INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

THE LATE DISTAL TUBULE consists of the short connecting tubule and the initial portion of the cortical collecting duct (CCD). The balance of ion transporting processes, combined with the high resistance to paracellular ion movement in this nephron segment, results in the development of a large, lumen-negative transepithelial potential difference (V_te), of the order of 30-50 mV. This arises primarily from the diffusion of sodium into the principal cell through amiloride-sensitive ion channels, thus depolarizing the apical membrane with respect to the basolateral membrane. However, accumulating evidence both from in vivo and in vitro studies suggests that electrogenic proton secretion may provide a positive component to the potential difference (14, 21, 28), which is normally obscured by the larger, negative component derived from transepithelial flux of sodium and other ions.

The processes underlying proton secretion in this nephron segment have now been partially defined. Active secretion by the "proton pump" (H⁺-ATPase) predominates, although there may be some additional contribution from Na⁺/H⁺ exchange. In addition, several groups have described an H⁺-K⁺-ATPase in the late distal tubule of both hypokalemic and normokalemic animals (11, 30). Of these mechanisms of proton secretion, the H⁺-ATPase is the only transporter known to be intrinsically electrogenic (3, 15). Nevertheless, attempts to demonstrate this electrogenicity in the late distal tubule of the anesthetized rat had been unsuccessful (5, 14), even when sodium channels were blocked with amiloride. However, it has recently been demonstrated that further addition of 5-nitro-2-(3-phenylpropylamino)benzoic acid (NPPB) and barium, blocking chloride and potassium channels respectively, allows an assessment of the magnitude of the electrogenicity of H⁺-ATPase activity (13). In the rat late distal tubule, H⁺-ATPase accounted for a potential difference of ~2 mV.

The late distal tubular V_te in potassium-depleted rats has been found to be significantly less lumen negative than that of potassium-replete controls (2, 25). Although this may reflect the maintenance of a high electrochemical gradient across the apical membrane in potassium-depleted rats, such a dietary regime has also been shown to increase bicarbonate reabsorption, and thus by inference proton secretion, in this nephron segment (9, 10, 14). Moreover, the late distal V_te of the potassium-depleted rat became significantly lumen positive during perfusion with a solution containing barium and amiloride (2), an effect that could be abolished by intravenous infusion of acetazolamide (R. J. Unwin, unpublished data). This suggests that potassium depletion may be associated with an increase in electrogenic proton secretion. Circumstantial evidence for increased H⁺-ATPase activity during potassium depletion also comes from histological studies. Hypokalemia was found to be associated with both hyperplasia of the -intercalated cell apical membrane and an increase in the density of electron-dense studs in this membrane (23, 27). As these studs may represent the morphological manifestation of H⁺-ATPase (8), these results suggest that potassium depletion induces the fusion of pump-laden vesicles into the apical membrane in a manner analogous to that reported during acidosis (6).

