Properties of single K+ channels in the basolateral membrane of rabbit proximal straight tubules

The basolateral membrane of rabbit straight proximal tubules, which were cannulated and perfused on one side, was investigated with the patch clamp technique. Properties of inward and outward directed single K⁺ channel currents were studied in cell-attached and insideout oriented cell-excised membrane patches. In cell-attached patches with NaCl Ringer solution both in pipette and bath, outward K⁺ currents could be detected after depolarization of the membrane patch by about 20–30 mV. The current-voltage (i/V) relationship could be fitted by the Goldman-Hodgkin-Katz (GHK) current equation, with the assumption that these channels were mainly permeable for K⁺ ions. A permeability coefficientP _K of (0.17±0.04) · 10^–12 cm³/s was obtained, the single channel slope conductance at infinite positive potentialg(V ) was 50±12 pS and the single channel conductance at the membrane resting potentialg(V _bl) was 12±3 pS (n=4). In cell-excised patches, with NaCl in the pipette and KCl in the bath, the data could also be fitted to the GHK equation and yieldedP _K = (0.1 ±0.01) ·10^–12 cm³/s,g(V ) = 40 ± 4 pS andg(V _bl) = 7 ± 1 pS (n=8). In cell-attached patches with KCl in the pipette and NaCl in the bath, inward K⁺ channels occurred at clamp potentials 60 mV, whereas outward K⁺ channel current was detected at more positive voltages. The current-voltage curves showed slight inward rectification. The single channel conductance, obtained from the linear part of the i/V curve by linear regression, was 46±3 pS and the reversal potential was 59±6 mV (n=9). In cell-excised patches with KCl in the pipette and NaCl in the bath, inward directed K⁺ channel currents could again be described by the GHK equation. The single channel parameters were similar to those recorded for outward K⁺ currents (see above). In inside-out oriented cell-excised patches with NaCl in the pipette and KCl in the bath, reducing bath (i.e. cytosolic) Ca²⁺ concentration from 10^–6 mol/l to less than 10^–9 mol/l did not affect the open state probability of single channel currents. These results demonstrate that the observed channels are permeable for K⁺ ions in both directions and that these basolateral K⁺ channels in rabbit proximal straight tubule are not directly dependent on Ca²⁺ ions.     

K+ channels in the basolateral membrane of rat cortical collecting duct

Impalement studies in isolated perfused cortical collecting ducts (CCD) of rats have shown that the basolateral membrane possesses a K⁺ conductive pathway. In the present study this pathway was investigated at the single-channel level using the patch-clamp technique. Patch-clamp recordings were obtained from enzymatically isolated CCD segments and freshly isolated CCD cells with the conventional cell-free, cell-attached and the cell-attached nystatin method. Two K⁺ channels were found which were highly active on the cell with a conductance of 67±5 pS (n=18) and 148±4 pS (n=21) with 145 mmol/l K⁺ in the pipette. In excised patches the first channel had a conductance of 28±2 pS (n=15), whereas the second one had a conductance of 85±1 pS (n=53) at 0 mV clamp voltage with 145 mmol/l K⁺ on one side and 3.6 mmol/l K⁺ on the other side of the membrane. So far it has not been possible to characterize the smaller channel further. Excised, and with symmetrical K⁺ concentrations of 145 mmol/l, the intermediate channel had a linear conductance of 198±19 pS (n=5). After excision in the inside-out configuration the open probability (P _o) of this channel was low (0.18±0.05, n=13) whereas in the outside-out configuration this channel had a threefold higher P _o (0.57±0.04, n=12). Several inhibitors were tested in excised membranes. Ba²⁺ (1 mmol/l), tetraethylammonium (TEA⁺, 10 mmol/l) and verapamil (0.1 mmol/l) all blocked this channel reversibly. Furthermore P _o was reversibly reduced by 10 nmol/l charybdotoxin (outside-out). This K⁺ channel of the basolateral membrane was regulated by cellular pH. P _o was reduced to 26±3% at pH 6.5 (n=6) and increased to 216±18% at pH 8.5 (n=7) compared to pH 7.4. Half-maximal inhibition was reached at pH 7.0. The channel had its highest P _o at a Ca²⁺ activity of less than 10^–8 mol/l (n=13). Increasing the Ca²⁺ activity to 1 mmol/l on the cytosolic side of the membrane resulted in a reduction of P _o to 13±3% (n=11). Half-maximal inhibition was reached at a Ca²⁺ activity of 10^–5 mol/l. The high activity of both K⁺ channels of the basolateral membrane on the cell indicates that they may serve for K⁺ recirculation across the basolateral membrane.     

