Properties of an inwardly rectifying ATP-sensitive K+ channel in the basolateral membrane of renal proximal tubule

Abstract

The potassium conductance of the basolateral membrane (BLM) of proximal tubule cells is a critical regulator of transport since it is the major determinant of the negative cell membrane potential and is necessary for pump-leak coupling to the Na+,K+-ATPase pump. Despite this pivotal physiological role, the properties of this conductance have been incompletely characterized, in part due to difficulty gaining access to the BLM. We have investigated the properties of this BLM K+ conductance in dissociated, polarized Ambystoma proximal tubule cells. Nearly all seals made on Ambystoma cells contained inward rectifier K+ channels (gammaslope, in = 24.5 +/- 0.6 pS, gammachord, out = 3.7 +/- 0.4 pS). The rectification is mediated in part by internal Mg2+. The open probability of the channel increases modestly with hyperpolarization. The inward conducting properties are described by a saturating binding-unbinding model. The channel conducts Tl+ and K+, but there is no significant conductance for Na+, Rb+, Cs+, Li+, NH4+, or Cl-. The channel is inhibited by barium and the sulfonylurea agent glibenclamide, but not by tetraethylammonium. Channel rundown typically occurs in the absence of ATP, but cytosolic addition of 0. 2 mM ATP (or any hydrolyzable nucleoside triphosphate) sustains channel activity indefinitely. Phosphorylation processes alone fail to sustain channel activity. Higher doses of ATP (or other nucleoside triphosphates) reversibly inhibit the channel. The K+ channel opener diazoxide opens the channel in the presence of 0.2 mM ATP, but does not alleviate the inhibition of millimolar doses of ATP. We conclude that this K+ channel is the major ATP-sensitive basolateral K+ conductance in the proximal tubule.

Figure 1

Dissociated Ambystoma proximal tubule cells retain epithelial cell polarity. (A) Light photomicrograph of a single dissociated Ambystoma proximal tubule cell shows distinct apical and basolateral membrane surfaces. The apical surface of these bilobated cells is the smaller lobe with a microvillar brush border. The robust cytoskeleton of these cells includes an actin-rich “waist-band” (arrowheads) between the two membrane domains that is important in retention of epithelial polarity. Giga-ohm seals can be made on both surfaces. Scale bar, 10 μm. (B and C) Scanning electron micrographs showing the sharp transition between the membrane domains and the detailed topology of the apical surface invested with its microvillar brush border, and the basolateral surface with its folds and projections. Scale bars: 10 μm in B, 1 μm in C.

Figure 2

The BLM K⁺ channel appears to be an inward rectifier. (A) Representative current records at various command potentials (−V_pip) from a cell-attached basolateral membrane patch containing at least two K⁺ channels. Note the increase in channel activity with hyperpolarization. The patch pipette contains 95 mM K⁺ (solution d) and the bath is NaCl (solution c). The dashed line represents the all channels closed (leak) current at each potential. Each open channel level is denoted by a dotted line. (B) Current–voltage relation for the BLM K⁺ channel for the conditions described in A. The limiting inward slope conductance is 22.2 ± 1.4 pS (n = 8) and the outward chord conductance 3.5 ± 0.1 pS (between 120 and 180 mV, n = 5). Symbols represent mean (•) ± SEM (bars).

Figure 3

The BLM K⁺ channel is a true inward rectifier. Representative current records at various command potentials (−V_pip) from an inside-out basolateral membrane patch in symmetrical [K] containing at least three K⁺ channels. The patch pipette and bath each contain 95 mM K⁺ (solution d) with 0.2 mM ATP added to the bath. The dashed line represents the all channels closed (leak) current at each potential. Each open channel level is denoted by a dotted line. (B) Effect of [Mg²⁺]_i on the I-V relation of the BLM K⁺ channel. The inward rectification evident in 1 mM [Mg²⁺]_i (○) is relieved when [Mg²⁺]_i is lowered to 200 nM (▪). (inset) Slope conductance–voltage (g-V) relation for the BLM K⁺ channel in 1 mM [Mg²⁺]_i (○) and 200 nM [Mg²⁺]_i (▪). Symbols represent mean ± SEM. The K⁺ channel prefers Tl⁺ over K⁺. The I-V relation from inside-out patches with Tl-acetate (solution g) in the pipette and K-acetate (solution h) in the bath is also shown (▴). The limiting inward slope conductance is 29.0 ± 1.0 pS (n = 4) for Tl⁺ compared with 24.5 ± 0.6 pS (n = 8) for K⁺. (C) Channel activity (nP _o) increases with hyperpolarization. Data from four inside-out membrane patches in symmetrical [K] are plotted. Channel activity at command potentials of −120, −100, −80, −60, and −40 mV was normalized to that at −100 mV for comparison. Solid line is a single exponential fit with a voltage constant of ∼83 mV. Symbols represent mean (•) ± SEM (bars).

