METHODS

Five-Sixths Nephrectomy

Adult male Sprague-Dawley rats, born and raised in a climate-controlled facility at the University of Ottawa and weighing between 230 and 300 g, underwent five-sixths nephrectomy, as previously described (16). The rats were allowed 13–16 days recovery after surgery prior to microperfusion or remnant kidney removal.

K Depletion

Dietary K depletion was induced in Nx rats (Nx-K) by ad libitum ingestion of a K-free synthetic diet and distilled water for 7 days prior to the experiment. The synthetic diet consisted of (in g/kg diet) 12.90 NaH₂PO₄, 4.07 NaCl, 9.60 MgCl₂ ⋅ 6H₂O, 14.23 CaCO₃, and 959.19 basal electrolyte-free diet powder (TD-78093; Harlan-Teklad, Madison, WI) and contained 163.13 mmol Na/kg diet, 164.08 mmol Cl/kg diet, and 0 mmol K/kg diet. Control Nx rats remained on ad libitum rat chow and tap water for the 7 days prior to experimentation.

Microperfusion

Rats were housed in individual stainless steel metabolic cages for 16 h (overnight) prior to microperfusion, allowing measurement of ingested food and drink and collection of urine under oil, using thymol as a preservative. The following morning, the animals were anesthetized with 100 mg/kg thiobutabarbital sodium (Inactin; Research Biochemical International, Natick, MA) and prepared for microperfusion, as described previously (16). Briefly, the rat was placed on a heated operating table, and a tracheostomy was performed, using PE-240 tubing. The left carotid artery was cannulated for continuous blood pressure measurement and collection of blood for acid-base and electrolyte analyses, while the left jugular vein was cannulated with three lines for infusion of fluid, pentobarbital sodium anesthetic (Somnotol; MTC Pharmaceuticals, Cambridge, ON, Canada), and 10% Lissamine green. The left kidney was exposed by flank incision, carefully dissected from the adrenal gland, and immobilized in a stainless steel cup covered with mineral oil. The ureter was catheterized with PE-50 tubing to ensure proper urine flow.

To replace surgical fluid losses, the rats were infused at 1% body wt/h for 30 min via the jugular vein, with donor plasma from a control rat (non-nephrectomized and K replete). The animal was then maintained on 0.9% saline at 1% body wt/h for the remainder of the experiment.

Perfusable surface two-loop DT were identified by injecting a 0.02-ml bolus of 1% Lissamine green into surface proximal loops and observing its passage through the nephron. DT were perfused at 15 nl/min with a hypotonic solution containing (in mM) 28 HCO₃, 26 Cl, 56 Na, 2 K, 1.8 Ca, 22 urea, and 4 gluconate. FD & C green no. 3 dye (0.05%; Keystone, Chicago, IL) and bovine serum albumin (0.1%; Intergen, Purchase, NY) were also added to the perfusate.${[}^{3}H]$inulin (DuPont Canada, Mississauga, ON) was added to the perfusate as a marker of water reabsorption. The perfusion rate of 15 nl/min was chosen on the basis of preliminary experiments on Nx rats, which showed early DT free-flow rates of 13.3 ± 0.7 nl/min (n = 12). The perfused bicarbonate load (28 mM HCO₃ × 15 nl/min = 420 pmol/min HCO₃), considerably higher than free-flow (2.7 mM HCO₃× 13.3 nl/min = 36 pmol/min HCO₃), was chosen to more easily reveal effects of inhibitors, although exaggerating proton secretion and possibly impeding bicarbonate secretion. A 10-min preperfusion period preceded all collections. At the end of the experiment, tubules that provided samples were back filled from the collection site with Latex (Microfil; Flow Tech, Carver, MA) to confirm surface direction of flow and to provide a hardened cast which, when removed from the digested kidney, allows measurement of the length of the perfused segment.

Groups

Group 1 consisted of control Nx rats, whereas Nx-K rats were divided into four groups: one control ( group 2) and three experimental ( groups 2B–2D). Rats ingroups 1–2 were perfused with the control solution (above), while groups 1B and 2B–2D were perfused with a modified solution, to which one of three transport inhibitors was added (Table 1). The H⁺-K⁺-ATPase inhibitor, Sch-28080 (gift from Schering Canada, Pointe-Claire, PQ) was dissolved in DMSO (final concentration DMSO in perfusate was 0.5%) and added at 10⁻⁵ M to the perfusate used in groups 1B and2B rats. Concanamycin A (Sigma-Aldrich Canada, Mississauga, ON) was used in group 2C to inhibit V-type H⁺-ATPase activity. It was first dissolved in DMSO (final concentration DMSO in perfusate was 0.1%) and then added to the control perfusate to a final concentration of 5 × 10⁻⁹ M. Finally, ingroup 2D, the AT₁ receptor inhibitor, losartan (generously donated by the DuPont-Merck Pharmaceutical, Wilmington, DE), was dissolved in distilled water and perfused at 10⁻⁶ M.

