are shown (). Solid line, linear fit to average data. b, Ca2+ dependence of high-conductance K+ channels. Top, sample currents corresponding to 5-sec recordings obtained from a single excised patch (inside-out) with 0, 1, or 10 µM Ca2+ in the bath solution; solid line, closed level. The patch holding potential was +60 mV. The patch was held in each condition for >4 min, and that time of recording was used to calculate NPo for each condition according to the equation in Materials and Methods. NPo in this patch amounted to 0.03, 0.31, and 0.95 when recording in solutions 0Ca, 1µMCa, and 10µMCa, respectively. Mean NPo values obtained in the three conditions also are plotted. Each value represents the mean ± standard error of 4-12 patches.

View larger version (29K): [in this window] [in a new window]   Fig. 5.   Effect of almitrine on Ca2+-dependent K+ channel activity. NPo was measured every 20 sec and plotted against time in an excised patch with at least two channels present. Channel activity was recorded in symmetrical K+, with 1 or 10 µM Ca2+ in the bath solution as indicated. Almitrine (0.01 or 10 µM) was applied when marked. The effect of 0.1 mM malic acid also was tested.

View larger version (24K): [in this window] [in a new window]   Fig. 6.   Dose-response curve of Ca2+-dependent K+ channel activity in response to increasing doses of almitrine. a, Sample recordings of channel activity during 4 sec in the same excised patch in control conditions (symmetrical K+ and 1 µM Ca2+ in the bath), with 0.1 and 10 µM almitrine, and 10 min after removal of the drug from the bath solution. Solid line, closed state. Short marks on the left, two opening levels in this patch. All-points histograms obtained in each condition from 4 min of recording also are shown. Pooled data from several patches were expressed as NPo/NPocontrol and represented as a function of almitrine concentration. Numbers in parentheses, number of averaged data. Solid line, nonlinear fit to the sigmoidal equation described in Results. Solutions: 0 Ca (pipette)/1 µM Ca (bath).

View larger version (14K): [in this window] [in a new window]   Fig. 7.   Ca2+ dependence of the effect of 10 µM almitrine on BK channel activity recorded in excised membrane patches from rat CB chemoreceptor cells or GH3 cells at two different Ca2+ concentrations: 1 and 10 µM. The percent inhibition was calculated as 100·(NPocontrol  NPoAlmitrine)/NPocontrol, with NPocontrol being the average open probability before and after almitrine application. In all cases, almitrine was present in the bath for 4-6 min. Each bar, mean ± standard error of three to six patches. Solutions: 0 Ca (pipette)/1 µM Ca or 10 µM Ca (bath).

Effects of almitrine on ionic currents from rabbit chemoreceptor cells. The effect of almitrine (10 µM) on whole-cell voltage-dependent currents of rabbit CB chemoreceptor cells is shown in Fig. 8. Voltage-dependent I_K were studied in seven cells. A protocol similar to the one described above for I_K in rat cells was used, but current families were obtained every 2 min. For each cell, the I-V relationship was obtained by determining I_peak and the amplitude just before the end of the 200-msec pulses (I_ss). I-V relationships obtained before, during, and after the application of 10 µM almitrine were normalized to the peak current at +80 mV in control conditions (i.e., before the application of the drug), and the results obtained with the seven cells (mean ± standard error) are represented in Fig. 8a. In the presence of almitrine, there is a small reduction in the current amplitude at very depolarized values (>+40 mV). However, this reduction is due to the rundown of the currents along the experiment, and its time course is the same in almitrine-treated and untreated cells. Traces in Fig. 8a show the I_K elicited by depolarizing pulses to +40 mV before, during, and after the application of almitrine in one of the studied cells.

