Induction of a novel cation current in cardiac ventricular myocytes by flufenamic acid and related drugs

Abstract

Background and purpose: Interest in non-selective cation channels has increased recently following the discovery of transient receptor potential (TRP) proteins, which constitute many of these channels.

Experimental approach: We used the whole-cell patch-clamp technique on isolated ventricular myocytes to investigate the effect of flufenamic acid (FFA) and related drugs on membrane ion currents.

Key results: With voltage-dependent and other ion channels inhibited, cells that were exposed to FFA, N-(p-amylcinnamoyl)anthranilic acid (ACA), ONO-RS-082 or niflumic acid (NFA) responded with an increase in currents. The induced current reversed at +38 mV, was unaffected by lowering extracellular Cl(-) concentration or by the removal of extracellular Ca(2+) and Mg(2+), and its inward but not outward component was suppressed in Na(+)-free extracellular conditions. The current was suppressed by Gd(3+) but was resistant to 2-aminoethoxydiphenyl borate (2-APB) and to amiloride. It could not be induced by the structurally related non-fenamate anti-inflammatory drug diclofenac, nor by the phospholipase-A(2) inhibitors bromoenol lactone and bromophenacyl bromide. Muscarinic or alpha-adrenoceptor activation or application of diacylglycerol failed to induce or modulate the current.

Conclusions and implications: Flufenamic acid and related drugs activate a novel channel conductance, where Na(+) is likely to be the major charge carrier. The identity of the channel remains unclear, but it is unlikely to be due to Ca(2+)-activated (e.g. TRPM4/5), Mg(2+)-sensitive (e.g. TRPM7) or divalent cation-selective TRPs.

Figure 1

Chemical structures of flufenamic acid and related drugs. Flufenamic (2-{[3-(trifluoromethyl)phenyl]amino}benzoic) acid is a typical fenamate, containing a phenyl ring attached, via an amino group, to benzoic acid (the aminobenzoic acid group is also called anthranilic acid). ACA (N-(p-amylcinnamoyl)anthranilic acid and ONO-RS-082 (2-(p-amylcinnamoyl)amino-4-chloro benzoic acid) contain the aminobenzoic acid moiety but with a cinnamoyl, instead of a phenyl group attached to the amino group. The two drugs differ by the substituents (H vs. Cl) in position 4 of the benzoic moiety. Diclofenac (2-[2-(2,6-dichlorophenyl)aminophenyl]acetic acid) differs from flufenamic acid by having a phenylacetyl instead of the benzoyl (phenylcarboxyl) group, and by the substitutions in the second phenyl group. Niflumic acid (2-{[3-(trifluoromethyl)phenyl]amino}nicotinic acid) is similar to flufenamic acid but its benzoic group is replaced by a nicotinic group.

Figure 2

Flufenamic acid (FFA)- and N-(p-amylcinnamoyl)anthranilic acid (ACA)-induced current in ventricular myocytes. A, C: Effects of extracellular application of FFA (100 µM; A) or ACA (50 µM; C) on whole-cell membrane currents. Traces of current–voltage relationships obtained using voltage ramps from +80 mV to −120 mV, under control extracellular conditions (○), in the presence of either drug (•) and after drug washout (□). B, D: Time course of whole-cell currents measured at +80 mV and −120 mV in the same cells as in A and C respectively. The period of superfusion with drug is indicated by horizontal bar.

Figure 3

Effect of ONO-RS-082 and niflumic acid (NFA), but lack of effect of diclofenac in inducing current in ventricular myocytes. A, C: Traces of current–voltage relationships obtained using voltage ramps from +80 mV to −120 mV, under control extracellular conditions (○) and in the presence of either drug (•): 100 µM ONO-RS-082 (A) or 300 µM NFA (C) respectively. B, D: Time course of whole-cell currents measured at +80 mV and −120 mV in the same cell as in A and C respectively. The period of superfusion with drugs is indicated by horizontal bar. E, F: Failure of diclofenac to increase currents or to prevent the effect of flufenamic acid (FFA). ACA, N-(p-amylcinnamoyl)anthranilic acid.

