Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Cl secretion in airway epithelial cells requires parallel activation of luminal Cl channels and basolateral K⁺ channels. Whereas Cl channels are required for exit of Cl to the luminal side of the epithelium, K⁺ channels are required for recycling of K⁺ via the basolateral membrane (1, 2). These K⁺ channels hyperpolarize the cell membrane voltage and thus maintain transepithelial electrolyte secretion. CF is one of the most common autosomal recessive diseases, characterized by a defect of cyclic adenosine monophosphate (cAMP)-dependent Cl conductance in the apical membrane of airway epithelia. Thus, identification of basolateral K⁺ channels involved in activation of cAMP-dependent Cl secretion could help to develop new pharmacologic strategies to circumvent the cystic fibrosis (CF) defect. So far, at least two different types of K⁺ channels have been detected in the basolateral membrane of secretory epithelial cells in previous studies. In colonic epithelia of mouse, rat, and human, Ca²⁺- and cAMP-dependent K⁺ conductances have been described (3). These K⁺ conductances are formed by separate individual K⁺channel entities. This has been demonstrated by the differential effects of K⁺ channel inhibitors on both cAMP- and Ca²⁺-activated K⁺ conductances. Typically, Ca²⁺-activated K⁺ conductance is inhibited by the compound clotrimazole, whereas cAMP-dependent K⁺ conductance is inhibited by the chromanol 293B (7, 8). Additional evidence for different populations of basolateral K⁺ channels came from patch-clamp experiments on isolated colonic crypt cells. According to these studies, Ca²⁺-activated K⁺ channels are of larger single-channel amplitude and can be resolved as single-channel currents, whereas cAMP-dependent K⁺ channels are of very small single-channel conductance and need to be analyzed by noise analysis (4 - 6, 9).

Both types of K⁺ conductance have been identified on the molecular level. The Ca²⁺-activated K⁺ channel was shown recently to be formed by a member of the hSK K⁺ channel family (10). This channel has been detected in human and rat colonic crypt cells (11). When expressed in Xenopus oocytes, a large Ca²⁺-activated K⁺ conductance is formed that is inhibited by clotrimazole and activated by the compound 1-EBIO (6, 7, 11, 12). The same compound, clotrimazole, also inhibits Ca²⁺-activated K⁺ channels and short-circuit currents in the colonic epithelium. The basolateral cAMP-activated K⁺ channel is blocked by the chromanol compound 293B with a concentration for half-maximal inhibition (IC₅₀) below 1 µM (8). The very same compound has been found to block K_VLQT1 K⁺ channels expressed in Xenopus oocytes (13, 14). K_VLQT1 (KCNQ1) is a voltage-gated K⁺ channel, cloned initially from heart cells where it is essential for repolarization of the heart action potential. When the channel is mutated this process is defective, as in certain forms of cardiac arrhythmia. Other members of this type of K⁺ channel were detected in the vestibular organ, brain, and sensory outer hair cells (15- 18). Cloning and expression of K_VLQT1 in colonic epithelial cells was shown in a recent study (19). These findings prompted us to assume a similar K⁺ channel in airway epithelial cells. We now deliver data indicating that in fact K_VLQT1 is forming the basolateral cAMP-activated K⁺ conductance in human airway epithelia. According to our data, parallel activation of K_VLQT1 and alternative Cl channels or residual CF transmembrane conductance regulator (CFTR) Cl channel activity may help to restore the secretory defect in CF airways.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Cell Culture

Cells from the immortalized human bronchial epithelial cell line 16HBE14_O(HBE) (kindly provided by Dr. D. Gruenert, University of California at San Francisco, San Francisco, CA) (20) were grown on plastic tissue-culture dishes and kept in an atmosphere of 5% CO₂/95% air in modified Eagle's medium containing (per liter) 5 mmol D-glucose, 20 mmol D-galactose, 2 mmol L-glutamine, 100 g fetal calf serum, 100 mg streptomycin, and 10,000 U penicillin.

