INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

CHLORIDE CHANNELS PLAY AN important role in volume regulation and electrolyte and water balance in a variety of epithelial cells. Several Cl channels have been identified on either the apical or basolateral membrane of epithelial cells, such as the adenosine 3',5'-cyclic monophosphate-sensitive cystic fibrosis transmembrane conductance regulator (CFTR; Refs. 11, 13) and the ClC family of Cl channels (20). Cl transport can also be activated by Ca²⁺, which in some cases is attributed to activation of other Ca²⁺-sensitive processes, such as opening of Ca²⁺-activated K⁺ channels, thereby changing the driving force for Cl secretion (10). Bona fide Ca²⁺- and Ca²⁺/calmodulin-dependent protein kinase II (CaMKII)-sensitive Cl channels have been identified in a variety of epithelia, including colonic cells (5, 24, 39), exocrine gland cells (1, 15), biliary and renal epithelia (22, 30), and airway epithelia (4, 34).

Several lines of evidence have suggested that the Ca²⁺-activated Cl channel (CaCC) pathway may be a promising alternate therapeutic target in cystic fibrosis (CF) to compensate for defective Cl transport via CFTR. Studies in freshly excised and cultured CF airway epithelia and in CF sweat glands have shown that the Ca²⁺-sensitive Cl conductance functions normally, whereas CFTR channel activity is compromised or absent (26, 33, 34). Furthermore, the expression of a CaCC seems to be an important determinant of the severity of organ-level disease in the CF knockout mouse model (6, 21, 38). Ca²⁺ ionophores have also been shown to activate a Cl conductance in CF airway epithelial cells (17), and a Ca²⁺-sensitive Cl conductance is upregulated in the nasal mucosa of CF mice (16, 32).

We have previously reported the isolation of a Cl channel protein from bovine tracheal epithelia (27, 28). When incorporated into planar lipid bilayers, the protein formed a small (25-30 pS) linear (under symmetrical conditions), 4,4'-diisothiocyanostilbene-2,2'-disulfonic acid (DIDS)-sensitive, anion-selective channel that was also sensitive to reduction by dithiothreitol (DTT) (29) and could be activated by Ca²⁺ and phosphorylation by CaMKII (14). The native protein, which migrates with a relative molecular mass of 140 kDa on nonreducing sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis (PAGE), migrated at 38 kDa when separated under reducing (50 mM DTT) conditions. This shift in relative molecular mass was paralleled by a loss of all channel activity in the presence of DTT (29). We have recently reported the cloning of a Ca²⁺-sensitive Cl channel from a bovine tracheal cDNA expression library (8). The cloned channel was inhibited by DIDS and DTT, was insensitive to niflumic acid, had a linear current-voltage (I-V) plot under symmetrical conditions, and could be activated by Ca²⁺ and CaMKII (8, 19). The sequence of the cloned channel also contained several consensus sites for phosphorylation by CaMKII, protein kinase C (PKC), and for N-linked glycosylation.

However, a major difference between the cloned and native CaCCs was their relative molecular masses. The primary transcript of the cloned CaCC codes for a protein that migrates at 100 kDa under both reduced and nonreducing conditions (in the absence of glycosylation), significantly larger than the smallest polypeptide component of the native channel, which migrates at 38 kDa under reducing conditions. This may reflect posttranslational processing of the primary translated product or that the native and cloned CaCC polypeptides are unrelated although functionally similar. To test the hypothesis that a much smaller form of the primary translated product could form a viable channel, we have generated a mutant CaCC (CaCCX) truncated at both the NH₂ and COOH termini. The predicted molecular mass of the protein encoded by CaCCX is 42.5 kDa. Therefore, the purpose of the present study was to compare the electrophysiological and pharmacological properties of the wt (wild-type) CaCC and CaCCX constructs functionally expressed in Xenopus oocytes.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Construction and in vitro translation of a truncated CaCC. We designed two primers to generate the truncated CaCC construct, CaCCX, in a polymerase chain reaction (PCR). Both 5' and 3' primers included a Bgl II site at the 5' end. The 5' primer (5'-GAAGATCTTCACCATGGATGTAATCATG-3') annealed to the wtCaCC cRNA sequence between bases 819 and 831; translation would be predicted to initiate at Met-277. The 3' antisense primer (5'-GAAGATCTTCTTATGAATAGATGCCATC-3') annealed to the CaCC coding region between bases 1984 and 1998, converting Arg-667 to a premature stop codon. The predicted length of the PCR product was 1179 base pairs, coding for a polypeptide of 390 amino acids. This PCR-generated fragment preserved the four predicted transmembrane regions of the wtCaCC and consensus sites for N-linked glycosylation (6 sites), CaMKII phosphorylation (3 sites), PKC phosphorylation (4 sites), and tyrosine kinase phosphorylation (1 site). Using the wtCaCC as the template in a PCR reaction with the above primers, we generated a product of the correct size under the following conditions: 94°C for 3 min × 1 cycle; 94°C for 1 min, 58°C for 2 min, 72°C for 3 min × 30 cycles; and 72°C for 15 min × 1 cycle, using Pfu thermostable polymerase (Stratagene). The product was digested with Bgl II, phosphorylated, ligated into a modified pGEM 11 vector (a gift of Dr. D. A. Melton, Harvard University, Cambridge, MA), and transformed into XL1-Blue. Positives were selected on the basis of restriction mapping (Bgl II, BamH I, Xba I, and Xho I).