In the present study, we have used bafilomycin A₁, a specific inhibitor of H⁺-ATPase, to measure electrogenic proton secretion during blockade of sodium, potassium, and chloride channels in the late distal tubule of control and potassium-depleted rats. To complement these in vivo investigations, a direct assessment of the effect of potassium depletion on the distribution of the proton pump in CCD was made by using a monoclonal antibody raised against its 31-kDa subunit.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Microperfusion experiments. Experiments were performed on male Sprague-Dawley rats that had been maintained on a potassium-deficient diet (<750 µmol K/kg dry wt, Teklad 88239; Bicester, Oxon, UK) for 10-15 days prior to anesthesia and on weight-matched animals that had been allowed free access to a companion control diet (250 mmol K/kg dry wt, Teklad 88238) prior to experimentation. Rats were anesthetized with 5-sec-butyl-5-ethyl-2-thiobarbituric acid (Inactin; Byk-Gulden, Constance, Germany; 110 mg/kg body wt ip) and prepared surgically for microperfusion of superficial distal tubules, as follows. The left jugular vein and left femoral artery were cannulated for infusions and for withdrawal of blood, respectively. A tracheotomy was performed, and the bladder was catheterized. The left kidney was exposed by a flank incision, cleared of perirenal fat, and placed in a Perspex dish that was clamped to the operating table. Following immobilization of the kidney in a 5% agar solution, the exposed surface was bathed in heavy mineral oil for the duration of the experiment. The ureter was cannulated close to the hilum. All animals received an intravenous infusion of NaCl (125 mmol/l) and NaHCO₃ (25 mmol/l) throughout the course of the experiment at 2 ml/h; during the final hour of surgery, an additional volume of infusate (1 ml) was administered to compensate for surgical losses. From 1 h after the completion of surgery, [³H]inulin (Amersham International, Aylesbury, UK) was included in the infusate (2 µCi primer, 2 µCi/h); a further 60 min were allowed for equilibration of the isotope. Superficial distal tubules were then identified and, following insertion of a proximal oil block, perfused using a thermally shielded microperfusion pump (Hampel, Neu-Isenburg, Germany) inserted in an early loop.

The distal tubule was perfused sequentially in an orthograde manner with two perfusates designed to mimic native early distal tubular fluid, which contained (in mmol/l) 50 NaCl, 2 KCl, 2 NH₄Cl, 1 CaCl₂, 100 urea, and 5 HEPES. In addition, amiloride (100 µmol/l; Research Biochemicals International, Natick, MA), barium chloride (2 mmol/l), and NPPB (100 nmol/l; a gift from Dr. R. Greger, Freiburg, Germany) were included to inhibit ion movement through sodium, potassium, and chloride ion channels, respectively. The pH of the perfusates was 7.4.

The two perfusates used were identical except for the presence or absence of bafilomycin A₁ (1 µmol/l; Sigma, Poole, UK), an H⁺-ATPase inhibitor. The vehicle alone (DMSO; 0.02% vol/vol) was included in the control infusate. Each tubule was perfused with both solutions.

Voltage measurements were made using single-barrel micropipette electrodes with tip diameters of 1-2 µm, pulled from borosilicate glass capillaries with an internal glass filament (GC120-F10, 1.2 mm OD × 0.69 mm ID; Clark Electromedical Instruments, Pangbourne, UK), filled with a solution containing sodium acetate (250 mmol/l), sodium chloride (250 mmol/l), and potassium chloride (2 mmol/l), and connected to a high-impedance electrometer (World Precision Instruments, Stevenage, UK); only electrodes with an input resistance of <15 M were used.

Baseline, or "zero," voltage was initially measured in the fluid bathing the kidney surface. The electrode was then advanced into the lumen of the last accessible portion of the late distal tubule (the positioning of the electrode within the tubule lumen was verified by observation of the change in voltage in response to a change in perfusion rate; Ref. 25). Subsequently, late distal tubular V_te was recorded during perfusion at 10 nl/min. With the electrode remaining in place, the same nephron segment was perfused again with the second of the perfusate pair, and a second measurement of V_te was made. Following this a second baseline, or zero, measurement was made. The tubule was then filled with silicone rubber (Microfil; Flowtech, Carver, MA) so that the impalement site could later be confirmed by microdissection.

Voltages were recorded using a Macintosh LC computer and Maclab chart software (AD Instruments, Hastings, UK). A recording was accepted as valid if 1) the deflection from zero was stable for at least 1 min, 2) the first and second baselines differed by less than 2 mV, 3) the postimpalement tip resistance was identical to the preimpalement value, and 4) the impalement site was in the final surface loop of the distal tubule.

During microperfusion recordings, urine was collected separately from the micropuncture and contralateral kidney, and mean arterial blood pressure was recorded throughout the experiment. Small (40 µl) samples of arterial blood were taken at regular intervals throughout the experiment to determine plasma [³H]inulin activity. At the end of the experiment, a 75-µl sample of arterial blood was taken for measurement of bicarbonate concentration, and a large (2 ml) blood sample was taken for plasma sodium and potassium concentration.