Comparison of Ca2+-dependent K+ channels in the membrane of smooth muscle cells isolated from adult and foetal human aorta

Ca²⁺-activated K⁺ ionic currents in the membrane of cultured smooth muscle cells isolated from foetal and adult human aorta were studied using whole cell and single-channel patch-clamp techniques. Whole cell currents in adult smooth muscle cells were 3–8 times larger than in foetal cells of similar sizes. The elementary conductance and ionic selectivity of single Ca²⁺-activated K⁺ were identical for both types of cells. Channel openings occurred in burst, the duration of which was 3–5-fold longer in adult than in foetal cells. The voltage dependency of the channel activating mechanism and the dependency of the mean open time on the Ca²⁺ concentration on the inner side of the membrane were similar for both types of cells. These results suggest that the main reason for the increase in potassium conductance during development is an alteration in the open time of the Ca²⁺-activated K⁺ channels.     

Effects of dopamine on ion transport across the rat distal colon

Dopamine (5·10^–6–5·10^–4 M) induced a concentration-dependent decrease in short-circuit current (I_sc) across the rat distal colon. This response was preceded by a transient and inconsistent increase in I_sc. The -adrenoceptor blocker phentolamine and the inhibitors of dopamine-2-like (D₂-like) receptors L-741,626 and L-745,870 inhibited the dopamine response, suggesting a contribution of adrenergic and dopaminergic receptors. The decrease in I_sc evoked by dopamine was inhibited by bumetanide, an inhibitor of the basolateral Na⁺-K⁺-2 Cl^– cotransporter responsible for the uptake of K⁺, and by quinine, a blocker of apical K⁺ channels, indicating that stimulation of K⁺ secretion contributes to the measured change in I_sc. In patch-clamp experiments dopamine hyperpolarized the membrane and increased cellular K⁺ current. This response was not concomitant with a change in the intracellular [Ca²⁺] as demonstrated in parallel fura-2 experiments. These results demonstrate that dopamine, like other catecholamines, stimulates colonic K⁺ secretion.     

Small-conductance Ca2+-activated K+ channels in bovine chromaffin cells

Simultaneous whole-cell patch-clamp and fura-2 fluorescence [Ca²⁺]_i measurements were used to characterize Ca²⁺-activated K⁺ currents in cultured bovine chromaffin cells. Extracellular application of histamine (10 M) induced a rise of [Ca²⁺]_i concomitantly with an outward current at holding potentials positive to –80 mV. The activation of the current reflected an increase in conductance, which did not depend on membrane potential in the range –80 mV to –40 mV. Increasing the extracellular K⁺ concentration to 20 mM at the holding potential of –78 mV was associated with inwardly directed currents during the [Ca²⁺]_i elevations induced either by histamine (10 M) or short voltage-clamp depolarizations. The current reversal potential was close to the K⁺ equilibrium potential, being a function of external K⁺ concentration. Current fluctuation analysis suggested a unit conductance of 3–5 pS for the channel that underlies this K⁺ current. The current could be blocked by apamin (1 M). Whole-cell current-clamp recordings snowed that histamine (10 M) application caused a transient hyperpolarization, which evolved in parallel with the [Ca²⁺]_i changes. It is proposed that a small-conductance Ca²⁺-activated K⁺ channel is present in the membrane of bovine chromaffin cells and may be involved in regulating catecholamine secretion by the adrenal glands of various species.     

Differential distribution of two Ca2+-dependent and -independent K+ channels throughout receptive and basolateral membranes of bullfrog taste cells

We could identify two types of K⁺ channels, of 80 and 40 pS conductance, respectively, in the bullfrog taste cell membrane using excised and cell-attached configurations of the patch-clamp technique. The taste cell membrane could be divided into four membrane parts — receptive area, apical process, cell body and proximal process. The 80-pS K⁺ channels were dependent on voltage and Ca²⁺ and were located exclusively on the receptive membrane and the apical process membrane. The 40-pS K⁺ channels were independent of voltage and Ca²⁺. The open probability of 40-pS K⁺ channels was decreased by the simultaneous presence of cyclic adenosine monophosphate (cAMP) and adenosine triphosphate (ATP), and the suppressive effect was antagonized by protein kinase inhibitor (PKI). Although 40-pS K⁺ channels were found in a high density on the receptive and apical process membranes, the channels also were present in the other two parts of the taste cell membrane. These results suggest that the two different types of K⁺ channel in the bullfrog taste cells may play different roles in gustatory transduction.     