Figure 4

The operating surface of the BLM K⁺ channel. (A) The single channel conductance (γ) at a command potential of −100 mV plotted as a function of [K⁺] and fit to the Hill equation according to is the maximal value of the conductance, K _d is the apparent dissociation constant, and n _H is the Hill coefficient. For the chord conductance at −100 mV (▪) and its least-squares fit (solid line), γ_max,slope = 34.3 pS, K _d = 77 mM, and n _H = 0.94. (B) The plot of K _d versus V_pip shows that the apparent binding–unbinding rate of K⁺ to the channel is voltage dependent. The data (○) were fit with a single exponential (solid line) as follows: K _d = 73 + 141 · exp(−V_pip/30).

Figure 5

The BLM K⁺ conductance is highly selective for K⁺. The cationic selectivity of the BLM is demonstrated in this outside-out patch with KCl (solution d) + 0.2 mM ATP in the pipette and Cl-salt of the test cation in the bath. The running average of current at −40 mV shows that cationic selectivity is K⁺ >> Rb⁺ ≈ Cs⁺ ≈ NH₄ ⁺ > Na⁺ ≈ Li⁺. The dashed line is the zero-current line, and the dotted line is the all channels closed (leak) current at a command potential of −40 mV. The voltage protocol is indicated below the current data.

Figure 6

Kinetics of the BLM K⁺ channel. (A) Open-time histogram for the BLM K⁺ channel in a cell-attached patch at −60 mV with the time intervals logarithmically binned (Sigworth and Sine, 1987). The data were fitted with two probability density functions (dashed lines) to give the overall fit (solid line), yielding time constants of τ_o1 = 0.78 ms (78%) and τ_o2 = 4.7 ms (22%). Based on the bandwidth of the recording system, the data and the fit were cutoff at 500 μs (vertical dotted line). (B) Closed-time histogram for the BLM K⁺ channel in a cell-attached patch at −60 mV with the time intervals logarithmically binned. The data were fitted with two probability density functions (dashed lines) to give the overall fit (solid line), yielding time constants of τ_c1 = 1.27 ms (74%) and τ_c2 = 397 ms (26%). Based on the bandwidth of the recording system, the data and the fit were cutoff at 500 μs (dotted vertical line).

Figure 7

Inhibitors of the BLM K⁺ channel. (A) Dose– response curve for inhibition of the BLM K⁺ channel by external Ba²⁺. Relative nP _o (nP _o/nP _o,control) determined from outside-out patches at a command potential of −40 mV with KCl in the pipette and bath (solution d) is plotted versus [Ba²⁺]. The data at −40 mV were fitted with the Hill equation (solid curve), yielding a K _i = 460 μM and n _H = 0.90. The inhibition by Ba²⁺ is fully reversible. (B) Glibenclamide inhibits the BLM K_ATP channel. (top) A running average (window width 16 ms) of current versus time. Glibenclamide (500 μM) is added to the bath where indicated. (bottom) Sample traces from a representative experiment at −60 mV with KCl in the pipette and bath (solution d). Exposure of this inside-out patch to 500 μM glibenclamide reversibly decreases nP _o by ∼50%. The dashed line denotes the all channels closed (leak) current level and the dotted lines indicate each open channel level.

