The role of extracellular Mg²⁺ in the regulation of smooth muscle contractility and reactivity is relatively well characterised. The role of intracellular Mg²⁺ ([Mg²⁺]_i), maintained between 0.5 and 1 mM), in the control of vascular function, however, is less known (e.g. Altura et al. 1993). Recent evidence suggests that [Mg²⁺]_i can rapidly change in response to various vasoactive substances (Touyz & Schiffrin, 1996; Zheng et al. 2001). Also, Gelband et al. (1993) previously reported that K⁺ channels in canine PASMCs were inhibited by [Mg²⁺]_i. In this study, the effect of [Mg²⁺]_i on voltage-dependent properties of two types of K_V currents, I_K1 and I_K2 which we have recently described in rat conduit PASMCs (Smirnov et al. 2002), was characterised using the whole cell patch clamp technique.

Male Wistar rats (225-275 g) were humanly killed and PASMCs isolated as described in Smirnov et al. (2002). Experiments were performed at room temperature in the presence of 1 µM paxilline and 10 µM glibenclamide to block BK_Ca and ATP-senstivie K⁺ currents, respectively. Cells were perfused with a pipette solution containing either 0, 5 or 10 mM MgCl₂ to vary [Mg²⁺]_i. Comparison of the maximal whole-cell conductance and steady-state activation (calculated from the I-V relationships) and inactivation (measured at +60 mV after 10 s conditioning depolarisations) for I_K1 and I_K2, which is thought to be carried through K_V1 and K_V2 channel subtypes respectively (Smirnov et al. 2002), showed a differential sensitivity of these two types of K_V currents to variation in [Mg²⁺]_i.

The maximal conductance for I_K1 was significantly decreased from 1.14 ± 0.17 nS/pF (mean ± S.E.M., n = 12, 0 mM MgCl₂) to 0.42 ± 0.06 nS/pF (n = 11, 10 mM MgCl₂, P < 0.0007, unpaired Student’s t test), while that for I_K2, did not change significantly (0.13 ± 0.01 and 0.11 ± 0.01 nS/pF in 12 and 14 PASMCs studied in 0 and 10 mM MgCl₂, respectively). On the other hand, the mid-activation potential (V_a) for I_K2 was significantly shifted to more negative membrane potentials from 4.1 ± 2.8 mV (n = 13, 0 mM MgCl₂) to -8 ± 2.1 mV (n = 14, 10 mM MgCl₂, P < 0.002), while V_a for I_K1 was not affected. However, the opposite effect on the mid-inactivation potential (V_h) was observed, An increase of MgCl₂ concentration in the pipette from 0 to 10 mM shifted the I_K1 inactivation dependency to the left by 26 mV from -23 ± 2 mV (n = 6)to -49 ± 5 (n = 6) respectively (P < 0.008), while V_h for I_K2 was only marginally affected -54 ± 2 mV vs. -59 ± 4 mV in 10 and 6 PASMCs studied in 0 and 10 mM MgCl₂, respectively, P > 0.24).

Our findings demonstrate that two distinct K_V channel types in PASMCs are differentially regulated by [Mg²⁺]_i. A comparison of hypothetical K_V ‘window’ current predicts that I_K1 is more sensitive to changes in [Mg²⁺]_i than I_K2 in the voltage range close to the cell resting membrane potential.

This work was supported by the British Heart Foundation (grant FS/2000013).