Neuropeptide Y inhibits Ca2+-activated K+ channels in vascular smooth muscle cells from the rat tail artery

The effects of neuropeptide Y (NPY) on the Ca²⁺-activated K⁺ channel in smooth muscle cells from the rat tail artery were studied by whole-cell and single-channel patch-clamp recording techniques. In the presence of nifedipine (1 M), whole-cell outward currents through Ca²⁺-activated K⁺ channels were inhibited by NPY in a dose-dependent manner from 20 to 200 nM. A maximum inhibition to about 48% of the control current could be achieved. Recordings from outside-out patches showed that the open probability of Ca²⁺-activated K⁺ channels were similarly inhibited by NPY. At 200 nM NPY, the open probability was reduced to about 36% of the control value. NPY did not affect the open times or current amplitude, but increased significantly the short (from 0.49 to 0.58 ms) and long (from 441 to 728 ms) closed times. Inhibition of Ca²⁺-activated K⁺ channels by NPY may contribute to its excitatory action on vascular smooth muscle cells.     

L-alanine evokes opening of single Ca2+-activated K+ channels in rat liver cells

Cellular uptake of neutral amino acids via Na⁺ cotransporters is known to be associated with an increased membrane K⁺ conductance mediated by an unknown mechanism that is essential for avoiding excessive cell swelling. We now demonstrate by patch-clamp single-channel current recording that exposure of rat liver cells to L-alanine, but not the poorly transported D-stereoisomer, evokes opening of single K⁺ channels and that this effect is reversible upon removal of the amino acid. The nature of the conductance pathways opened in the intact cell by L-alanine has been investigated in cell-free excised membrane patches where it can be shown that the K⁺-selective channels are opened by Ca²⁺ acting from the inside of the membrane at a concentration as low as 0.1 M.     

Single Ca2+-activated K+ channels in excised membrane patches of hamster oocytes

The properties of the Ca²⁺-activated K⁺ channel in unfertilized hamster oocytes were investigated at the single-channel level using inside-out excised membrane patches. The results indicate a new type of Ca²⁺-activated K⁺ channel which has the following characteristics: (1) single-channel conductance of 40–85 pS for outward currents in symmetrical K⁺ (150 mM) solutions, (2) inward currents of smaller conductance (10–50 pS) than outward currents, i.e. the channel is outwardly rectified in symmetrical K⁺ solutions, (3) channel activity dependent on the internal concentration of free Ca⁺ and the membrane potential, (4) modification of the channel activity by internal adenosine 5 diphosphate (0.1 mM) producing a high open probability regardless of membrane potential.     

Tetraethylammonium-insetisitive,Ca2+-activated whole-cell K+ currents in rat submandibular acinar cells

Using whole-cell patch-clamp techniques, we demonstrate, for the first time, that rat submandibular acinar cells contain a tetraethylammonium (TEA)-insensitive, Ca²⁺-activated K⁺ conductance which is not attributable to large conductance, voltage-sensitive, Ca²⁺-dependent K⁺ channels (maxi-K⁺ channels). Taken together with our recent K⁺ efflux and fluid secretion studies in intact rat submandibular gland, we postulate that the K⁺ conductance reported here may be involved in the basolateral K⁺ efflux pathway activated by cytosolic Ca²⁺ concentration during secretion by this gland.     

Basolateral K+ efflux is largely independent of maxi-K+ channels in rat submandibular glands during secretion

