AICAR decreases the activity of two distinct amiloride-sensitive Na+-permeable channels in H441 human lung epithelial cell monolayers

Abstract

Transepithelial transport of Na(+) across the lung epithelium via amiloride-sensitive Na(+) channels (ENaC) regulates fluid volume in the lung lumen. Activators of AMP-activated protein kinase (AMPK), the adenosine monophosphate mimetic AICAR, and the biguanide metformin decreased amiloride-sensitive apical Na(+) conductance (G(Na+)) in human H441 airway epithelial cell monolayers. Cell-attached patch-clamp recordings identified two distinct constitutively active cation channels in the apical membrane that were likely to contribute to G(Na+): a 5-pS highly Na(+) selective ENaC-like channel (HSC) and an 18-pS nonselective cation channel (NSC). Substituting NaCl with NMDG-Cl in the patch pipette solution shifted the reversal potentials of HSC and NSC, respectively, from +23 mV to -38 mV and 0 mV to -35 mV. Amiloride at 1 microM inhibited HSC activity and 56% of short-circuit current (I(sc)), whereas 10 microM amiloride partially reduced NSC activity and inhibited a further 30% of I(sc). Neither conductance was associated with CNG channels as there was no effect of 10 microM pimoside on I(sc), HSC, or NSC activity, and 8-bromo-cGMP (0.3-0.1 mM) did not induce or increase HSC or NSC activity. Pretreatment of H441 monolayers with 2 mM AICAR inhibited HSC/NSC activity by 90%, and this effect was reversed by the AMPK inhibitor Compound C. All three ENaC proteins were identified in the apical membrane of H441 monolayers, but no change in their abundance was detected after treatment with AICAR. In conclusion, activation of AMPK with AICAR in H441 cell monolayers is associated with inhibition of two distinct amiloride-sensitive Na(+)-permeable channels by a mechanism that likely reduces channel open probability.

Fig. 1.

Effect of AICAR and metformin on G_Na+ in H441 cell monolayers. A: typical continuous apical current (I_ap) traces from untreated control, 2 mM AICAR-, or 2 mM metformin-treated monolayers bathed in a potassium gluconate solution with 11.5 mM Na⁺ (see materials and methods). Basolateral membranes were permeabilized with nystatin (75 μM), and the Na⁺ concentration of the apical bath raised to ∼50 mM to create a driving force for Na⁺ influx (Na⁺). Amiloride was then added to the apical bath (amiloride) to determine the amiloride-sensitive G_Na+. Arrows indicate the point of application of each drug. B: apical amiloride-sensitive G_Na+ determined from untreated control, AICAR-, and metformin-treated monolayers. Data are presented as means ± SE. *Significantly different from control, P ≤ 0.05, n = 3.

Fig. 2.

Properties of 2 distinct cation channels in cell-attached patches from apical membrane of H441 cell monolayers. A(i) and (ii): representative trace showing constitutive activity of cation channel currents in a cell-attached patch at −100 mV on different time scales that had unitary amplitudes of ∼−0.5 pA and long openings. A(iii) amplitude histogram of the cation channel currents shown in A(i) that had 1 closed (C) and 2 open levels (O₁ and O₂) indicating the patch contained at least 2 channels. A(iv): mean current/voltage (I/V) relationship of these cation channel currents illustrating that in 145 mM NaCl, the channel currents had a slope conductance and reversal potential (E_r) of 5 pS + 23 mV, respectively (each point from at least n = 5). In the presence of 145 mM NMDG-Cl, the I/V relationship had extrapolated E_r of −38 mV (each point from at least n = 4). B(i) and (ii): a cell-attached patch that contained 2 distinct channel currents at −100 mV. B(iii) amplitude histogram of the cation channel currents shown in (i) that had 2 major peaks illustrating that the unitary amplitudes of O₁ and O₂ were −0.58 pA and −1.76 pA, respectively. The other peaks represent multiple openings of both types of channels. In B, mean I/V relationship shows that the larger amplitude channel currents had a slope conductance of 18 pS and an E_r of +4 mV, and that in the presence of NMDG, the extrapolated E_r was shifted to −35 mV (each point from at least n = 4).

Fig. 3.