Patch-Clamp Experiments

The culture dishes were mounted on the stage of an inverted microscope (Axiovert 10; Zeiss, Obernochen, Germany) and kept at 37°C. The bath perfusion rate was 20 ml/min, corresponding to a bath exchange rate of twice per second. The standard bath solution contained (in mmol/liter): NaCl 145, K₂HPO₄ 1.6, KH₂PO₄ 0.4, CaCl₂ 1.3, MgCl₂ 1.0, and D-glucose 5. The pH was adjusted to 7.4. A flowing KCl electrode served as a reference and appropriate corrections for liquid junction potentials were made. The fast whole-cell (WC) patch-clamp method was used (21). The patch pipettes had an input resistance between 2 and 4 M and were filled with a solution containing (in mmol/liter): KCl 30, K-gluconate 95, NaH₂PO₄ 1.2, Na₂HPO₄ 4.8, ethyleneglycol-bis- (-aminoethyl ether)-N,N'-tetraacetic acid 1, CaCl₂ 0.7, MgCl 1.3, D-glucose 5, adenosine triphosphate 1. The pH was adjusted to 7.2 and the Ca²⁺ activity of this solution was 0.1 µmol/liter. The access conductance (G_A) was measured by applying sinus command voltages (800 Hz) in each experiment and was between 30 and 120 nS. Appropriate corrections for the measurement of G_A were made to obtain accurate WC conductance readings. Currents and voltages were recorded with a patch-clamp amplifier (Fröbe/Busch, this institute) and continuously displayed by a pen recorder. The membrane voltage (V_m) of the cells was recorded continuously using the current clamp mode of the patch-clamp amplifier and the WC current (I) was measured at regular intervals by clamping the membrane voltage (V_c) to ± 30 mV in steps of 10 mV. The WC conductance (G_m) was calculated according to Ohm's law from the measured I and V_c.

Ussing Chamber Experiments

Freshly excised nasal tissues were obtained from 20 non-CF individuals after surgery for plastic reconstruction or sleep apnea syndrome (15 subjects) or nasal polyps (5 subjects) and from 10 CF patients after polypectomy (Professor R. Laszig, ENT Clinic, University Hospital Freiburg; and Professor G. Münker, ENT Clinic University Hospital Ludwigshafen). The tissues were immediately put into ice-cold buffer solution with the following composition (in mmol/liter): NaCl 127, KCl 5, Glucose 5, MgCl₂ 1, Na-Pyruvate 5, N-2-hydroxyethylpiperazine-N'-ethane sulfonic acid 10, and CaCl₂ 1.25; and albumin 10 g/liter. A thin layer of respiratory epithelium was dissected from the polyp and mounted into a modified Ussing chamber as described previously (3). The open tissue area of the chamber was minimized to 0.95 mm², allowing for stable measurements of small pieces of native respiratory tissue. The luminal and basolateral baths were perfused continuously at a rate of 15 ml/min (chamber volume, 2 ml) allowing us to examine the effect of K⁺ channel blockers on Cl secretion in the absence and presence of cAMP stimulation in a paired fashion. The bath solution had the following composition (in mmol/liter): NaCl 145, KH₂PO₄ 0.4, K₂HPO₄ 1.6, glucose 5, MgCl₂ 1, Ca-gluconat 1.3. The pH was adjusted to 7.4. Bath solutions were heated by water jackets and all experiments were carried out at 37°C. Ussing chamber measurements were carried out under open-circuit conditions. Transepithelial resistance (R_te) was determined by applying short (1 s) current pulses (I = 0.5 µA) and by recording the corresponding changes in transepithelial voltage (V_te), or V_te, as well as the basal V_te continuously. Values for the V_te were referred to the serosal side of the epithelium. The voltage deflections obtained with the empty chamber (V_te) were subtracted from those obtained in the presence of the mucosa. R_te was calculated according to Ohms law (R_te = V_te  V_te)/I). The equivalent short-circuit current (I_sc) was determined from V_te and R_te, i.e., I_sc = V_te/R_te. After mounting the tissues in the Ussing chamber, an equilibration period of 60 min was allowed for stabilization of basal V_te and R_te. Then amiloride (10 µmol/liter; luminal side) was applied to block electrogenic Na⁺ absorption. cAMP-dependent Cl secretion was activated by adding isobutylmethylxanthine (IBMX) and forskolin (100/1 µmol/ liter; both sides). The K⁺ channel blockers Ba²⁺ (5 mmol/liter) and the chromanol 293B (0.01 to 100 µmol/liter) were added to the basolateral side of the epithelium.

