Divalent cation permeability and blockade of Ca2+-permeant non-selective cation channels in rat adrenal zona glomerulosa cells

Abstract

1. The effects of the divalent cations Ca2+, Mg2+ and Ni2+ on unitary Na+ currents through receptor-regulated non-selective cation channels were studied in inside-out and cell-attached patches from rat adrenal zona glomerulosa cells. 2. External Ca2+ caused a concentration-dependent and voltage-independent inhibition of inward Na+ current, exhibiting an IC50 of 1.4 mM. The channel was also Ca2+ permeant and external Ca2+ shifted the reversal potential as expected for a channel exhibiting a constant Ca2+ : Na+ permeability ratio near to 4. 3. External and internal 2 mM Mg2+ caused voltage-dependent inhibition of inward and outward Na+ current, respectively. Modelling Mg2+ as an impermeant fast open channel blocker indicated that external Mg2+ blocked the pore at a single site exhibiting a zero voltage Kd of 5.1 mM for Mg2+ and located 19 % of the distance through the transmembrane electric field from the external surface. Internal Mg2+ blocked the pore at a second site exhibiting a Kd of 1.7 mM for Mg2+ and located 36% of the distance through the transmembrane electric field from the cytosolic surface. 4. External Ni2+ caused a voltage- and concentration-dependent slow blockade of inward Na+ current. Modelling Ni2+ as an impermeant slow open channel blocker indicated that Ni2+ blocked the pore at a single site exhibiting a Kd of 1.09 mM for Ni2+ and located 13.7% of the distance through the transmembrane electric field from the external surface. 5. External 2 mM Mg2+ increased the Kd for external Ni2+ binding to 1.27 mM, consistent with competition for a single binding site. Changing ionic strength did not substantially affect Ni2+ blockade indicating the absence of surface potential under physiological ionic conditions. 6. It is concluded that at least two divalent cation binding sites, separated by a high free energy barrier (the selectivity filter), are located in the pore and contribute to Ca2+ selectivity and permeability of the channel.

Scheme 1

Figure 1. Effects of extracellular Ca²⁺ on unitary Na⁺ currents

A, current traces illustrating representative single channel events measured at -80 mV for several Ca²⁺ concentrations (indicated at the left of each trace). The closed channel current level is indicated by c at the left of the traces; each trace was taken from a different patch. The external (pipette) solution composition for each Ca²⁺ concentration is described in Table 1; the bath (cytosolic face) saline was nominally Ca²⁺ free (solution 15). Current records were low-pass filtered at 500 Hz for display. B, I-V relationships are plotted from data obtained with 0.1 mM Ca²⁺ (▴, n= 5), 1.0 mM Ca²⁺ (•, n= 5) and 27 mM Ca²⁺ (▪, n= 3) pipette solutions. Symbols indicate the mean (±s.e.m.) unitary current amplitude; error bars are not shown for errors less than symbol size. Smooth curves represent polynomial regression fits to each data set. C, fractional inward slope conductance is plotted as a function of Ca²⁺ concentration: γ_B represents mean conductance at each Ca²⁺ concentration and γ the mean maximal conductance in the absence of external Ca²⁺. The smooth curve represents the best fit to a Langmuir isotherm, indicating an IC₅₀ of 1.4 mM for Ca²⁺ inhibition. D, V_rev of unitary currents is plotted as a function of external Ca²⁺ activity. Symbols represent the mean (±s.e.m.) of 3-5 determinations at each concentration; symbols lacking error bars represent the means of 2 determinations. The smooth curve represents the change in V_rev predicted from the modified GHK equation (eqn (1)) assuming a constant Ca²⁺:Na⁺ permeability ratio of 4.

Figure 2. Voltage-dependent inhibition of unitary Na⁺ currents by external and internal 2 mM Mg²⁺

A, current traces illustrating representative single channel events obtained from inside-out patches at membrane potentials of 60 and -60 mV in the absence of both external and internal Mg²⁺ (left traces), with 2 mM internal Mg²⁺ (middle traces), and with 2 mM external Mg²⁺ (right traces). The patch recording for 2 mM external Mg²⁺ exhibited two open channel current levels. The closed channel current level is indicated by c at the left of each trace. Subscripts o and i designate external and internal Mg²⁺, respectively. Pipettes contained either solution 5 (0 Mg²⁺) or solution 14 (2 mM Mg²⁺) and the bath contained either solution 13 (0 Mg²⁺) or solution 15 (2 mM Mg²⁺) (Table 1). Current traces were low-pass filtered at 500 Hz for display. B, I-V relationship for single channel currents measured for each Mg²⁺ condition described in A. Symbols represent the mean (±s.e.m.) of unitary current amplitude from 3 or 4 separate patches in each configuration; error bars were omitted when error was less than symbol size. Lines represent polynomial least-squares regression fits to each data set. C, linearized logarithmic plot (eqn (2)) of voltage-dependent inhibition of unitary Na⁺ curent amplitude by extracellular and cytoplasmic 2 mM Mg²⁺. Symbols represent the reduction in current (i_o/i) - 1, where i_o represents the mean maximal current in the absence of Mg²⁺ and i represents the mean current recorded with 2 mM Mg²⁺_o/0 mM Mg²⁺_i (▪) or 0 mM Mg²⁺_o/2 mM Mg²⁺_i (▴); data from B. Curves represent the linear regression fit to the data.