Immunocytochemistry. Experiments were performed on male Sprague-Dawley rats maintained on either a low-potassium or control diet as described above. Following anesthesia (Inactin, 110 mg/kg body wt ip), the left femoral artery was cannulated with polyethylene tubing. Both kidneys were then removed, and a 2-ml sample of arterial blood was immediately taken for the measurement of blood pH and plasma sodium, potassium, bicarbonate, and aldosterone concentrations. The animal was then killed with an overdose of anesthetic.

Kidneys were sectioned into 2-mm-thick slices and incubated overnight at room temperature in mercuric chloride (B5) fixative before paraffin embedding. Sections (4 µm) were cut, placed on coated slides (BDH Pharmaceuticals, London, UK) and dried overnight at 37°C. The sections were subsequently dewaxed in xylene solution, rehydrated in decreasing concentrations of ethanol, and incubated in Lugol's iodine solution, 5% sodium thiosulfate, and PBS (pH 7.4). Endogenous peroxidase activity was blocked by incubating the sections in freshly made 3% H₂O₂ (in methanol). Following rinsing in PBS, nonspecific binding was blocked by exposing the sections to the blocking solution (20% calf serum; Life Technologies, Paisley, UK) in PBS, with 1% polyethylene glycol for 30 min. Subsequently, the sections were incubated in the primary antibody (E11, 1:50 dilution, a gift from Dr. S. Gluck, St. Louis, MO) at room temperature for 1 h.

After incubation with the primary antibody, the sections were rinsed for 60 min (3 changes) in PBS and then immersed in the secondary antibody (horseradish-peroxidase-conjugated rabbit anti-mouse immunoglobulin IgG; 1:150; Dako, High Wycombe, Bucks, UK) for 30-45 min at room temperature. The sections were again rinsed in PBS for 30 min.

The peroxidase reaction was developed by exposing the sections to 3',3'-diaminobenzidine tetrahydrochloride and H₂O₂ for 5-10 min. Following the peroxidase reaction, sections were counterstained with hematoxylin, washed in water, dehydrated in an ascending alcohol series, and cleared in xylene. Finally, the sections were mounted in a xylene-based medium.

Cell counting. Fifteen to twenty CCDs were chosen at random in each kidney under low-power magnification. The number of cells with a distinct nucleus in each segment was counted under high-power magnification, together with the number of cells showing distinct H⁺-ATPase staining. Each section was counted in a single blind manner; all cells displaying distinct staining for H⁺-ATPase were included in one of the following five categories, adapted from the counting method of Gluck and associates (6, 8): apical (stain accentuated in the apical region of the cell), basolateral (stain accentuated in the basolateral region of the cell), apical/basolateral (simultaneous distinct staining of the apical and basolateral poles), diffuse (diffuse homogenous staining), and indeterminate (cells that could not be safely assigned to any of the above categories). The antibody used in this study, E11, was raised in mice against a 10 amino acid synthetic peptide derived from the known sequence of the carboxy-terminal region of the bovine H⁺-ATPase 31-kDa subunit (31). This antibody, which has been used extensively by Gluck and colleagues (16, 17, 31), is extremely well characterized, and its specificity has been confirmed. Moreover, it has been shown to give the same staining pattern as antibodies against other proton pump subunits, indicating that E11 staining represents constitutive proton pump localization (7).

Analyses. Sodium and potassium concentrations in urine and plasma were measured by flame photometry (model 543; Instrumentation Laboratory, Warrington, UK). For microperfusion experiments, [³H]inulin activities in urine and plasma were determined by beta-emission spectroscopy (model 2000 CA; Canberra Packard, Pangbourne, UK) after dispersal in Aquasol 2 scintillation cocktail (DuPont, Stevenage, UK).