Oocyte expression. Toads were obtained from Xenopus I (Ann Arbor, MI) and were kept in circulating dechlorinated tap water at 18°C. Toads were fed two times weekly with beef liver. The ovarian tissue was removed from toads under hypothermal anesthesia through a small incision in the lower abdomen as previously described (12). Oocytes at maturation stages V and VI were carefully defolliculated by hand in Ca²⁺-free OR-2 solution [in mM: 85.2 NaCl, 2.5 KCl, 10 N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid (HEPES), and 1.0 Na₂HPO₄, pH 7.5] and maintained in OR-2 medium (Ca²⁺-free OR-2, 1.0 CaCl₂, pH 7.5). Both OR-2 media contained 0.5% streptomycin (Sigma). Oocytes were incubated in OR-2 for several hours or overnight before cRNA injection. A Nanoject (Drummond Science, Broomall, PA) was used for injection of cRNA or ribonuclease-free water (control). The optimized concentrations (5-500 ng) of cRNA in 50 nl were injected into oocytes in Ca²⁺-free OR-2 solution (310 mosM) and then transferred to normal OR-2 medium; 5% heat-inactivated horse serum (Sigma) was added to OR-2 culture medium to improve the expression efficiency and cell viability (25).

Electrophysiological measurements. After 48-h incubation in OR-2 at 18°C, whole cell currents were recorded from the oocytes following equilibration at room temperature (at least 10 min at 20-25°C) and analyzed using pCLAMP version 5.5 software (Axon Instruments, Foster City, CA). Voltage-clamp potentials were evoked using a TEA-200 voltage clamp (Dagan, Minneapolis, MN) controlled by a personal computer connected via a TL-1 interface (Axon Instruments). The injected oocyte was placed in a small chamber (1 ml) and perfused with ND-96 medium (in mM: 96 NaCl, 1 MgCl₂, 2 KCl, 5.0 BaCl₂, 5.0 CaCl₂, and 5 HEPES, pH 7.4) at a flow rate of 1.5-2 ml/min for at least 5 min before recording. Microelectrodes filled with 3 M KCl had a resistance of 0.5-3.0 M. The bath was clamped by two silver-plated wires through 3% agar bridges in 3 M KCl. The oocytes were clamped at a holding potential of 60 mV. Data filtered at 0.5-1 kHz were digitized and stored on disk for off-line analysis. Test voltages were stepped from the holding potential to 100 through +80 mV in 20-mV increments for 1 s. The currents at 80 and +80 mV were monitored at 30-s intervals. An average of samples from 500 to 700 ms of each episode was used to plot I-V curves and to calculate the effect of inhibitors and regulatory agents. Linear components of capacitance and leak currents were not subtracted.