The involvement of large-conductance, voltage- and Ca²⁺-activated K⁺ channels (maxi-K⁺ channels) in basolateral Ca²⁺-dependent K⁺-efflux pathways and fluid secretion by the rat submandibular gland was investigated. Basolateral K⁺ efflux was monitored by measuring the change in K⁺ concentration in the perfusate collected from the vein of the isolated, perfused rat submandibular gland every 30 s. Under conditions in which the Na⁺/K⁺-ATPase and Na⁺-K⁺-2Cl^– cotransporter were inhibited by ouabain (1 mmol/l) and bumeta-nide (50 mol/l) respectively, continuous stimulation with acetylcholine (ACh) (1 mol/l) caused a transient large net K⁺ efflux, followed by a smaller K⁺ efflux, which gradually returned to the basal level within 10 min. These two components of the K⁺ efflux appear to be dependent on an increase in cytosolic Ca²⁺ concentration. The initial transient K⁺ efflux was not affected by charybdotoxin (100 nmol/l) or tetraethylammonium (TEA) (5 mmol/l) but the smaller second component was strongly and reversibly inhibited by charybdotoxin (100 nmol/l) and TEA (0.1 and 5 mmol/l). The initial K⁺ efflux transient induced by ACh was inhibited by quinine (0.1–3 mmol/l), quinidine (1–3 mmol/l) and Ba²⁺ (5 mmol/l), but not by verapamil (0.1 mmol/l), lidocaine (1 mmol/l), 4-aminopyridine (1 mmol/l) or apamin (1 mol/l). Ca²⁺-dependent transient large K⁺ effluxes induced by substance P (0.01 mol/l) and A23187 (3 mol/l) were not inhibited by TEA (5 mmol/l or 10 mmol/l). A23187 (3 mol/l) evoked a biphasic fluid-secretory response, which was not inhibited by TEA (5 mmol/l). Patch-clamp studies confirmed that the whole-cell outward K⁺ current attributable to maxi-K⁺ channels obtained from rat submandibular endpiece cells was strongly inhibited by the addition of TEA (1–10 mmol/l) to the bath. It is concluded that maxi-K⁺ channels are not responsible for the major part of the Ca²⁺-dependent basolateral K⁺ efflux and fluid secretion by the rat submandibular gland.     

Ca2+-activated K+ channels in airway smooth muscle are inhibited by cytoplasmic adenosine triphosphate

Large-conductance Ca²⁺-activated K⁺ channels were studied in membranes of cultured rabbit airway smooth muscle cells, using the patch-clamp technique. In cell-attached recordings, channel openings were rare and occurred only at very positive potentials. Bradykinin (10 M), an agonist which releases Ca²⁺ from the sarcoplasmic reticulum, transiently increased channel activity. The metabolic blocker 2,4-dinitrophenol (20 M), which lowers cellular adenosine triphosphate (ATP) levels, induced a sustained increase of channel activity in cell-attached patches. In excised patches, these channels had a slope conductance of 155 pS at 0 mV, were activated by depolarization and by increasing the Ca²⁺ concentration at the cytoplasmic side above 10^–7 mol/l. ATP, applied to the cytoplasmic side of the patches, dose-dependently decreased the channel's open-state probability. An inhibition constant (K _i) of 0.2 mmol/l was found for the ATP-induced inhibition. ATP reduced the Ca²⁺ sensitivity of the channel, shifting the Ca²⁺ activation curve to the right and additionally reducing its steepness. Our results demonstrate that cytoplasmic ATP inhibits a large-conductance Ca²⁺-activated K⁺ channel in airway smooth muscle. This ATP modulation of Ca²⁺-activated K⁺ channels might serve as an important mechanism linking energy status and the contractile state of the cells.     

ATP-Dependent inhibition of Ca2+-activated K+ channels in vascular smooth muscle cells by neuropeptide Y

Neuropeptide Y(NPY) inhibits Ca²⁺-activated K⁺ channels reversibly in vascular smooth muscle cells from the rat tail artery. NPY (200 M) had no effect in the absence of intracellular adenosine 5triphosphate (ATP) and when the metabolic poison cyanide-M-chlorophenyl hydrozone (10 M) was included in the intracellular pipette solution. NPY was also not effective when ATP was substituted by the non-hydrolysable ATP analogue adenosine 5-[, -methylene]-triphosphate (AMP-PCP). NPY inhibited Ca²⁺-activated K⁺ channel activity when ATP was replaced by adenosine 5-O-(3-thiotriphosphate) (ATP [-S]) and the inhibition was not readily reversed upon washing. Protein kinase inhibitor (1 M), a specific inhibitor of adenosine 3, 5-cyclic monophosphatedependent protein kinase, had no significant effect on the inhibitory action of NPY. The effect of NPY on single-channel activity was inhibited by the tyrosine kinase inhibitor genistein (10 M) but not by daidzein, an inactive analogue of genistein. These observations suggest that the inhibition by NPY of Ca²⁺-activated K⁺ channels is mediated by ATP-dependent phosphorylation. The inhibitory effect of NPY was antagonized by the tyrosine kinase inhibitor genistein.     