RNA Isolation and Reverse Transcriptase/Polymerase Chain Reaction

Total RNA was isolated from the mucosal layer of freshly excised non-CF and CF nasal polyps and from HBE cells using RNeasy spin columns (Quiagen, Hilden, Germany) as described previously and was reverse transcribed at 37°C for 1 h using random primer and reverse transcriptase (RT) (Superscript RT; Life Technologies, Inc., Karlsruhe, Germany). The size of the expected 738-base pair (bp) fragment of K_VLQT1 was amplified by polymerase chain reaction (PCR) using the sense primer 5'-TTCTGGATGGAGATCGTG-3' and antisense primer 5'-GCCTTCCGGATGTAGATC-3' (95°C for 30 s, 63°C for 1 min, and 72°C for 1 min; 35 cycles). PCR products were visualized by loading an 8-µl sample on a 0.9% agarose gel using a 123-bp marker as a standard. The PCR product was subcloned into pBluescript SK () vector and sequenced using Thermo Sequenase I (Pharmacia, Freiburg, Germany) and a 373A DNA sequencer (Applied Biosystem, Norwalk, CT).

Compounds and Statistics

Amiloride and IBMX were obtained from Sigma and Merck (Deisenhofen and Darmstadt, Germany, respectively). Forskolin and 293B were obtained from Hoechst (Frankfurt/Main, Germany). All chemicals used were of highest grade of purity available. From some individuals, transepithelial measurements were performed on more than one nasal tissue. When multiple tissue samples from one individual were studied by the same experimental protocol, data were averaged to obtain a single value for each individual subject. Data are shown as individual recordings or as means ± standard error of the mean; n = number of subjects for transepithelial measurements, or n = number of experiments for WC patch-clamp studies. Statistical analysis was performed using paired Student's t test. Data obtained from CF and non-CF tissues were compared by unpaired Student's t test. A P value < 0.05 was accepted to indicate statistical significance.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Expression of K_VLQT1 in Airway Epithelium

Expression of K_VLQT1 in respiratory epithelium was verified by isolation of RNA from mucosa of non-CF and CF nasal polyps and HBE cells. After reverse transcription and PCR, transcripts of K_VLQT1 were detected in freshly isolated non-CF and CF airway epithelium as well as in HBE cells (Figure 1). Sequencing of the PCR generated a 738-bp fragment that confirmed expression of human K_VLQT1.

View larger version (43K): [in this window] [in a new window]   Figure 1.   RT-PCR analysis of RNA obtained from freshly isolated CF and non-CF nasal airway epithelia and from HBE cells. A 738-bp fragment of human KVLQT1 was obtained after reverse transcription of total RNA (+) but not without RT (). Subsequent sequencing of the amplified fragment confirmed amplification of human KVLQT1.

Contribution of Basolateral cAMP-Dependent K⁺ Conductance to Ion Transport in Non-CF and CF Airways

Basal properties of non-CF and CF airways were assessed. In epithelia obtained from non-CF subjects I_sc was 41.5 ± 7.4 µA/cm² (n = 16). V_te and R_te were 0.8 ± 0.1 mV and 23.9 ± 3.6 cm², respectively. Amiloride (10 µmol/liter) added to the mucosal side of the epithelium reduced V_te and I_sc and increased R_te (Figures 2A and 2C). In CF, basal V_te and I_sc were significantly increased (2.3 ± 0.5 mV and 159.7 ± 31.1 µA/cm², respectively) and basal R_te was reduced (15.9 ± 2.5 cm²) when compared with non-CF subjects (n = 8). Amiloride-sensitive I_sc was significantly increased in nasal tissues from CF patients (Figures 2B and 2C). When tissues from non-CF subjects were stimulated by IBMX (100 µmol/liter) and forskolin (1 µmol/ liter), both enhancing intracellular cAMP, I_sc and V_te were enhanced, which was paralleled by a decrease in R_te. In CF subjects, no significant changes of V_te, R_te, or I_sc were observed upon cAMP-dependent stimulation, due to the lack of luminal CFTR-Cl channels (Figure 2C). These data demonstrate typical properties of CF and non-CF airways used in the present study.