Solutions and chemicals. Stock solutions of 1 M DTT in H₂O, 10 mM phorbol 12-myristate 13-acetate (PMA; Calbiochem, La Jolla, CA) in dimethyl sulfoxide (DMSO), and 200 mM DIDS (Sigma) in ethanol or freshly prepared in H₂O, 1 mM A-23187, or 1 mM ionomycin (both Calbiochem) in DMSO were stored at 0°C in light-protected vials. The working concentrations were freshly prepared in the perfusing media for each experiment. Niflumic acid (Sigma; 100 mM stock in DMSO) was diluted to the final concentration in the perfusing medium immediately before use. The PKC inhibitor, chelerythrine chloride (10 mM stock in DMSO; Research Biochemicals International, Natick, MA), was used by pretreating the cells for 20 min at a final concentration of 10 µM in ND-96 medium before addition of the PKC activator, PMA. All other reagents not listed above were obtained either from Sigma (St. Louis, MO) or from Fisher.

Statistics. Results are expressed as means ± SE, where n is the number of oocytes. Student's t-test was performed to calculate the significant differences of the recordings before and after each drug application. Curve fitting of the concentration-response relationship of PMA was fit with the following equation where I_max stands for the current at the highest concentration of PMA, I_min is the current in the absence of PMA, [PMA] is the concentration applied, EC₅₀ represents the required concentration of PMA to activate the half-maximal conductance, and n is the slope of the fitting curve. Due to the variability of oocyte current expression, the current amplitudes of CaCC and CaCCX were normalized to the I_max of the outward currents (at +80 mV, for Cl influx) or to the I_max of the inward currents (at 100 mV, for Cl efflux).

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

In vitro translation of CaCCX. The full-length open reading frames of the CaCC and CaCCX predict that these polypeptides should migrate (in the absence of cotranslational glycosylation) at ~100 and 42.5 kDa, respectively. As shown in Fig. 1, in vitro translation of the two cDNAs resulted in polypeptides that migrated at 102 and 45 kDa, consistent with their predicted molecular masses under nonreducing SDS-PAGE.

View larger version (31K): [in this window] [in a new window]   Fig. 1.   In vitro translation of wild-type (wt; full length) Ca2+-activated Cl channel (CaCC) and truncated CaCC mutant construct (CaCCX). Appropriate cDNAs were transcribed and translated in the presence of [35S]methionine as described in MATERIALS AND METHODS. Translated wtCaCC migrated with a relative molecular mass of 102, whereas the truncated CaCCX polypeptide migrated with a relative molecular mass of 45, consistent with the predicted values of 100 and 43 kDa, respectively.

Niflumic acid blocks the endogenous CaCC. Oocytes of the South African clawed toad Xenopus laevis have been widely used for Cl channel expression studies. There are several endogenous Cl channels in the oocytes, including a CaCC and a Ca²⁺-inactivated Cl channel. The former is activated by increased intracellular Ca²⁺ (23), and the latter is activated by decreasing extracellular Ca²⁺ (35). However, both channels can be inhibited by niflumic acid and flufenamic acid (37). In the presence of 2 µM ionomycin (Fig. 2A), the current of H₂O-injected oocytes at +80 mV was 653 ± 40 nA compared with a basal current of 420 ± 16 nA (n = 4; control). After the addition of 100 µM niflumic acid, the current returned to the control level (481 ± 26 nA). The same pharmacological effects of Ca²⁺ ionophores and niflumic acid were observed at a membrane potential of 80 mV (Fig. 2B). Figure 2C shows similar effects on the endogenous CaCC activated by a second Ca²⁺ ionophore, A-23187 (2 µM). We have previously demonstrated that niflumic acid did not affect the cloned CaCC expressed in oocytes (8). Thus all of the experiments on CaCC and CaCCX were done with 100 µM niflumic acid in the perfusing solutions to eliminate any endogenous CaCC from consideration.

View larger version (27K): [in this window] [in a new window]   Fig. 2.   Effects of Ca2+ ionophores and niflumic acid (NFA) on the endogenous CaCC in the membrane of Xenopus oocytes. A: after treatment of H2O-injected oocytes with 2 µM ionomycin (IONO; 5 mM extracellular CaCl2), oocyte whole cell currents increased above that seen in control (CTL) and are inhibited by 100 µM niflumic acid (n = 4). Currents recorded at a membrane potential of 80 mV also reveal activation by ionomycin and blockade by niflumic acid (B). C: with 2 µM A-23187, a second Ca2+ ionophore, the endogenous CaCC is activated and is also sensitive to 100 µM niflumic acid. To compare the endogenous CaCC with the expressed CaCC, y-axis scales are same as those in Fig. 5. Inset: the same plot at expanded scale. Results in A and B are ±SE.