Differential sensitivity of highly Na⁺ selective channel (HSC) and nonselective cation channel (NSC) activity to amiloride in cell-attached patches from H441 cell monolayers. A: inclusion of 1 μM amiloride in the patch pipette solution produced a pronounced inhibition of HSC activity after 10–20 s at −100 mV. B is a typical trace showing that NSC activity at −100 mV was not inhibited by inclusion of 1 μM amiloride in the patch pipette solution, whereas C illustrates that 10 μM amiloride induced partial inhibition of NSC activity. D: mean data of relative NP_o after application of 1 and 10 μM amiloride to the bath compared with normalized NP_o for patches with HSC alone or HSC + NSC before application of amiloride (control). ***Significantly different from control P < 0.001.

Fig. 4.

Effect of amiloride on transepithelial short-circuit current (I_sc) across H441 cell monolayers. A: typical trace showing changes in transepithelial I_sc measured across H441 cell monolayers in response to application of amiloride (0.01–100 μM) and ouabain (1 mM). B: mean data showing transepithelial I_sc from untreated monolayers (control) or those treated with 1 and 10 μM amiloride. Data are expressed as % control I_sc. ***Significantly different from control P < 0.001, †significantly different from 1 μM amiloride, P < 0.01, n = 6. C: amiloride concentration effect curve. Data were fitted with a sigmoid dose response curve with a Hill slope of −1.04 and an r² value of 0.9.

Fig. 5.

Effect of inhibitors of CNG channels on transepithelial I_sc across H441 cell monolayers. A: transepithelial I_sc measured across H441 cell monolayers that were untreated (control) or treated with 10 μM pimoside followed by 10 μM amiloride (pimoside + amiloride). B: transepithelial I_sc measured across H441 cell monolayers that were untreated (control) or treated with 50 μM l-cis-diltiazem followed by 10 μM amiloride (l-cis + amiloride). C: transepithelial I_sc measured across H441 cell monolayers that were untreated (control) or treated with 10 μM amiloride followed by 10 μM pimoside (amiloride + pimoside) or l-cis-diltiazem (amiloride+ l-cis). Data are expressed as % control I_sc to normalize between experiments. *Significantly different from control.

Fig. 6.

Effect of an inhibitor and activator of CNG channels in cell-attached patches from H441 cell monolayers. A: inclusion of 10 μM pimoside (a CNG channel inhibitor) in the patch pipette solution had no effect on either HSC or NSC activity at −100 mV. B and C: bath application of 500 μM 8-bromo-cGMP (a CNG channel activator) had no effect on HSC or NSC activity, respectively, at −100 mV.

Fig. 7.

Effect of AICAR, an activator of AMP-activated protein kinase (AMPK), and Compound C, an inhibitor of AMPK, on HSC and NSC in cell-attached patches from H441 cell monolayers. A(i) shows a control patch that contained constitutive HSC and NSC activity at −100 mV that was maintained for the duration of the recording. A(ii) illustrates a patch that contained infrequent activity of a small amplitude channel current following pretreatment with 2 mM AICAR for 1 h. A(iii) shows a patch that contained marked HSC and NSC activity after pretreatment with 80 μM Compound C for 30 min followed by incubation with 80 μM Compound C and 2 mM AICAR for a further 1 h. Note the scale bar refers to i, ii, and iii. B: mean cation channel activity in control patches (n = 6) in the presence of 2 mM AICAR alone (n = 12) and 80 μM Compound C and 2 mM AICAR (n = 6) at −100 mV (***P < 0.001). C: acute bath application of 2 mM AICAR produced a marked inhibition of both HSC and NSC activity at −100 mV after 5–10 min. D: bath application of 2 mM AICAR had no effect on HSC/NSC activity in an inside-out patch held at −100 mV.

Fig. 8.

Effect of AICAR on apical abundance of α-, β-, and γENaC proteins in H441 cell monolayers. A: typical Western blots of apical biotinylated (bound-biotinylated) and non-biotinylated (nonbound) protein samples from H441 cell monolayers. Samples were untreated (control) or treated with AICAR or MG132. Proteins were immunostained for αENaC using anti NH₂-terminal and COOH-terminal antisera, βENaC, γENaC, and β-actin. The position of protein size markers is shown at left and the sizes of predominant immunostained proteins at right of the images. Images shown for each antiserum are from the same immunoblot but may have been digitally reordered for consistency of presentation. B: densitometry analysis of immunostained proteins from apical biotinylated protein samples. Data are shown as means ± SE for n = 3 samples. *Significantly different from control P < 0.05.