View larger version (48K): [in this window] [in a new window]   View larger version (54K): [in this window] [in a new window]   View larger version (9K): [in this window] [in a new window]   Figure 2.   Effects of amiloride and IBMX/forskolin (IBMX/Fors) on Vte and Rte in human respiratory epithelium from non-CF (A) and CF (B) subjects. Rte was obtained from the Vte downward deflections obtained by pulsed current injection. (A) In non-CF subjects, lumen-negative Vte was reduced when amiloride (10 µmol/liter) was added to the luminal side of the epithelium. Subsequent stimulation with IBMX/forskolin (100/1 µmol/liter) induced a lumen-negative Vte, indicating activation of Cl secretion. (B) In CF subjects, lumen-negative Vte and the effect of amiloride were both increased compared with non-CF subjects. Subsequent addition of IBMX/forskolin failed to induce Cl secretion and thus had no effect on Vte. (C) Summary of equivalent Isc in non-CF and CF subjects under control conditions, in the presence of amiloride and IBMX/forskolin. *Statistical significance for the effects of amiloride and IBMX/forskolin (paired t test). §Statistical significance for basal properties and the effects of amiloride and IBMX/forskolin when compared with experiments obtained from non-CF tissues (unpaired t test). Number of subjects is in parentheses.

Contribution of basolateral K⁺ channels to cAMP-dependent Cl secretion in human airway epithelium was assessed by using the K⁺ channel blockers Ba²⁺ and the chromanol 293B, a specific inhibitor of cAMP-activated K_VLQT1 K⁺ channels (8, 13). In the presence of amiloride, non-CF tissues were stimulated with IBMX and forskolin (100/1 µmol/liter), which increased I_sc significantly from 10.9 ± 1.6 to 36.0 ± 7.3 µA/cm² (n = 10). Addition of 293B (10 µmol/liter) reduced cAMP-activated I_sc almost completely to 16.2 ± 3.2 µA/cm². When Ba²⁺ (5 mmol/ liter) was added in the presence of 293B, the remaining I_sc was further reduced to 1.3 ± 2.2 µA/cm² (n = 10) (Figures 3A and 3B). The inhibitory effect of 293B was concentration-dependent (Figure 3C) and an IC₅₀ of approximately 0.3 µmol/liter (n = 7) was obtained (Figure 3D). This value compares well with that obtained in previous studies for human and rat colon epithelium (3, 8). In CF tissue, 293B (10 µmol/liter) had no effect on I_sc in the presence of amiloride or after stimulation with IBMX/forskolin. Ba²⁺ (5 mmol/liter) reduced I_sc from 19.4 ± 4.6 to 7.9 ± 3.1 µA/cm² (n = 4) (Figure 3B).

View larger version (84K): [in this window] [in a new window]   View larger version (10K): [in this window] [in a new window]   View larger version (72K): [in this window] [in a new window]   View larger version (12K): [in this window] [in a new window]   Figure 3.   Effects of the chromanol 293B and Ba2+ on ion transport in respiratory epithelia from non-CF and CF subjects. (A) In non-CF tissue, in the presence of amiloride (10 µmol/liter) and IBMX/forskolin (100/1 µmol/liter), lumen-negative Vte was largely reduced by addition of 293B (10 µmol/liter). The remaining Vte was almost completely inhibited by Ba2+ (5 mmol/liter). (B) Summary of effects of 293B and Ba2+ on upper airway tissues from non-CF and CF individuals. (C) Original recording and (D) summary of the concentration dependency of the inhibitory effects of 293B obtained from respiratory epithelia from non-CF individuals. The IC50 was in the range of 0.3 µmol/liter. *Statistical significance for the effects of 293B and Ba2+ (paired t test). §Statistical significance for the effects of IBMX/forskolin and 293B when compared with experiments with tissues obtained from non-CF individuals (unpaired t test). Number of subjects is in parentheses.

