Characterization of a charybdotoxin-sensitive intermediate conductance Ca2+-activated K+ channel in porcine coronary endothelium: relevance to EDHF

Abstract

1. This study characterizes the K(+) channel(s) underlying charybdotoxin-sensitive hyperpolarization of porcine coronary artery endothelium. 2. Two forms of current-voltage (I/V) relationship were evident in whole-cell patch-clamp recordings of freshly-isolated endothelial cells. In both cell types, iberiotoxin (100 nM) inhibited a current active only at potentials over +50 mV. In the presence of iberiotoxin, charybdotoxin (100 nM) produced a large inhibition in 38% of cells and altered the form of the I/V relationship. In the remaining cells, charybdotoxin also inhibited a current but did not alter the form. 3. Single-channel, outside-out patch recordings revealed a 17.1+/-0.4 pS conductance. Pipette solutions containing 100, 250 and 500 nM free Ca(2+) demonstrated that the open probability was increased by Ca(2+). This channel was blocked by charybdotoxin but not by iberiotoxin or apamin. 4. Hyperpolarizations of intact endothelium elicited by substance P (100 nM; 26.1+/-0.7 mV) were reduced by apamin (100 nM; 17.0+/-1.8 mV) whereas those to 1-ethyl-2-benzimidazolinone (1-EBIO, 600 microM, 21.0+/-0.3 mV) were unaffected (21.7+/-0.8 mV). Substance P, bradykinin (100 nM) and 1-EBIO evoked charybdotoxin-sensitive, iberiotoxin-insensitive whole-cell perforated-patch currents. 5 A porcine homologue of the intermediate-conductance Ca(2+)-activated K(+) channel (IK1) was identified in endothelial cells. 6. In conclusion, porcine coronary artery endothelial cells express an intermediate-conductance Ca(2+)-activated K(+) channel and the IK1 gene product. This channel is opened by activation of the EDHF pathway and likely mediates the charybdotoxin-sensitive component of the EDHF response.

Figure 1

Effect of iberiotoxin, charybdotoxin and apamin on ‘linear' and ‘non-linear' whole-cell K⁺ currents in porcine coronary artery endothelial cells. Currents were elicited by 10 mV step pulses between −150 and +150 mV from a holding potential of −60 mV using 500 nM free Ca²⁺ pipette solution. Cells displayed ‘linear' (A and B) and ‘non-linear' (C and D) I/V relationships. Representative current traces (A and C) and the mean data (B, n=12 and D, n=9) obtained under control conditions (Con) and after the cumulative application of iberiotoxin (100 nM; IbTX), charybdotoxin (100 nM; ChTX) and apamin (100 nM; Apa).

Figure 2

Single-channel recordings obtained from outside-out patches taken from porcine coronary artery endothelial cells. Activity was recorded at the indicated holding potentials under asymmetrical K⁺ gradients with Ca²⁺ fixed at 250 nM in the pipette solution. Representative traces are shown, with ‘c' indicating the closed stated (A). Mean unitary currents were plotted against holding potential and fitted with a linear function (B). The slope conductance obtained was 17.1±0.4 pS.

Figure 3

Effect of Ca²⁺ and toxins on 17.1 pS channel activity recorded from outside-out patches taken from porcine coronary artery endothelial cells. Distributions of unitary conductance amplitude recorded at 0 mV are shown, together with representative traces (closed state is indicated by ‘c'). A Gaussian fit was performed for every single channel experiment, and the amplitude of single channels was determined with the best fit. Then, the mean of the obtained amplitudes was calculated. The effect of Ca²⁺ was determined using pipette solutions with free Ca²⁺ fixed at 100, 250 and 500 nM (panel A). The sensitivity of the channel to the cumulative application of iberiotoxin (IbTX, 100 nM) and charybdotoxin (ChTX, 100 nM) was recorded using 250 nM free Ca²⁺ in the pipette solution (panel B).

Figure 4

Effect of apamin and the combination of apamin plus charybdotoxin on substance P- and 1-EBIO-evoked hyperpolarizations of porcine coronary artery endothelium. Substance P and 1-EBIO were added as bolus injections, calculated to give transiently the required final concentrations (100 nM and 600 μM, respectively), to a recording chamber perfused with Kreb's solution containing apamin (Apa, 100 M) and charybdotoxin (ChTX, 100 nM) as indicated. Membrane potential responses were recorded using microelectrodes filled with 3 M KCl. A representative trace (A) and mean data derived from four such experiments (B: columns represent the mean±s.e.mean resting membrane potential immediately before exposure (top of column) and mean±s.e.mean peak hyperpolarization (bottom of column) during exposure to each agent) are shown. Levcromakalim (10 μM) was included as an internal control.

Figure 5

Characterization of substance P-evoked currents in porcine coronary artery endothelial cells determined using the perforated-patch configuration (holding potential of −40 mV using 500 nM free Ca²⁺ in the pipette solution). Panel A: Outward currents resulting from the application of substance P (SP, 100 nM, indicated by horizontal bar) to the bathing solution in the presence of iberiotoxin (IbTX, 100 nM) and charybdotoxin (ChTX, 100 nM) are shown. Panel B: During the steady-state phase of the SP-evoked currents, 10 mV voltage steps between −120 mV and +80 mV were applied from a holding potential of −80 mV. The effect of charybdotoxin is shown.

Figure 6

Characterization of bradykinin and 1-EBIO evoked currents in porcine coronary artery endothelial cells recorded using the perforated-patch configuration. K⁺ currents were elicited by voltage steps to +50 mV from a holding potential of −80 mV (500 nM free Ca²⁺ in the pipette solution) in control conditions (Con), after application of bradykinin (BK, 100 nM) and after application iberiotoxin (IbTX, 100 nM), apamin (Apa, 100 nM) and charybdotoxin (ChTX, 100 nM) Ms shown (panel A, representative traces and mean data are shown, n=7). Identical experiments were performed using 1-EBIO (300 μM) in place of BK (panel B, n=7).

Figure 7

Comparison of deduced amino acid sequences of porcine and human IK1. Porcine IK1 (submitted as GenBank AY062036) was sequenced from coronary artery endothelium cDNA using 3′ and 5′ RACE with gene-specific primers designed to homologous regions of human and rat IK1 sequences. Underlined residues constitute transmembrane spans (S1-S6) and the pore (P) region according to Vandorpe et al. (1998); boldfaced residues are conserved consensus calmodulin-binding domain (Conserved Domain Database, NCBI).