Figure 3. External Ni²⁺-induced fluctuations in unitary Na⁺ currents

Current traces illustrating representative single channel events recorded from cell-attached patches at -60 mV (upper traces) and -120 mV (lower traces) in the absence of Ni²⁺ (control traces, left), in 0.05 mM Ni²⁺ (middle traces), and in 0.5 mM Ni²⁺ (right traces). The pipettes contained solution 5 (Table 1) with Ni²⁺ added hyperosmotically. Resting membrane potential was set to 0 mV with a high K⁺ saline (solution 3). Expanded time scale segments of current traces recorded at -120 mV are shown to illustrate Ni²⁺-induced blocking events. The closed and blocked channel current levels are indicated by c at the left of each trace. Current traces were low-pass filtered at 2 kHz.

Figure 4. Open and blocked state dwell time histograms for 0.1 mM Ni²⁺ blockade of unitary Na⁺ currents

A, blocked state dwell time histograms measured at -60 mV (left graph) and -120 mV (right graph) from a single cell-attached patch with 0.1 mM Ni²⁺ in the pipette solution (solution 5, Table 1). B, open state dwell time histograms measured at -60 mV (left graph) and -120 mV (right graph) from the same patch as in A. The resting membrane potential was set to 0 mV with a high extracellular K⁺ saline (solution 3). The smooth curves in each graph represent the maximum likelihood fit of the data to a single exponential relaxation. The time constants of the fitted exponential relaxations for τ_b (A) and τ_o (B), and membrane potentials (V_m) are indicated in the inset of each histogram.

Figure 5. Effects of membrane potential and Ni²⁺ concentration on τ_b

Mean blocked times measured in the presence of 0.05 mM (•), 0.2 mM (▪) and 0.5 mM (▴) Ni²⁺ at -60, -80, -100 and -120 mV; symbols are offset to illustrate clearly standard deviations of the data. Symbols represent the mean (±s.d.) of 3-4 separate patches at each voltage and Ni²⁺ concentration; symbols lacking error bars represent the mean value from two separate patches.

Figure 6. Determination of K_d(0) and location (δ) of the Ni²⁺ binding site in the absence and presence of external Mg²⁺

A, the reciprocal of τ_o for Ni²⁺ blockade in the absence of external Mg²⁺ (pipette solution 5) is plotted as a function of external Ni²⁺ concentration. Data are shown for membrane potentials of -120 mV (•), -100 mV (▪), -80 mV (▴) and -60 mV (▾). Values of τ_o represent single exponential maximum likelihood fits to pooled data from 2-4 patches at each voltage and Ni²⁺ concentration; the number of pooled blocking events for each τ_o ranged between 300 and 1400. Values of τ_o were corrected for missed blocking events. Error bars indicate the standard deviation and are not shown for values less than symbol size. B, reciprocal τ_o values for Ni²⁺ blockade, measured in the presence of 2 mM Mg²⁺ (solution 14), are plotted as a function of Ni²⁺ concentration. Data are shown for membrane potentials of -120 mV (•), -80 mV (▴), -60 mV (▾) and -40 mV (♦). Values of τ_o and error bars are as described in A; each value represents fits to pooled data from 2-4 patches containing 500-700 blocking events. C, voltage-dependent association rate constants, k₁(V), determined from the slope of the regression lines in A and B for Ni²⁺ blockade in the absence (•) and presence (▪) of 2 mM external Mg²⁺, respectively. Error bars represent the standard error of the regression coefficients. D, the voltage dependence of the dissociation constant, K_d, for Ni²⁺ binding in the absence and presence of 2 mM Mg²⁺. Open symbols represent K_d(0) determined from extrapolation of the linear regression fits to each data set (continuous curves). Error bars represent the error extrapolated from the standard errors for k₁(V). For all figures, smooth curves represent the linear regression fits to the data.

Figure 7. Effects of ionic strength on Na⁺ conductance and Ni²⁺ blockade

A, effect of decreasing external NaCl concentration and ionic strength on the I-V relationships of unitary currents obtained from inside-out patches; pipette contained 150 mM NaCl (•), 75 mM NaCl (▴) and 37.5 mM NaCl (▾). Ionic strength was decreased by substitution of glucose for NaCl (mixing solutions 5 and 18, Table 1); the bath contained solution 15. Symbols represent the mean (±s.d.) current amplitude from 3-5 patches at each NaCl concentration; errors less than symbol size are not shown. Smooth curves represent 3rd order or 2nd order (75 and 37.5 mM NaCl) regression fits to the data. B, effect of decreasing external NaCl concentration at constant ionic strength on the I-V relationship; pipette contained 150 mM NaCl (•), 112.5 mM NaCl (▪) and 75 mM NaCl (▴). Ionic strength was maintained by substitution of NMDG for NaCl (mixing of solutions 5 and 19); the bath contained solution 15. Symbols represent the mean (±s.d.) from 3-6 patches at each NaCl concentration. Smooth curves represent 3rd order or linear (75 mM NaCl) regression fits to the data. C, Na⁺ concentration dependence of inward slope conductance for variable ionic strength (glucose substitution, ○) and constant ionic strength (NMDG substitution, □) data from A and B. Filled circles represent channel conductance in 150 and 225 mM NaCl (pipette solution 16/bath solution 17). Symbols represent the mean (±s.d.) from 3-6 patches for each ionic condition. Continuous and dashed curves represent fits to the Michaelis-Menton equation for the glucose-substituted and NMDG-substituted data, respectively. D, effect of ionic strength on the voltage dependence of reciprocal τ_o for 0.1 mM Ni²⁺ blockade of unitary Na⁺ currents in cell-attached patches: 225 mM NaCl (pipette solution 16, ○), 150 mM NaCl (solution 5, □), or 112.5 mM NaCl (mixing solutions 5 and 18, ▵). Symbols represent reciprocal of corrected τ_o values determined from data pooled from 3 separate patches containing between 400 and 1500 blocking events at each ionic strength and membrane potential. Standard deviations were less than symbol size.