We further examined the impact of cAMP-dependent stimulation on basolateral K⁺ conductance in strictly paired experiments. To that end, and after inhibition of the Na⁺ conductance by amiloride, 293B and Ba²⁺ were applied in the absence or presence of IBMX/forskolin. In non-CF tissue, 293B reduced I_sc slightly but significantly from 19.6 ± 2.9 to 14.5 ± 2.1 µA/cm² (n = 16) under basal conditions. The effect of 293B was completely reversible upon washout for 30 min. IBMX and forskolin increased I_sc from 17.1 ± 2.5 to 44.3 ± 6.1 µA/cm², which was reduced by 293B to 20.3 ± 2.9 µA/cm² (Figures 4A and 5A). Thus, 293B-sensitive I_sc was significantly enhanced after cAMP-dependent stimulation. Similar observations were made when the nonspecific K⁺ channel blocker Ba²⁺ (5 mmol/liter) was applied instead of 293B. In the absence of IBMX/forskolin, Ba²⁺ inhibited I_sc by 10.9 ± 2.2 µA/cm² (n = 10). After stimulation with IBMX and forskolin, Ba²⁺-sensitive I_sc was significantly increased to 19.3 ± 2.4 µA/cm², demonstrating activation of a cAMP-dependent K⁺ conductance (Figure 5B). In CF tissue, no significant changes were observed when 293B was added to respiratory tissues, either under control conditions or after cAMP-dependent stimulation (n = 8) (Figures 4B and 5A). As shown in Figures 3B and 5B, under control conditions the inhibitory effects of Ba²⁺ were comparable with that observed in non-CF tissues: I_sc = 16.4 ± 1.9 µA/cm² (n = 7). However, stimulation with IBMX and forskolin did not augment Ba²⁺-sensitive I_sc (I_sc = 15.4 ± 2.2 µA/cm²; n = 7) in CF tissue (Figure 5B). The data therefore clearly indicate activation of 293B-sensitive I_sc in the basolateral membrane of non-CF respiratory epithelia. Due to defective CFTR Cl channels, such an activation was not observed in epithelia from CF patients.

View larger version (52K): [in this window] [in a new window]   View larger version (37K): [in this window] [in a new window]   Figure 4.   Effects of 293B in the absence and presence of cAMP stimulation in nasal epithelia from non-CF (A) and CF (B) subjects. (A) A small but significant inhibition of lumen-negative Vte was observed in non-CF subjects under basal conditions. The effect was entirely reversible on washout for 30 min. Stimulation with IBMX and forskolin (IBMX/Fors) increased lumen-negative Vte. The effect of 293B (10 µmol/liter) was enhanced after cAMP-dependent stimulation of the tissue. (B) In CF subjects, IBMX/forskolin failed to induce Cl secretion and 293B had no effect on epithelial transport. Experiments were performed in the presence of 10 µmol/liter amiloride.

View larger version (14K): [in this window] [in a new window]   Figure 5.   Summary of the 293B-sensitive (A) and Ba2+-sensitive (B) Isc in nasal tissues obtained from non-CF and CF subjects in the absence (open bars) or presence (filled bars) of cAMP stimulation. (A) In the presence of amiloride (10 µmol/liter), 293B- and Ba2+-sensitive Isc were significantly enhanced after incubation with IBMX and forskolin. In CF subjects, 293B- and Ba2+-sensitive Isc were not affected by cAMP-dependent stimulation. *Statistical significance for the effects of 293B and Ba2+ (paired t test). Open square indicates statistical significance for the difference of the effects of 293B and Ba2+ in the absence and presence of cAMP stimulation (paired t test). §Statistical significance for the effects of 293B when compared with experiments with nasal tissues obtained from non-CF subjects (unpaired t test). Number of subjects is in parentheses.