Potassium transport in epithelial cells isolated from small intestine of the chicken

The transport of potassium has been studied in epithelial cells isolated from chicken small intestine using⁸⁶Rb as a tracer for K⁺. (i) The uptake studies revealed that about 60% of the total K⁺ net flux is inhibited by ouabain and therefore mediated by the Na⁺–K⁺-ATPase. About 20% of the ouabain-insensitive K⁺ net influx was inhibited by furosemide, bumetanide and by either Na⁺ or Cl^– removal from the incubation solution, suggesting that a Na⁺/Cl^–/K⁺ cotransport system might be present in chicken enterocytes. (ii) The efflux of K⁺ was measured from cells preloaded with⁸⁶Rb. K⁺ efflux was inhibited by Ba²⁺, quinine and verapamil; it was stimulated by A23187, and it was unaffected by 3,4-diaminopyridine. Apamin, that has no effect on basal rates of K⁺ efflux, abolished the effect of A23187. These findings suggest that K⁺ efflux appears to occur through Ca²⁺-activated K⁺ channels.     

Ca2+-activated K+-channels in the nuclear envelope isolated from single pancreatic acinar cells

The patch-clamp techniques are applied to the outer membrane of the nuclear envelope isolated from rat pancreatic acinar cells. The nucleus identified under an inverted microscope was removed by cell surgery from enzymatically dispersed single cells. All the patch-clamp techniques, in situ, excised, and whole-material recordings were applied to the envelope. We have found voltage- and Ca²⁺-activated K⁺-channels with an unitary conductance of 200 pS in the outer membrane. The channels are activated by lumen positive potentials and by an increase in luminal Ca²⁺ concentration. They may play a role for controlling Ca²⁺-release from the lumen of the nuclear envelope (endoplasmic reticulum) to the nucleoplasm and the perinuclear cytoplasm.     

Potassium channels in the basolateral membrane of the rectal gland ofSqualus acanthias

The present study examines the influences of pH and Ca²⁺ and several putative inhibitors on the basolateral K⁺ channel of the rectal gland ofSqualus acanthias. Excised membrane patches were examined using the patch clamp technique. It is shown that reduction of the calcium activity on the cytosolic side to less than 10^–9 mol/l has no detectable inhibitory effect on this channel. Conversely, increase in calcium activity to some 10^–3 mol/l reduced the activity of this channel. Variations in cytosolic pH had only a moderate effect on the current amplitude: alkalosis by one pH unit increased and acidosis reduced the single current amplitude by some 15%. Several inhibitors were tested in excised patches when added to the cytosolic side. Ba²⁺ (5·10^–3 mol/l), quinine (10^–3 mol/l), quinidine (10^–4 mol/l), lidocaine (1 mmol/l), tetraethylammonium (10 mmol/l), Cs⁺ (10 mmol/l), and Rb⁺ (20 mmol/l) all blocked this K⁺ channel reversibly. We conclude that the basolateral K⁺ channel of the rectal gland is distinct from other epithelial K⁺ channels inasmuch as it is not stimulated by Ca²⁺ directly, but that it is qualitatively similar to many other known K⁺ channels with respect to its sensitivity towards blockers.This study was supported by Deutsche Forschungsgemeinschaft Gr 480/8 and by NSF and NIH grants to the Mount Desert Island Biological Laboratory     

Variations of membrane cholesterol alter the kinetics of Ca2+-dependent K+ channels and membrane fluidity in vascular smooth muscle cells

The patch-clamp technique and fluorescence polarization analysis were used to study the dependence of Ca²⁺-dependent K⁺ channel kinetics and membrane fluidity on cholesterol (CHS) levels in the plasma membranes of cultured smooth muscle rabbit aortic cells. Mevinolin (MEV), a potent inhibitor of endogenous CHS biosynthesis was used to deplete the CHS content. Elevation of CHS concentration in the membrane was achieved using a CHS-enriching medium. Treatment of smooth muscle cells with MEV led to a nearly twofold increase in the rotational diffusion coefficient of DPH (D) and to about a ninefold elevation of probability of the channels being open (P _o). The addition of CHS to the cells membrane resulted in a nearly twofold decrease in D and about a twofold decrease in P _o. Elementary conductance of the channels did not change under these conditions. These data suggest that variations of the CHS content in the plasma membrane of smooth muscle cells affect the kinetic properties of Ca²⁺-dependent K⁺ channels presumably due to changes in plasma membrane fluidity. Our results give a possible explanation for the reported variability of Ca²⁺-dependent K⁺ channels kinetics in different preparations.     