Plasma aldosterone concentration was measured by radioimmunoassay (Coat-A-Count; Diagnostic Products, Caenarfon, UK). The pH and the bicarbonate concentration in arterial blood were measured by blood-gas analysis (ABL 500 blood gas system; Radiometer, Copenhagen, Denmark).

Calculations. For microperfusion experiments, the clearance of [³H]inulin was used as a measure of glomerular filtration rate; this and the excretion rates of sodium and potassium were calculated by standard formulas. Distal tubular V_te was taken as the mean deflection from the mean of the pre- and postimpalement baselines. This change in voltage is the sum of two components: the first is the true V_te, and the second results from altered tip potentials that may arise from differences in the ionic composition of the interstitial fluid (in which baseline voltage was measured) and that of the perfusate. To estimate the magnitude of the second component, the electrode was moved from the surface fluid to a pool of the perfusate, and the change in voltage was measured. In both groups of rats, the change was always less the 0.2 mV. For the immunocytochemistry, counts made in each CCD were combined to give a mean value per kidney. These data in turn were used to calculate a mean for each group.

Statistics. All values are presented as means ± SE. For both series of experiments, plasma data were compared using Student's t-test for independent samples. For micropuncture experiments, comparisons within groups were made using Student's t-test for paired samples; an unpaired t-test was used for comparisons between groups. For the immunocytochemical series, a two-way analysis of variance was used to compare the distribution of H⁺-ATPase staining in the two groups of rats. The least significant difference post hoc test was used for within-category comparisons. In all cases, a difference was taken as statistically significant if P < 0.05.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Blood/plasma data. Blood and plasma data for the two groups of animals are shown in Table 1. As expected, the animals maintained on the potassium-deficient diet were markedly hypokalemic compared with the potassium-replete controls. In addition, potassium-depleted rats were alkalotic and had an elevated plasma bicarbonate concentration, whereas plasma aldosterone, measured only in immunocytochemistry experiments, was significantly reduced. It should be noted that although the potassium-deficient diet resulted in a very severe hypokalemia, animals in this group gained weight normally and were indistinguishable in appearance and behavior from potassium-replete controls.

Whole kidney data. As there was no significant difference in body weight between the two groups, whole kidney data are presented in Table 2 as absolute values. Potassium excretion was markedly lower in the potassium-depleted rats than in the control group; there were no other significant differences between the two groups of rats. There were no differences between measurements in the micropuncture and contralateral kidneys (data not shown).

Distal tubule transepithelial potential difference. The V_te of the late distal tubule was measured in groups of potassium-replete and potassium-depleted rats during distal tubular perfusion with solutions containing either bafilomycin A₁ or the DMSO vehicle alone. Each tubule was perfused with the two solutions. Since both contained the ion channel blockers amiloride, barium, and NPPB, the late distal tubular V_te, normally highly lumen negative (2), was markedly reduced in both groups of rats. The effect of bafilomycin A₁ was found to be independent of the order in which the perfusates were used; the data presented in Figs. 1 and 2 therefore represent pooled values.

View larger version (17K): [in this window] [in a new window]   View larger version (22K): [in this window] [in a new window]   Fig. 1.   Late distal tubular transepithelial potential difference (Vte) in potassium-replete rats (A; n = 5, 22 recordings) or potassium-depleted rats (B; n = 5, 23 recordings), during paired perfusion with solutions containing either bafilomycin A1 (Baf A1; 1 µmol/l) or vehicle alone. Individual values and means ± SE are shown.

As shown in Fig. 1, bafilomycin A₁ caused a significant change in the potential difference toward lumen negativity in both groups of rats. However, the net effect of bafilomycin A₁ on V_te was significantly greater in the hypokalemic animals (Fig. 2). The mean value for late distal V_te during perfusion with the control perfusate was different in the two groups of animals (P < 0.001), being lumen negative in the potassium-replete group and lumen positive in potassium-depleted animals. Although the net effect of bafilomycin A₁ was greater in the low-potassium rats, the measured potential difference during perfusion with the inhibitor was still significantly less negative than the corresponding value in the potassium-replete group (P < 0.01).