293B-Sensitive K⁺ Conductance in HBE Cells

Additional patch-clamp experiments were performed on cultured non-CF HBE cells. The WC conductance was largely increased and the cell membrane potential was depolarized by stimulation of the cells with 10 µmol/liter forskolin (Figures 6A and 6B). When extracellular Cl concentration was reduced to 30 mmol/liter and was replaced by the impermeant anion gluconate, the WC conductance was inhibited and the membrane voltage depolarized, indicating activation of a WC Cl conductance (Figures 6C and 6D). Next, we examined the effects of 293B on HBE cells. As shown in Figure 7A, a small but significant fraction of the WC conductance was inhibited dose-dependently by 293B after stimulation of HBE cells by 10 µmol/liter forskolin, which was paralleled by depolarization of the membrane voltage (data not shown). For 293B, an IC₅₀ of about 6 µmol/ liter was detected. In paired experiments, the amount of WC conductance that was inhibited by 293B was significantly enhanced upon stimulation with forskolin (Figure 7A). These data indicate that 293B-sensitive K⁺ channels in HBE cells are activated by cAMP-dependent stimulation.

View larger version (23K): [in this window] [in a new window]   Figure 6.   Patch-clamp WC recordings in HBE cells. Continuous recording of (A) Gm and (B) membrane voltage (Vm) in HBE cells. The effects of partial replacement of extracellular Cl by both gluconate (30Cl) and 293B (10 µmol/liter) were rather weak under control conditions and were enhanced after stimulation of the tissue with forskolin (10 µmol/liter). (C) Summary of the Gms in the absence and presence of forskolin indicating activation of a WC Cl conductance. (D) Summary of the changes in Vm caused by forskolin and 30Cl. Forskolin depolarized Vm significantly. Depolarization of Vm was significant in the presence of forskolin, whereas no effects were observed under control conditions. *Statistical significance for the effects of 30Cl and forskolin, respectively. # Statistical significance when compared with experiments obtained under control conditions (paired t tests). Number of experiments is in parentheses.

View larger version (12K): [in this window] [in a new window]   Figure 7.   (A) Concentration dependency of the inhibitory effects of 293B on Gm in HBE cells. The IC50 of Gm was in the range of 6 µmol/liter. (B) Activation of 293B-sensitive K+ conductance by increase in intracellular cAMP. Summary of the Gms obtained in HBE cells in the presence and absence of forskolin (strictly paired experiments). *Statistical significance for the effects of 293B (paired t test). §Significant increase of Gm by forskolin (paired t test). #Statistical significance when compared with experiments obtained under control conditions (paired t test). Number of experiments is in parentheses.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Transepithelial secretion of fluid, mucins, and electrolytes is essential in the formation of airway surface liquid. Cl secretion has been shown to depend on parallel activation of luminal Cl channels and basolateral K⁺ channels in various epithelia. In this secretory model, activation of a basolateral K⁺ conductance is indispensable for supplying the driving force for Cl exit through the luminal membrane. The ion transport defect underlying CF airway disease is characterized by a defect of cAMP-mediated CFTR Cl channels in the apical membrane of airway epithelia. In this study we have examined the basolateral K⁺ conductance that is activated during cAMP-dependent stimulation in human airway epithelial cells. To be able to assess the impact of basolateral K⁺ channels on cAMP-mediated Cl secretion in native human respiratory tissue we made use of a perfused micro-Ussing chamber with an exposed area of less than 1 mm² (22). All measurements were performed under open circuit conditions, resembling the physiologic situation in vivo and allowing for stable measurements of V_te and R_te over several hours. Due to imperfect edge-sealing, the absolute magnitude of V_te detected ex vivo was decreased compared with values reported from previous in vivo studies (23). However, transepithelial ex vivo measurements allowed to quantify cAMP-dependent Cl secretion on the basis of equivalent I_sc, and continuous bath perfusion, enabled us to investigate the basolateral K⁺ conductance involved in this process. It is shown here that airway Cl secretion that is turned on by an increase of cAMP is sustained essentially by parallel activation of a basolateral K⁺ conductance. A previous study on murine nasal epithelia arrived at a similar conclusion (6). In this earlier report, basolateral application of the compound 293B, which inhibits K_VLQT1 type K⁺ channels, effectively blocked cAMP-activated I_sc in murine colonic epithelium but not in nasal epithelia. The authors suggested that basolateral cAMP-dependent K⁺ channels are different in both murine colonic and airway epithelia. In contrast to the measurements on mouse airways, the data obtained in the present study on native human respiratory tissue suggest that K_VLQT1 plays a major role in maintaining cAMP-dependent Cl secretion in human airways. In non-CF tissue, the chromanol 293B completely inhibited cAMP-stimulated I_sc and reduced total I_sc by about 60%. The observation that Ba²⁺ had an additional inhibitory effect on I_sc supports the concept of different types of basolateral K⁺ channels contributing to airway Cl secretion. A similar mechanism has been demonstrated for colonic epithelium of human, mouse, and rat (3, 4, 6, 24).

