Abstract
Key points: Na+ current inactivation is biphasic in insulin-secreting cells, proceeding with two voltage dependences that are half-maximal at ∼-100 mV and -60 mV. Inactivation of voltage-gated Na+ (NaV ) channels occurs at ∼30 mV more negative voltages in insulin-secreting Ins1 and primary β-cells than in HEK, CHO or glucagon-secreting αTC1-6 cells. The difference in inactivation between Ins1 and non-β-cells persists in the inside-out patch configuration, discounting an involvement of a diffusible factor. In Ins1 cells and primary β-cells, but not in HEK cells, inactivation of a single NaV subtype is biphasic and follows two voltage dependences separated by 30-40 mV. We propose that NaV channels adopt different inactivation behaviours depending on the local membrane environment. Pancreatic β-cells are equipped with voltage-gated Na+ channels that undergo biphasic voltage-dependent steady-state inactivation. A small Na+ current component (10-15%) inactivates over physiological membrane potentials and contributes to action potential firing. However, the major Na+ channel component is completely inactivated at -90 to -80 mV and is therefore inactive in the β-cell. It has been proposed that the biphasic inactivation reflects the contribution of different NaV α-subunits. We tested this possibility by expression of TTX-resistant variants of the NaV subunits found in β-cells (NaV 1.3, NaV 1.6 and NaV 1.7) in insulin-secreting Ins1 cells and in non-β-cells (including HEK and CHO cells). We found that all NaV subunits inactivated at 20-30 mV more negative membrane potentials in Ins1 cells than in HEK or CHO cells. The more negative inactivation in Ins1 cells does not involve a diffusible intracellular factor because the difference between Ins1 and CHO persisted after excision of the membrane. NaV 1.7 inactivated at 15--20 mV more negative membrane potentials than NaV 1.3 and NaV 1.6 in Ins1 cells but this small difference is insufficient to solely explain the biphasic inactivation in Ins1 cells. In Ins1 cells, but never in the other cell types, widely different components of NaV inactivation (separated by 30 mV) were also observed following expression of a single type of NaV α-subunit. The more positive component exhibited a voltage dependence of inactivation similar to that found in HEK and CHO cells. We propose that biphasic NaV inactivation in insulin-secreting cells reflects insertion of channels in membrane domains that differ with regard to lipid and/or membrane protein composition.