Figure 4

Concentration dependence of the effect of N-(p-amylcinnamoyl)anthranilic acid (ACA). Pooled data (n = 3–23) on the effects of different concentrations of the drugs tested. Data points for ACA are fitted by the Hill equation (see Methods), with V_max = −0.50 pA/pF, K_0.5 = 24 µM; and n_Hill = 1.5 (continuous curve). Data for the other drugs were not fitted because only two concentrations of each were tested. FFA, flufenamic acid; NFA, niflumic acid.

Figure 5

N-(p-amylcinnamoyl)anthranilic acid (ACA)-induced current in sheep ventricular myocytes. A: Traces of currents obtained using voltage steps between −100 mV and +70 mV, with 10 mV increments (but displayed with 20 mV increments), under control conditions (left), in the presence of 30 µM ACA (middle), and following drug washout (right). B: Current–voltage relationships obtained using end-of-pulse currents in A: under control conditions (○), in the presence of ACA (•) and after drug washout (□). C: Time course of currents at +80 mV and −120 mV obtained using voltage ramps in the same cell as in the other panels. The period of superfusion with ACA is indicated by horizontal bar.

Figure 6

Cation permeability of the N-(p-amylcinnamoyl)anthranilic acid (ACA)-induced conductance. A, B, C: Effects of ACA (30 µM) during superfusion with Na⁺-free (Na⁺_o replaced by NMDG⁺_o) or Na⁺-containing extracellular solutions. Data from one same single cell. A: Time course of current at +80 mV and −120 mV. B, C: Current–voltage relationships under basal condition (○), in the presence of ACA (•) and after drug washout (□), during superfusion with Na⁺-free extracellular solution (B) or during superfusion with standard (Na⁺-containing) extracellular solution (C). D: Failure of Ca²⁺ to increase inward current in the presence of ACA (30 µM) in a cell superfused with Na⁺-free extracellular solution. E: Failure of removal of intracellular Na⁺ to suppress the outward current induced by ACA (30 µM) at positive potentials. F: Pooled data of ACA-induced currents measured at −120 mV in the presence of different types of extracellular solutions. **P < 0.01 versus control; anova with post hoc test.

Figure 7

Lack of involvement of diacylglycerol or phospholipase A₂. A: Lack of effect of extracellularly applied oleoyl acetyl glycerol (OAG; 100 µM) on membrane currents measured at +80 mV and −120 mV using voltage ramps. B: Lack of effect of α-adrenoceptor (phenylephrine, PE; 40 µM) and muscarinic (carbachol, CCh; 100 µM) agonists on the current induced by N-(p-amylcinnamoyl)anthranilic acid (ACA). C, D: Lack of effect of extracellular application of bromoenol lactone (BEL; 25 µM; C) or bromophenacyl bromide (BPB; 50 µM; D) on membrane currents measured at −120 mV. Periods of drug application are indicated by horizontal bars.

Figure 8

N-(p-amylcinnamoyl)anthranilic acid (ACA) and diclofenac effects on the action potential. A, D: Superimposed action potentials recorded before and during application of increasing concentrations of ACA (A) or of diclofenac (D). The symbol and trace colours in A and D correspond to the column colours for the same concentrations in B and E respectively. The horizontal dashed lines indicate the 0 mV level. B, E: Action potential duration measured at 90% repolarization (APD₉₀) in presence of various concentrations of ACA (B) or of diclofenac (E). (*P < 0.05 for ACA vs. control; anova with post-test) C, F: Effect of 100 µM ACA (C) or 300 µM diclofenac (F) on the resting membrane potential (*P < 0.05 for ACA vs. control). Notice the depolarization of the resting membrane and the relative lengthening of the action potential as the concentration of ACA but not that of diclofenac is raised. Number of cells for each drug concentration indicated above the columns. Stimulation at 1 Hz.