View larger version (24K): [in this window] [in a new window]   Fig. 8.   Effect of 10 µM almitrine on whole-cell currents from rabbit chemoreceptor cells. a, I-V relationships of IK measured at the peak (Ipeak) or at 200 msec (ISS) and obtained before, during, and after the application of 10 µM almitrine. Current amplitudes were normalized in respect to Ipeak at +80 mV under control conditions. Values are mean ± standard error for seven cells. Solutions: standarde/125Ki. Traces obtained in a single cell during a 200-msec step from 60 to +40 mV are shown. b, I-V relationships of ICa measured during the application of almitrine were normalized against the average of those obtained under control and recovery conditions to correct for rundown. Values are mean ± standard error for six cells. Solutions: 10Cae, 300 nM TTX/0Ki. Traces obtained in one cell in steps to +10 mV before, during, and after the application of 10 µM almitrine are shown. c, I-V relationships of INa measured under control conditions, during the application of almitrine, and after washing of the drug were normalized against the maximal current under control conditions. Values are mean ± standard error for four cells. Solutions: standarde, 100 µM CdCl2/0Ki. Traces, obtained in one cell with voltage steps to +10 mV are shown.

	Discussion

Top Summary Introduction Materials & Methods Results Discussion References

We examined the effect of almitrine on ionic currents from chemoreceptor cells from rat and rabbit carotid bodies. In whole-cell recordings, the only significant effect observed has been an inhibition of IK_Ca in rat cells. Almitrine does not affect IK_v in rabbit cells (Fig. 8), or in rat cells, as suggested for the voltage dependence of the drug effect (Fig. 1b) and for the lack of effect in the presence of ChTX (Fig. 2). Furthermore, the effect on IK_Ca is not due to an inhibition of I_Ca (Figs. 3 and 8) but to a direct action of the drug on high-conductance Ca²⁺-activated K⁺ channels (Figs. 5-7). The effect of almitrine on BK channels is fully reversible, but the washout of the drug is very slow. The apparent lack of reversibility in whole-cell recordings certainly is due to the overlapping of the slow recovery and the washout of IK_Ca. The effect of almitrine on IK_Ca from rabbit cells has not been studied because the Ca²⁺-dependent component of IK is present in a very variable amount and I_K is mainly a voltage-dependent current (Ureña et al., 1989; Pérez-García et al., 1992). In fact, the lack of a marked hump in the I-V relationships measured in the rabbit cells in which almitrine was tested (Fig. 8a) clearly suggests a minor presence of IK_Ca in the studied cells. Analogously, the effect of almitrine on I_Na was studied only in rabbit CB cells because only a small percentage of rat cells (<10%) exhibits I_Na (López-López et al., 1997). Taken together, our data show that in chemoreceptor cells, almitrine, at concentrations up to 10 µM, inhibits selectively BK channels.

The single-channel properties of BK currents in our preparation show several differences with a previous work in neonatal rat CB cells (Wyatt and Peers, 1995), including a smaller unitary conductance (152 versus 190 pS) and a higher open probability for a given intracellular Ca²⁺ concentration (NPo = 0.3 at 1 µM Ca²⁺ versus 0.04). These discrepancies could reflect differences in the developmental stage between both preparations, because the experimental conditions and recording solutions are quite similar.

The selective effect of almitrine on BK channels in chemoreceptor cells provides a molecular target that can contribute to understanding of the reported chemostimulant effects of the drug in several preparations. We know that the main function of BK in excitable cells is to contribute to action potential repolarization (Garcia et al., 1995; Sah, 1996), and thereby selective inhibition of this channel in spontaneously active cells increases action potential frequency. In addition, there is clear evidence for the contribution of BK to the maintenance of resting membrane potential in some preparations (Carl et al. 1996).