Expression of wtCaCC and CaCCX in Xenopus oocytes. The whole cell currents of oocytes expressing either the wtCaCC or CaCCX were measured after at least a 48-h incubation following cRNA injection. Only those oocytes injected with CaCC or CaCCX that had a basal conductance of at least 1 µA at +80 mV were identified as functionally expressing cells and selected for further experiments. The background currents of cRNA-injected oocytes were much higher than those of H₂O-injected cells during perfusion with ND-96 medium. Figure 3 shows sample records for wtCaCC (left) and CaCCX (right) currents. The whole cell currents in Xenopus oocytes injected with wtCaCC and CaCCX cRNA at +80 mV were 1,972 ± 282 (n = 7) and 1,783 ± 297 nA (n = 20), respectively, markedly higher (P < 0.001) than those of H₂O-injected oocytes (124 ± 34 nA, n = 7). The current amplitudes at 80 mV for oocytes injected with H₂O, wtCaCC cRNA, or CaCCX cRNA averaged 40 ± 2, 399 ± 30, and 466 ± 38 nA, respectively. These results are also similar to those previously reported for H₂O-injected and wtCaCC cRNA-injected oocytes (8).

View larger version (56K): [in this window] [in a new window]   Fig. 3.   Representative pharmacological responses recorded from Xenopus oocytes expressing wtCaCC or CaCCX. Left: traces recorded from 1 oocyte injected with 25 ng wtCaCC cRNA. Right: records from 1 oocyte injected with 25 ng CaCCX cRNA. Cells were held at 60 mV with a 20-mV step test potentials to +80 mV from 100 mV (see MATERIALS AND METHODS). Perfusing medium was ND-96 plus 100 µM niflumic acid. Ionomycin-sensitive and dithiothreitol (DTT)-sensitive currents are the differences before and after application of 2 µM ionomycin or 2 mM DTT.

Extracellular Cl sensitivity. For oocytes injected with wtCaCC cRNA or CaCCX cRNA, large currents were recorded at the test membrane potential of 100 mV. To ensure that these currents were due to a Cl conductance, sodium gluconate was used to replace sodium chloride isosmotically in the superfusing solutions. Three Cl concentrations, 114, 50, and 6 mM, were used to assess the Cl sensitivity of the current (Fig. 4). The current for the wtCaCC at 100 mV in the presence of 114 mM Cl (n = 6) was 1,924 ± 108 nA and 2,819 ± 263 nA at 50 mM Cl and 4,240 ± 433 nA at 6 mM Cl. Readdition of Cl led to the restoration of the normal Cl current level (data not shown). Therefore, the increased inward currents recorded for oocytes injected with cRNA are specific Cl-dependent conductance pathways. Subsequently, the same procedure was applied to oocytes injected with CaCCX cRNA and perfused with ND-96 containing variable extracellular Cl concentrations. As shown in Fig. 5, the expressed inward currents were increased when the normal ND-96 was replaced with the low-Cl medium. The current amplitude at 100 mV in 114 mM Cl (n = 7) was 2,259 ± 71 nA and increased to 2,838 ± 108 nA in 50 mM Cl. The largest current averaged 3,441 ± 118 nA and was recorded after 6 mM Cl ND-96 was perfused for more than 5 min (Fig. 5). These results indicate that CaCCX cRNA-injected oocytes exhibited a similar dependence on external Cl as did wtCaCC.

View larger version (23K): [in this window] [in a new window]   Fig. 4.   External Cl dependence of wtCaCC. A: summary of currents recorded at 100 mV (n = 6). B: current-voltage (I-V) relationships for 1 typical cell in the presence of different extracellular Cl concentrations ([Cl]o). Reversal potential shifted from 31.2 to 12.5 and +2.4 mV, respectively, when the superfusing medium was switched to 50 mM [Cl]o and 6 mM [Cl]o from 114 mM [Cl]o, suggesting that these currents represent a Cl channel. Current increased significantly when the oocytes were perfused in 6 mM Cl. * P < 0.05 for 6 mM vs. 114 mM Cl. Data in A are ±SE.