Figure 8

Nucleotides reversibly inhibit the BLM K⁺ channel. (A) ATP_i reversibly inhibits the BLM K⁺ channel in inside-out patches. In the experiment depicted, up to 13 channel open levels are seen under control conditions (top) at −60 mV with KCl (solution d) in the pipette and NaCl (solution c) plus 0.2 mM ATP in the bath. (middle) Addition of 5 mM ATP to the cytoplasmic side almost completely blocks channel activity (98% decrease) with only rare openings to one open level. (bottom) The inhibition is readily reversed upon returning to 0.2 mM ATP_i. (B) Dose–response curve for inhibition by ATP_i. The inhibitory effect of ATP_i was determined in inside-out patches under the conditions described in A. Relative nP _o (nP _o/nP _o,control) is plotted versus [ATP]_i. The data (•) was fitted with the Hill equation (solid line) yielding a K _i ∼ 2.4 mM and n _H = 3.95. Adenosine nucleotides and nucleoside triphosphates (each at 5 mM) inhibit the BLM K⁺ channel. The inhibitory effect of ATP, ADP, AMP, and adenosine was determined under the conditions described in A. Values plotted are average (nP _o,test/nP _o,control) SEM (n = 3–9). The rank order of inhibition is ATP (93.3%) > ADP (65.6%) > AMP (38.7%). Adenosine has no inhibitory effect on nP _o of the BLM K⁺ channel. Although all the nucleoside triphosphates inhibit the channel, ATP exerts a significantly stronger block than the other NTPs tested (P < 0.02). There is no significant difference among the other NTPs.

Figure 9

ATP but not ATP-γS prevents and rescues channel rundown. Hydrolyzable nucleoside triphosphates prevent and rescue BLM K⁺ channel rundown. A running average (current versus time, window width 768 ms) of a representative experiment is shown. The pipette is KCl (solution d), the bath is NaCl (solution c), the command potential is −60 mV. Upon excision in a nucleotide-free bath, channel activity decreases (channel rundown). After the addition of 0.2 mM ATP to the bath, channel activity slowly recovers. The addition of 0.2 mM ATP-γS (a poorly hydrolyzable ATP analogue) in the continued presence of 0.2 mM ATP has no effect. However, when ATP is removed, ATP-γS is not able to support channel activity, which rapidly declines and runs down. Readdition of ATP leads to full recovery of channel activity. Single-channel traces showing that ATP-γS has an inhibitory effect on K_ATP channel activity in excised inside-out BLM patches. When compared with control conditions (top), the addition of ATP-γS (middle) reduces nP _o. This inhibition is reversible as long as the exposure to ATP-γS is not prolonged (bottom). (C) Removal of Mg²⁺ does not prevent channel rundown in an ATP-free bath. The top panel shows that rundown of the BLM K_ATP channel upon excision into an ATP-free bath proceeds despite removal of bath Mg²⁺. Representative traces from the regions marked by α, β, and γ are shown at bottom.

Figure 10

The K channel opener diazoxide activates the BLM K⁺ channel. Diazoxide activates the BLM K⁺ channel in the presence of ATP. In the experiment depicted, the pipette contains KCl (solution d), the bath contains NaCl (solution c), and the command potential is −80 mV. (top) A running average (window width 16 ms) of current versus time. (bottom) Sample current traces from the same experiment. With diazoxide alone, the channel opens infrequently (nP _o = 0.15) and no more than two channels are open at a time. Diazoxide combined with 0.2 mM ATP, however, promptly increases channel activity and within 5 min, up to 13 simultaneously open channels are evident (nP _o = 5.27). The all channels closed level (dashed line) and open channel levels (dotted lines) are indicated.

Figure 11

Equivalent electrical cell/circuit. Adjacent proximal tubule cells are outlined as dashed lines, and the equivalent circuit discussed in the text is diagrammed on the left. The basolateral membrane potential is denoted by voltage source V_bl. The current produced by the Na,K-ATPase pump, I_net ^pump is given by I_Na ^pump + I_K ^pump. Basolateral ion channel currents are shown as their Thévenin equivalents with their conductances, G_K ^bl and G_Cl ^bl, represented by the equivalent resistor, in series with their Nernst potentials, E_K and E_Cl, respectively. The paracellular current, I_total ^shunt is represented by I_Na ^shunt + I_Cl ^shunt. R_A is the resistance of the apical membrane. Kirchoff's current law at Node A is given by Eq. 11.