Calcium- and cyclic-AMP-mediated secretory responses in isolated colonic crypts

Whole-cell recordings were performed at isolated crypts from the distal colon of the rat. Enterocytes in intact crypts, patched from the basolateral side, exhibited a gradient in the resting zero-current potential. Along the axis of the crypt, the highest potentials were measured in the ground region, the lowest in the surface region. The cholinergic agonist, carbachol, induced a hyperpolarization and an increase of the outward current in both the middle and the ground cells of intact crypts. This effect could be prevented by Ba²⁺ or by the intracellular Ca²⁺ antagonist, 8-(N, N-diethylamino)-octyl-3,4,5-trimethoxy-benzoate hydrochloride (TMB-8). Its action, however, was not dependent on the presence of external Ca²⁺. Both ground cells and the cells in the middle part of the crypt responded to forskolin, an activator of the adenylate cyclase, with a depolarization. In the middle part of the crypt, the depolarization induced by forskolin was associated with an increase of the outward current. It could be blocked by the Cl^– channel blocker, 5-nitro-2-(3-phenylpropylamino)-benzoate, indicating an increase of Cl^– conductance. In contrast, the forskolin-induced depolarization in the ground part of the crypt was associated with a decrease of the outward current. This effect could be prevented by Ba²⁺, indicating a decrease of a potassium conductance. The changes in outward current could be prevented by the presence of an inhibitor of protein kinase A in the pipette solution. In conclusion, these results suggest that carbachol, an agonist acting on the Ca²⁺ pathway, indirectly causes Cl^– secretion by an increase of the driving force, i.e. the membrane potential. Only the activation of cyclic AMP synthesis by forskolin is able to increase Cl^– conductance in the rat colon. The latter response seems to be dependent on the state of differentiation of the enterocytes.     

Acetylcholine and ATP hyperpolarize endothelium via activation of different types of Ca(2+)-activated K+ channels

Experiments on rat thoracic aorta showed that ATP and acetylcholine hyperpolarize endothelial cell via selective activation of low (SK) and intermediate (IR) conductance Ca²⁺-activated K⁺ channels, respectively. It was hypothesized that ATP- and acetylcholine-activated Ca²⁺-signals are spatially separated and generated in plasma membrane regions with predominant localization of SK- and IR-channels, respectively.     

ATP suppresses activity of Ca2+ -activated K+ channels by Ca2+ chelation

Ca²⁺-activated maxi K⁺ channels were studied in inside-out patches from smooth muscle cells isolated from either porcine coronary arteries or guinea-pig urinary bladder. As described by Groschner et al. (Pflügers Arch 417:517, 1990), channel activity (NP _o) was stimulated by 3 M [Ca²⁺]_c (1 mM Ca-EGTA adjusted to a calculated pCa of 5.5) and was suppressed by the addition of 1 mM Na₂ATP. The following results suggest that suppression of NP _o by Na₂ATP is due to Ca²⁺ chelation and hence reduction of [Ca²⁺]_c and reduced Ca²⁺ activation of the channel. The effect was absent when Mg ATP was used instead of Na₂ATP. The effect was diminished by increasing the [EGTA] from 1 to 10 mM. The effect was absent when [Ca²⁺]_c was buffered with 10 mM HDTA (apparent pK _Ca 5.58) instead of EGTA (pK _Ca 6.8). A Ca²⁺-sensitive electrode system indicated that 1 mM Na₂ATP reduced [Ca²⁺]_c in 1 mM Ca-EGTA from 3 M to 1.4 M. Na₂ATP, Na₂GTP, Li₄AMP-PNP and NaADP reduced measured [Ca²⁺]_c in parallel with their suppression of NP _o. After the Na₂ATP-induced reduction of [Ca²⁺]_c was re-adjusted by adding either CaCl₂ or MgCl₂, the effect of Na₂ATP on NP _o disappeared. In vivo, intracellular [Mg²⁺] exceeds free [ATP^4–], hence ATP modulation of maxi K⁺ channels due to Ca²⁺ chelation is without biological relevance.