View larger version (21K): [in this window] [in a new window]   Fig. 5.   External Cl dependence of CaCCX. A: averaged currents at 100 mV (n = 7) plotted as a function of the [Cl]o (* P < 0.05, 6 mM Cl vs. 114 mM Cl). B: I-V relationships of a typical recording. Reversal potentials in 114, 50, and 6 mM Cl were 29.8, 10.1, and +4.3 mV, respectively. Data in A are ±SE.

Ca²⁺ ionophore activation of CaCCX and CaCC. The hallmark of the epithelial Cl channel from bovine trachea is its Ca²⁺ sensitivity (14, 19). As expected, newly expressed Cl channels were activated by increasing intracellular Ca²⁺ via the application of the Ca²⁺ ionophores A-23187 or ionomycin. Figure 6 shows that the wtCaCC- and CaCCX-mediated Cl currents were elevated by 2 µM A-23187 or ionomycin. Activated currents in the presence of 2 µM ionomycin or A-23187 were recorded for oocytes expressing CaCCX (4,396 ± 551 nA, P < 0.01 vs. control, n = 7) or wtCaCC (5,832 ± 425 nA, P < 0.01 vs. control, n = 5) with 5 mM CaCl₂ in the perfusing ND-96 medium. The I-V curve of wtCaCC- and CaCCX-induced currents were plotted in the presence and absence of the Ca²⁺ ionophores, the anion channel inhibitor, and the reducing agent as shown in Fig. 6. The plots yielded an outwardly rectified current at depolarizing potentials. The reversal potential of the control oocytes expressing wtCaCC and CaCCX was approximately 30 mV.

View larger version (23K): [in this window] [in a new window]   Fig. 6.   I-V relationship of wtCaCC and CaCCX. Expression of wtCaCC or CaCCX induces an outwardly rectified conductance with a reversal potential of approximately 30 mV. A: activation effect of 2 µM A-23187 on the expressed wtCaCC currents in 20-mV steps from 100 to +80 mV. Blocking effect of 200 µM DIDS appears slightly greater than that of 2 mM DTT. B: I-V curve of wtCaCC in the presence of 2 µM ionomycin, 2 mM DTT, and 200 µM DIDS. C: I-V curve of CaCCX and the current change in the presence of 2 µM ionomycin, 200 µM DIDS, and 2 mM DTT. Data are ±SE.

DTT and DIDS inhibit wtCaCC and CaCCX. Studies on the native tracheal CaCC (29) and the cloned CaCC (8) revealed that both epithelial Cl channels are sensitive to the anion channel inhibitor DIDS and the reducing agent DTT. To verify the sensitivity of the expressing CaCC and CaCCX in Xenopus oocytes, 200 µM DIDS and 2 mM DTT were applied separately to the ionophore-activated oocytes (Figs. 2 and 6 and Table 1). DTT (2 mM) or DIDS (200 µM) significantly inhibited ionomycin-stimulated currents and further reduced the basal amplitudes. DIDS (200 µM) inhibited the currents to a greater extent (CaCCX: 565 ± 213 nA, n = 3; wtCaCC: 769 ± 241 nA, n = 3) than did 2 mM DTT (CaCCX: 1,222 ± 289 nA, n = 9; wtCaCC: 1,636 ± 254 nA, n = 10), indicating that both wtCaCC and CaCCX were sensitive to DIDS (P < 0.001; decrease of 87.1 ± 7.3% for CaCCX and 86.8 ± 3.1% for wtCaCC) and DTT (P < 0.001; decrease of 72.2 ± 5.8% for CaCCX and 71.9 ± 2.9% for wtCaCC; P < 0.001). These inhibitors had similar effects when A-23187 was used to increase intracellular Ca²⁺. These data are consistent with wtCaCC or CaCCX cRNA expressing a DTT- and DIDS-sensitive Cl conductance in oocytes. The inhibitory effect of DIDS could be reversed by washout unlike that of DTT. Addition of the same concentration of vehicle (0.1% ethanol) had no apparent effect on oocyte currents. In contrast, the endogenous CaCC was not sensitive to the reducing agent DTT (data not shown).