In the present experiments with CF respiratory tissues, activation of K_VLQT1 by IBMX and forskolin failed to induce Cl secretion. This is explained by the lack of luminal CFTR Cl channels. Therefore, 293B had no effect in CF tissues, although K_VLQT1 is expressed in human CF respiratory tissues as shown by RT-PCR analysis. Further, no evidence exists that would indicate that the level of expression of K_VLQT1 in CF airways is reduced. In a study with airways from CFTR (/) knockout mice, however, different results were obtained. Here, cAMP-dependent stimulation induced Cl secretion even in the airways of CF mice. This discrepancy could be explained by enhanced expression of an alternative Ca²⁺- dependent Cl conductance in the luminal membrane of mouse compared with human CF airways (6, 25).

In murine nasal epithelium of both wild-type (CFTR +/+) and CFTR (/) knockout mice the inhibitory compound 293B was largely ineffective in blocking cAMP-activated ion secretion (6). Further, in our experiments, the IC₅₀ values for 293B were ~ 10-fold higher in cultured HBE cells compared with native nasal tissues. Currently, there are no further data available that would explain the differences in 293B sensitivity obtained in native mouse and human airways and cultured airway cells. However, a likely explanation is that the chromanol sensitivity of K_VLQT1 is modulated by additional regulatory proteins such IsK or KCNE3 (26, 27) and that these proteins are expressed differently in native tissues from different species and cultured cells. KCNE3 has recently been demonstrated to colocalize with K_VLQT1 in colonic crypt cells and markedly change 293B sensitivity of the K⁺ currents (27).

An additional K⁺ conductance seems to exist in the basolateral membrane of airway epithelial cells that is blocked by Ba⁺ but not by 293B. Very similar observations have been made in human, mouse, and rat colonic epithelia (3, 4, 6). Although the results obtained from mouse tissue indicate the presence of Ca²⁺-activated K⁺ channels (6), there are currently no data available on human airways. The recently cloned Ca²⁺-activated human SK4 K⁺ channel would make a good candidate (10). This channel was also cloned recently from rat and was shown to be expressed in colonic epithelial cells. Further studies have to demonstrate whether this type of channel is also present in basolateral membranes of human airways. There is a large body of evidence that K_VLQT1 is forming the major component of basolateral cAMP-activated K⁺ channels in the colonic epithelium of rat, mouse, rabbit, and human (3, 5, 6, 19, 28). The present paper now indicates for the first time that a similar mechanism is present in human airways. The presence of K_VLQT1 in the basolateral membrane of human airways may challenge further research on activators of K_VLQT1/IsK K⁺ channels (29) that could be used for the treatment of patients with CF. These activators would be particularly beneficial in conjunction with compounds that stimulate residual Cl channel activity of mutant CFTR or that activate alternative Cl channels present in apical membranes of CF airways (30).