The results of electrophysiological studies have shown that most rat CB chemoreceptor cells possess whole-cell I_K with a ChTX-sensitive Ca²⁺-dependent (BK) component that represents a major percentage of the entire I_K at membrane voltages between 10 and +40 mV (Peers and Buckler, 1995; López-López et al., 1997; see Fig. 1). It is a well-established fact that BK currents in rat chemoreceptor cells are reversibly inhibited by low PO₂ and that this inhibition may play an important role in the modulation of the response of the cells to hypoxia (Peers and Buckler, 1995; Wyatt and Peers, 1995; López-López et al., 1997). However, because rat CB chemoreceptor cells in normoxic conditions do not present spontaneous activity (López-López and Peers, 1997), the functional significance of this inhibition is in dispute. Although some workers have shown that BK contributes significantly to the genesis and maintenance of resting membrane potential (Wyatt and Peers, 1995), others found that ChTX does not affect membrane potential (Buckler, 1997), implying that BK inhibition produced by low PO₂ cannot represent the trigger for the chemotransduction process. The chemostimulant action of almitrine in normoxic rats (Behm et al., 1993; Lagneaux, 1994), in light of the findings reported in the current work, could be accounted for if BK contributes to the genesis of membrane potential, but the action of almitrine potentiating hypoxic chemoreception (Lagneaux, 1994) can be satisfactorily explained regardless of whether BK participates in the genesis of membrane potential because rat chemoreceptor cells can generate action potentials during hypoxic stimulation (Peers and Buckler, 1995).

The data presented here indicate that almitrine behaves as a selective blocker of BK in rat chemoreceptor cells and that the inhibitory effect of almitrine is dependent on the intracellular Ca²⁺ levels, being less prominent at higher Ca²⁺ concentrations (Fig. 7). Although the mechanism of this blockade has not been characterized, it probably involves a direct interaction of almitrine with the channel, because the role of intracellular mediators can be excluded in the inside-out configuration. Regarding the tissue-specificity of the effect, we found that almitrine at 10 µM also inhibits BK in GH3 cells (Fig. 7), although a detailed characterization of this inhibition is lacking. Due to the variability of expression of BK in rabbit chemoreceptor cells (see above), we have not studied the effect of almitrine on this channel in this species, but our preliminary results in GH3 cells make conceivable that almitrine would have comparable effects. In the rabbit CB chemoreceptor cells, BK currents are not O₂ sensitive (Ganfornina and López-Barneo, 1991). However, contrary to those in rat, rabbit chemoreceptor cells generate action potentials both at rest and after hypoxic stimulation (López-López et al., 1989; Montoro et al., 1996), so BK currents, when present, should contribute to action potential repolarization, and their inhibition by almitrine would lead to increased Ca²⁺ entry and consequent activation of the cell. Moreover, considering that chemoreceptor cells are electrically coupled (Abudara and Eyzaguirre, 1994), cell activation could spread to adjacent cells and ultimately to the entire CB. This hypothesis is consistent with previous data showing that almitrine both promotes the release of neurotransmitters from rabbit CB and potentiates the secretory response induced by hypoxia (Almaraz et al., 1992).

In addition to its chemostimulant actions, almitrine improves ventilation/perfusion mismatching and increases the oxygenation of arterial blood through potentiation of the hypoxic pulmonary vasoconstriction (Chardon et al., 1980; Saadjian et al., 1994). This therapeutically important effect of almitrine also could be accounted for by the inhibition of BK because it is well documented that BK plays a very important role in regulation of pulmonary artery tone (Weir and Archer, 1995). Thus, the findings reported in the current work might represent the description of a common cellular mechanism generating the therapeutic effects of almitrine.

Based on the ubiquitous distribution of BK in the organism, it seems that almitrine would produce a wide spectrum of unwanted effects; however, this is not the case in laboratory animals or humans, indicating that at clinically useful doses, the function of most cells is not altered by almitrine. How this specificity is achieved remains unknown. It could be related to the tissue specificity of the properties of BK (Bolton and Beech, 1992; Tseng-Crank et al., 1994) or, more likely, to the conjunction of additional cell mechanisms that, as targets for almitrine, amplify the effects of BK inhibition. In fact, almitrine produces other cellular effects (Leverve et al., 1994) that might contribute to sharpen the specificity of almitrine actions at the level of CB chemoreceptors and pulmonary artery smooth muscle cells, minimizing simultaneously its actions in other tissues.

In conclusion, we demonstrate that almitrine at therapeutically useful doses inhibits the O₂-sensitive high-conductance Ca²⁺-dependent K⁺ channel in rat CB chemoreceptor cells without affecting other ionic currents. This effect of almitrine could represent an important mechanism underlying the chemostimulant action of almitrine.