Chord conductances. The chord conductances at +80 mV were estimated to evaluate the functional expression of CaCC cRNA- or CaCCX cRNA-injected Xenopus oocytes and the pharmacological responses of the channels. A concentration of 2 µM ionomycin induced conductances of wtCaCC (70.9 ± 7.4 µS, n = 5) and CaCCX (50.0 ± 8.3 µS, n = 8), which were double those observed in the absence of ionophore (31 ± 4 µS for wtCaCC, n = 9; and 26.1 ± 3.6 µS for CaCCX, n = 11). The outward conductances were blocked by application of 2 mM DTT (to 16.7 ± 1.8 µS for wtCaCC, n = 10; and 15.4 ± 5.6 µS for CaCCX, n = 7) and 200 µM DIDS (to 12.7 ± 3.0 µS for wtCaCC, n = 3; and 8.0 ± 2.4 µS for CaCCX, n = 3), suggesting that the outward conductances of wt and truncated Cl channel (Cl influx) were more strongly activated by increasing intracellular Ca²⁺ and more sensitive to DIDS (results not shown).

PMA activates CaCC- and CaCCX-associated currents. Activation of PKC by phorbol esters such as PMA has been demonstrated to induce Cl secretion across salt-secreting epithelia (2). Because several PKC consensus sequences were identified in the CaCC, we tested the hypothesis that PKC may be a regulatory factor of this cloned epithelial Cl channel. Addition of 100 nM PMA to H₂O-injected oocytes (Fig. 7) resulted in no detectable increase in current at either +80 or 100 mV (n = 4). The stimulatory effect of PMA on CaCC- and CaCCX-expressing oocytes is also shown in Fig. 7 and summarized in Table 1. The average of CaCC- and CaCCX-associated currents at +80 mV in the presence of PMA alone increased significantly (P < 0.05) to 8,391 ± 761 nA from 2,888 ± 204 nA (n = 3) and to 8,001 ± 674 nA from 3,226 ± 314 nA (n = 5), respectively. PMA also induced a larger current at a hyperpolarizing membrane potential (80 mV). For the CaCCX, PMA caused a threefold increase (P < 0.05) to 9,217 ± 691 from 2,913 ± 524 nA (n = 5). As for the wtCaCC (n = 3), a fourfold activation by PMA was observed from 1,580 ± 230 to 6,593 ± 759 nA (P < 0.05). The activating effect of PMA was found across the whole range of step-clamping voltages (from 100 to +80 mV), and no significant difference was found between the responses of wtCaCC and CaCCX to PMA treatment.

View larger version (22K): [in this window] [in a new window]   Fig. 7.   Effect of phorbol 12-myristate 13-acetate (PMA) on the endogenous CaCC, the wtCaCC, or CaCCX. Injected oocytes were exposed to 100 nM PMA, which stimulated both the wtCaCC and CaCCX currents but slightly inhibited the endogenous CaCC. Left: bar plot of the averaged currents at 80 mV. Right: averaged currents at +80 mV. * P < 0.05. Data are ±SE.

DIDS and DTT inhibit the PMA-activated currents of wtCaCC and CaCCX. To further determine if the currents elicited from CaCC or CaCCX cRNA-injected oocytes with 100 µM PMA were mediated by Cl channels, 2 mM DTT and 200 µM DIDS were applied to the PMA-pretreated oocytes. Both DTT and DIDS significantly inhibited PMA-induced currents of the wtCaCC and CaCCX (Table 1). In the presence of 2 mM DIDS, the current at +80 mV of wtCaCC decreased to 2,184 ± 145 nA from 7,092 ± 168 nA (PMA-pretreated control, n = 4), whereas 2 mM DTT decreased the magnitude of the wtCaCC-associated currents from 6,929 ± 246 to 2,184 ± 145 nA (n = 3). For CaCCX-expressing oocytes, the currents decreased to 1,583 ± 213 from 5,427 ± 1,632 nA (n = 3) following the addition of 2 mM DTT and to 1,343 ± 93 from 5,352 ± 374 nA (n = 3) after the blockade of 200 µM DIDS (P < 0.05).

