High concentrations of caffeine are known to inhibit inward rectifier (I_K1) potassium currents (Varro et al. 1993; Cui & Terrar, 1995). In the present study we investigated whether caffeine also inhibted I_to and I_K.

Rats were stunned and humanely killed by cervical dislocation (Schedule 1, Scientific Procedures Act, 1986). The heart was digested with collagenase and protease to isolate ventricular myocytes. Voltage clamp was performed using the perforated patch method using amphotericin B (240 mg ml^-1). Internal pipette solutions contained (mmol l^-1): potassium glutamate, 120; KCl, 20; NaCl, 10 and Hepes, 10; pH 7.25 with KOH. External solutions contained (mmol l^-1): NaCl, 140; KCl, 5.4; MgCl₂, 1; CaCl₂, 1; glucose, 10; Hepes, 10; pH 7.35 with NaOH. Experiments were carried out at 35 °C. Data are presented as means ± S.E.M.

Initial experiments verified the inhibitory action of 10 mmol l^-1 caffeine on I_K1, and showed that these effects were independent of [Ca²⁺]_i, as described previously (Cui & Terrar, 1995). Simultaneous measurement of sacolemmal current and [Ca²⁺]_i with fura-2 showed that inhibition of the outward component of I_K1 could account for the inhibitory effect of caffeine on the background holding current observed at -40 mV: the holding current was inhibited by 0.53 ± 0.10 pA pF^-1 (n = 4) in the absence and by 0.48 ± 0.20 pA pF^-1 (n = 4) in the presence of 10 µmol l^-1 BAPTA.

I_to, elicited by square voltage pulses of 1 s duration from -40 to +60 mV, was inhibited by 1.75 ± 0.52 pA pF^-1 (n = 4) in the presence of 10 mmol l^-1 caffeine, without altering the voltage dependence of activation or inactivation. Steady-state I_K, recorded at the end of clamp pulse, were inhibited by 2.15 ± 0.25 pA pF^-1 (n = 7) in the presence of 10 mmol l^-1 caffeine (Fig. 1). These data show that caffeine has a small but significant inhibitory action on three different types of K channels in cardiac muscle.

All procedures accord with current UK legislation.