Patch‐clamp recordings

Coverslips with attached cells were transferred to the recording chamber where they were perfused with DMEM or Tyrode solution (146 mm NaCl, 4.5 mm KCl, 1.1 mm CaCl₂, 1 mm MgCl₂, 10 mm Hepes, 10 mm glucose, pH 7.4 adjusted with NaOH). All Hepes‐buffered solutions were continuously bubbled with 100% oxygen. Carbogen gas was used for DMEM. An Axopatch 200B patch‐clamp amplifier in combination with a Digidata 1440A and Clampex 10 software (Molecular Devices Inc., Sunnyvale, CA, USA) were used for the whole cell currents and action potential (AP) recordings as previously (Karppinen et al. 2015). Ag/AgCl electrodes (World Precision Instruments Inc., Sarasota, FL, USA) were used for recording and bath electrodes. Patch pipettes were pulled from borosilicate glass capillary tubing (1.5 mm o.d., 0.86 mm i.d., Harvard Apparatus, Edenbridge, UK) with a micropipette puller (Sutter P‐97, Sutter Instrument Company, Novato, CA, USA) and fire polished. Patch electrode resistances were 1.5–3 MΩ when filled with any pipette solution. The junction potential was corrected by setting the baseline of the test potential as zero before breaking the cell membrane. After establishment of a ruptured patch, the Tyrode solution was changed to the recording solution depending on the current in question. Recordings were carried out after a period of 5 min to allow adequate intracellular dialysis. The cell capacitance and series resistance were compensated electronically, monitored throughout the experiments, and corrected for time‐dependent changes. Recordings with an unstable or poor access resistance (10% change over the period of recording or initial access resistance larger than 10 MΩ) were discarded. Typically, the access resistance was 3–6 MΩ. In voltage clamp, control cells were held at −80 mV (excluding Na⁺/Ca²⁺‐exchanger (NCX) current recordings) and membrane capacitance and membrane resistance were calculated in response to a 5 mV pulse using the Membrane Test function of Clampex. To correct for variability in cell size, current amplitudes were normalized by cell capacitance. Recordings were carried out at a sampling rate of 10 kHz, low‐pass Bessel filtered at 5 kHz, stored on a computer and analysed off‐line with pCLAMP 10 software (Molecular Devices Inc.).

A previously described protocol was used (Xu et al. 2011). Briefly, L‐type Ca²⁺ currents were measured using 200 ms voltage steps ranging from −30 to +50 mV in 10 mV increments at a frequency of 0.2 Hz after an initial 1 s pre‐pulse at −40 mV. The voltage dependence of inactivation was assessed by application of 2 s pre‐pulses of different voltage amplitudes (in a range from −40 mV to +10 mV), which were followed by a 100 ms test pulse to +10 mV. The solutions used were: internal solution (110 mm CsOH, 90 mm aspartic acid, 20 mm CsCl, 10 mm tetraethyl ammonium chloride (TEA), 10 mm Hepes, 10 mm EGTA, 5 mm Mg‐ATP₂, 5 mm sodium creatine phosphate, 0.4 mm GTP‐Tris, 0.1 mm leupeptin, pH 7.2 adjusted with CsOH) and bath solution (125 mm N‐methyl‐glucamine, 5 mm 4‐aminopyridine (4‐AP), 20 mm TEA chloride, 2 mm CaCl₂, 2 mm MgCl₂, 10 mm glucose, 10 mm Hepes, pH 7.4 adjusted with HCl).

The solutions and protocol for the NCX current (I _NCX) recordings were as described previously (Despa et al. 2003). The cells were held at −15 mV. I _NCX was measured as Ni²⁺‐sensitive current using a voltage ramp from +100 to −100 mV. The bath solution contained 140 mm NaCl, 1 mm MgCl₂, 1 mm CaCl₂, 0.01 mm nifedipine, 0.01 mm niflumic acid, 10 mm glucose and 10 mm Hepes, pH 7.4 adjusted with Tris. NiCl₂, 10 mm, was added to block I _NCX. The pipette solution contained 10 mm NaCl, 20 mm CsCl, 80 mm CsOH, 80 mm glutamic acid, 20 mm TEA chloride, 5 mm Tris‐ATP, 5.7 mm MgCl₂, 2.69 mm CaCl₂, 5 mm BAPTA (200 nm free Ca²⁺) and 10 mm Hepes, pH 7.2 adjusted with CsOH.

The protocol for potassium current measurements was adopted from Rivard and colleagues (Rivard et al. 2009). The total K⁺ current (I _peak) was elicited by a series of 500 ms voltage steps varying from −110 to +50 mV in 10 mV increments at a frequency rate of 0.2 Hz. The current density of the inward rectifier K⁺ current (I _K1) was determined at the end of the voltage steps, ranging from −110 to −40 mV. To eliminate the transient outward K⁺ current (I _Kto), an inactivating pre‐pulse was applied (50 ms, −40 mV). The remaining current consists of the ultra‐rapid delayed rectifier K⁺ current (I _Kur) and the steady‐state outward K⁺ current (I _Kss). 4‐AP (100 μm; I _Kur inhibitor) was applied to record I _Kss. The density of I _Kto was obtained by subtracting the current traces measured with and without the inactivating pre‐pulse, while I _Kur was measured as a subtraction of currents recorded in the absence and presence of 4‐AP. The current densities of each of the three components of the outward K⁺ currents were determined at the peak current. The internal solution contained 110 mm potassium aspartate, 20 mm KCl, 8 mm NaCl, 1 mm MgCl₂, 1 mm CaCl₂, 10 mm BAPTA, 4 mm K₂ATP, and 10 mm Hepes, pH 7.2 adjusted with KOH. Cells were perfused with Tyrode solution (130 mm NaCl, 5.4 mm KCl, 1 mm CaCl₂, 1 mm MgCl₂, 0.3 mm Na₂HPO₄, 10 mm Hepes, and 5.5 mm glucose, pH 7.4 adjusted with NaOH).

APs were elicited by a 1 ms current injection and measured using the current clamp mode. The intracellular solution used was the same as described previously (120 mm potassium aspartate, 25 mm KCl, 1 mm MgCl₂, 2 mm sodium creatine phosphate, 4 mm Na₂‐ATP, 2 mm NaGTP, 10 mm EGTA and 5 mm Hepes, pH 7.2 adjusted with KOH) (Yang et al. 2005) and the bath solution was DMEM.