Dose-response relationships of PMA on wtCaCC and CaCCX. To test whether the pharmacological potency of PMA on the expressed CaCC and CaCCX was affected by truncation, the concentration response of PMA up to 1 µM was determined. The half-maximal concentration (EC₅₀) for wtCaCC was 9.2 ± 3.1 nM at +80 mV and 10.1 ± 2.1 nM at 100 mV. PMA (100 nM) stimulated ~90.3% of the maximal PMA-sensitive current at 100 mV. The EC₅₀ for PMA activation of the CaCCX was 21.0 ± 9.8 nM at +80 mV and 19.2 ± 4.8 nM at 100 mV. In the presence of 100 nM PMA, 78.8% of the maximal PMA-induced current was activated. However, the difference between the EC₅₀ of CaCC and CaCCX was not statistically significant (P > 0.05). The dose-response curves for PMA stimulation of Cl current for the wtCaCC and CaCCX are shown in Fig. 8.

View larger version (14K): [in this window] [in a new window]   Fig. 8.   Dose-response curve of PMA on the wtCaCC and CaCCX. Protein kinase C (PKC) activator PMA evoked wtCaCC () or CaCCX () currents in a concentration-dependent fashion from 0.1 to 1,000 nM PMA recorded at +80 mV. Effect of half-maximal concentrations of PMA on the wtCaCC and CaCCX are similar. Currents are normalized to the maximal inward currents. See the text for half-maximal effective concentrations. Data are ±SE.

Chelerythrine chloride prevents the activation of PMA on wtCaCC and CaCCX. To ascertain whether the stimulatory effect of PMA on the CaCC- or CaCCX-associated currents was due to activation of PKC, we pretreated the oocytes with 10 µM chelerythrine chloride, a specific PKC inhibitor, for 20 min before electrophysiological recording. The currents of CaCC- and CaCCX-expressing oocytes at +80 mV were 1,341 ± 239 nA (n = 3) and 1,649 ± 142 nA (n = 4), respectively, when treated with chelerythrine chloride (Fig. 9). After a 10-min incubation of 100 nM PMA in the presence of chelerythrine chloride, the currents were 1,594 ± 301 and 1,759 ± 162 nA for CaCC- or CaCCX-expressing oocytes, respectively. There were no statistically significant differences between currents recorded before and after addition of PMA in the presence of the inhibitor. We also tested the inhibitory effect of chelerythrine chloride on the currents evoked by PMA. Currents of wtCaCC and CaCCX in the PMA-pretreated oocytes diminished gradually following the addition of 10 µM chelerythrine chloride (data not shown). Addition of 10 mM chelerythrine chloride to CaCC- and CaCCX-expressing oocytes before exposure to 100 nM PMA was without effect on the basal current (data not shown).

View larger version (17K): [in this window] [in a new window]   Fig. 9.   Chelerythrine chloride prevents the activation of PMA on expressed wtCaCC and CaCCX in oocytes. wtCaCC- or CaCCX-injected oocytes were exposed to chelerythrine chloride (10 µM) for 20 min before the addition of agonist (100 nM PMA). Averaged currents at +80 mV are shown for both wtCaCC and CaCCX. Data are ±SE.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

In the present study, we have examined the roles that the NH₂ and COOH termini of a cloned CaCC play in essential channel functions such as sensitivity to Ca²⁺ and to physiological regulators and inhibitors and in the maintenance of biophysical characteristics such as ion selectivity. These studies were initiated because our earlier observations showed that a native tracheal CaCC migrates at 38 kDa on reducing SDS-PAGE but as a 140-kDa complex under nonreducing conditions (8). We therefore designed a construct based on the cloned CaCC cDNA sequence that might conceivably mimic a polypeptide that had been processed to form a smaller 38-kDa subunit. Because membrane proteins are frequently glycosylated in the endoplasmic reticulum and Golgi and the primary amino acid sequence of the cloned CaCC protein could be glycosylated in vitro, the truncated construct was designed to contain the four transmembrane domains and six glycosylation sites, consistent with a membrane-spanning protein. Because we have previously shown that the cloned and native CaCCs are activated by CaMKII-dependent phosphorylation (14, 19), this construct also preserved 3 of the 10 consensus CaMKII sites. In addition to determining if a smaller form of the CaCC could make a viable channel, we also examined whether PKC could be a physiological regulator of this conductance, given that the primary cDNA sequence encodes 13 consensus sites for PKC phosphorylation.