View larger version (23K): [in this window] [in a new window]   Fig. 2.   Change in late distal tubular Vte in control rats (open bar; n = 22 recordings) or potassium-depleted rats (Low-K, hatched bar; n = 23 recordings) from control values in response to perfusion of the same nephron segment with bafilomycin A1 (1 µmol/l). Data are means ± SE.

Immunocytochemistry. The percentage of cells in the CCD that showed H⁺-ATPase staining was similar in the two groups of animals (control = 39.3 ± 1.2% of 2,725 cells counted; low K = 38.1 ± 1.4% of 3,089 cells counted; not significant). All reported categories were observed, as shown in Fig. 3. The distribution of H⁺-ATPase labeling within these groups is shown in Fig. 4. Two-way analysis of variance indicates a highly significant (P < 0.001) redistribution of H⁺-ATPase in the potassium-depleted rats. The percentage of stained cells with H⁺-ATPase predominantly in the apical membrane was significantly higher and the percentage of cells showing diffuse staining significantly lower than the corresponding values in the control group.

View larger version (91K): [in this window] [in a new window]   Fig. 3.   H+-ATPase immunoreactivity in intercalated cells of the cortical collecting duct (CCD), identified using E11 murine monoclonal antibody. Top: immunoreactivity in apical region of the cell (A) and in basolateral region of the cell (B). Middle: simultaneous staining of apical and basolateral poles (AB). Bottom: diffuse (D), homogenous staining of cytoplasm.

View larger version (28K): [in this window] [in a new window]   Fig. 4.   Distribution of H+-ATPase immunoreactivity (E11 monoclonal antibody) within subtypes of intercalated cells of the CCD in potassium-replete (n = 5 rats; open columns) and potassium-depleted rats (n = 5; hatched columns). Cells were categorized into 5 distinct groups with regard to the polarity of the proton pump labeling.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

Plasma and whole kidney data. Dietary potassium depletion for 10-17 days resulted in marked hypokalemia and metabolic alkalosis. In addition, plasma aldosterone activity was significantly reduced, and the animals were excreting very little potassium in the urine. There were no other significant differences.

Late distal tubular transepithelial potential difference during perfusion with vehicle perfusate. In these experiments, amiloride, BaCl₂, and NPPB were included in the perfusate to inhibit the movement of sodium, potassium, and chloride through their respective ion channels, a protocol which has previously been shown to allow an estimation of the contribution of electrogenic proton secretion to the generation of V_te (13). In potassium-replete animals, this had the effect of reducing the potential close to zero (see Fig. 1). Yet, despite the cocktail of ion channel blockers, the potential difference remained lumen negative, contrasting with analogous experiments performed in vitro. The reason for this is not clear, but the persistence of the lumen-negative V_te may represent residual ionic diffusion across the tubular epithelium. Despite this, the V_te measured in the late distal tubule of potassium-depleted animals was lumen positive. This suggests enhanced proton secretion in these animals (2).

Effect of bafilomycin A₁. Until recently, the contribution of the V-type H⁺-ATPase to the generation of late distal tubular V_te had only been estimated on theoretical grounds: attempts at empirical measurements had proved inconclusive. However, Fernandez et al. (13), using inhibitors to block the main ion channels of this segment, have shown that H⁺-ATPase activity has the effect of depolarizing the lumen-negative V_te by ~2 mV. Data from the present series of experiments are consistent with these findings, since microperfusion of the distal tubule with bafilomycin A₁ caused the late distal tubular V_te of control rats to become more lumen negative by ~2 mV. In addition, we have investigated the contribution of H⁺-ATPase to V_te in potassium-depleted animals and have observed that bafilomycin A₁ causes a negative deflection significantly greater than that measured in potassium-replete controls, suggesting an upregulation of H⁺-ATPase activity during potassium depletion.