We chose Xenopus oocytes as the expressing cell system because oocytes promiscuously translate and process injected RNA and because niflumic acid inhibits the endogenous oocyte CaCC but not the epithelial CaCC (8). Our results demonstrated that the endogenous CaCC is evoked by an increase in intracellular Ca²⁺ triggered through incubation with the Ca²⁺ ionophore and is inhibited by niflumic acid as previously reported (37). The possibility that the recorded currents in CaCC- or CaCCX-expressing oocytes were contaminated by the endogenous CaCC could therefore be excluded by inclusion of 100 µM niflumic acid in the bath (Fig. 3), enough to inhibit all of the endogenous CaCC-associated current (37). It was easy to distinguish the H₂O injected from the CaCC or CaCCX cRNA-injected oocytes from the respective current amplitudes and chord conductances. In contrast, there were no significant differences between wtCaCC- and CaCCX-expressing oocytes as determined from the current amplitudes, the chord conductances, the pharmacological profile, or the Ca²⁺ ionophore-sensitive components. The inhibitory actions of DTT and DIDS and the activating effect of ionomycin are consistent with our previous observations obtained from oocyte membrane vesicles containing wtCaCC reconstituted into planar lipid bilayers (19), two-electrode voltage-clamp recordings from wtCaCC-expressing oocytes, and whole cell currents from transfected COS-7 cells (8). Because we observed identical pharmacological properties of the currents in both wtCaCC- and CaCCX-expressing oocytes, the truncated construct forms a functional channel.

Our previous results on the native CaCC isolated from the trachea and the cloned CaCC suggested that these epithelial Cl channels are regulated by at least two signal transduction systems: Ca²⁺ and inositol phosphate messengers. Intracellular Ca²⁺ and CaMKII activate the native and cloned CaCCs, and D-myo-inositol 3,4,5,6-tetrakisphosphate downregulates the cloned CaCC incorporated into planar lipid bilayers (14, 19). In epithelial cells, PKC and CaMKII are activated by increasing intracellular Ca²⁺ and are involved in the regulation of Cl secretion (36). In contrast, the endogenous CaCC of Xenopus oocytes is inhibited by PKC-activating phorbol esters, as has been demonstrated by Dascal and colleagues (3). To test the hypothesis that activation of the PKC pathway could stimulate the expressed CaCC, the potent PKC activator PMA was used to pretreat oocytes expressing either the wtCaCC or CaCCX. The upregulatory influence of PMA (Figs. 7 and 8) supports the hypothesis that PKC can regulate this expressed epithelial CaCC and its truncated construct. Moreover, the PKC-specific inhibitor, chelerythrine chloride, prevents the PMA-induced increase in current in wtCaCC- or CaCCX-expressing oocytes, consistent with PKC-dependent activation of current in these cells.

The significance of our observations that a severely truncated protein can form a functional channel lies in the role that posttranslational processing may play in determining channel structure. Several membrane receptors and glycoproteins are translated as large precursor molecules that are subject to posttranslational cleavage to a functional product. These include the human insulin proreceptor (31), cadherin precursors (18), and the hepatocyte growth factor receptor (7). Proteolytic cleavage would be a novel processing mechanism for an ion channel. However, our present results suggest that processing of the wtCaCC polypeptide may account for the disparity in molecular mass of the native and cloned CaCC, as a severely truncated form of the protein makes a channel that is indistinguishable from the wild-type channel when expressed in Xenopus oocytes. Furthermore, the four PKC sites that are preserved in CaCCX respond to PMA as does the wtCaCC construct, although 13 predicted PKC sites are available in the full-length protein. These results suggest that the regions of the channel critical to regulation are preserved in the truncated CaCC, although we cannot discount the possibility that subtle changes in kinetics or submicromolar Ca²⁺ affinities have been altered by this truncation and would therefore reveal differences between full-length and truncated proteins. In addition to sites for kinase phosphorylation and glycosylation, analysis of the cDNA sequence of the tracheal CaCC revealed the presence of 13 possible monobasic cleavage sites (9). Thus a large portion of the wtCaCC could be cleaved without deleterious effects on channel function. Whether the truncated construct truly represents the smallest 38-kDa subunit of the native tracheal CaCC chloride channel or whether the cloned and native channels represent two unrelated polypeptides remains to be determined.