Analyses

Whole blood and urine pH and $P{\text{CO}}_2$were measured quantitatively by electrode (IL 1610 blood gas system; Instrumentation Laboratory, Milano, Italy), and HCO₃ concentrations were calculated. Plasma and urine Na and K concentrations were measured by flame photometry (IL 943 flame photometer; Instrumentation Laboratory), and Cl concentrations were measured by electrotitration (CMT 10 chloride titrator; London Scientific, London, ON). Plasma total protein concentrations and urine specific gravity were measured by refractometry (10400A TS meter; Cambridge Instruments, Buffalo, NY), and hematocrits were determined by microcapillary reader (International Equipment, Needham Heights, MA). Urine osmolalities were determined by freezing-point osmometry (advanced model 3MOplus MicroOsmometer; Advanced Instruments, Norwood, MA). Plasma creatinine concentrations were measured by the kinetic Jaffé method, without deproteinization (BM/Hitachi System 717; Boehringer-Mannheim, Laval, PQ). ${[}^{3}H]$inulin in perfusates and samples was counted using the Beckman model 3801 liquid scintillation system (Beckman Instruments Canada, Mississauga, ON).

Perfusate and sample total carbon dioxide (tCO₂) concentrations were measured by microcalorimetry, as previously described (16). A standard curve (10, 20, 30, and 40 mM NaHCO₃) was run before sample analysis, and standards bracketed the determination of sample and perfusate tCO₂ concentration. Perfusate and sample chloride concentrations were determined by constant current electrotitration with potentiometric end-point sensing, as described previously (16). A series of four standards (20, 40, 60, and 80 mM NaCl) was run before perfusate and sample analysis to generate a regression line, and, in analytes suspected to have a low chloride concentration, a mid-curve standard was added to the sample or perfusate before titration.

Calculations. The perfusion rate (R_P) was calculated as the product of the collected rate (R_C) and the ratio of sample inulin concentration over perfusate inulin concentration. The difference between the calculated perfusion rate and the measured collected rate (R_P − R_C) provided a measurement of water reabsorption (J_v) along the length of the perfused DT.$J_{{\text{tCO}}_2}$was calculated as

Western Analysis

Nx and Nx-K rats were anesthetized by intraperitoneal injection of 100 mg/kg Somnotol and perfused via the abdominal aorta with cold PBS at pH 7.4. The remnant left kidney was removed, and portions of the cortex were frozen in liquid nitrogen. Kidney tissue was added to 2× sample buffer (13) at 1 g/10 ml and solubilized by intermittent sonication (Microson Cell Disrupter; Heat Systems Ultrasonics, Farmingdale, NY) and subsequent heating to 100°C for 10–15 min (10). Samples of stomach antrum and distal colon from control and Nx-K rats were processed in the same manner. Protein concentration was determined by bicinchoninic acid protein assay kit (Pierce, Rockford, IL), according to the manufacturer’s protocol, as modified for microplate reader (10).

Samples prepared from cortical tissue (50 μg protein) were separated on a 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis slab gel according to the method of Laemmli (13) and electrophoretically transferred to nitrocellulose (25). For immunoblotting, the nitrocellulose was blocked with 3% bovine serum albumin in 0.1% Tween 20 in PBS for 1 h at room temperature, followed by overnight incubation in H⁺-ATPase primary antibody at 1:2,000 dilution at 4°C or monoclonal anti H⁺-K⁺-ATPase (gastric β-subunit, Affinity Bioreagents MA3–923) at 1:2,500 dilution for 2 h. The H⁺-ATPase antiserum was raised in rabbits against the COOH-terminal decapeptide of the 31-kDa subunit of the bovine renal V-type H⁺-ATPase, as described by Sullivan et al. (24). The nitrocellulose was rinsed in 0.1% Tween 20 in PBS and incubated with horseradish peroxidase-conjugated donkey anti-rabbit IgG (Amersham NA-9340) or sheep anti-mouse IgG (Amersham NA-9310) at a 1:10,000 dilution for 30 min at room temperature. Immunoreactive protein bands were detected by the enhanced chemiluminescence method (ECL; Amersham Life Sciences, Oakville, ON), according to the manufacturer’s instructions. Controls for the specificity of the antiserum and densitometric measurement of immunoreactive bands were performed as described previously (16).

Immunocytochemistry

For immunocytochemistry to detect the H⁺-ATPase, kidneys were harvested from Nx and Nx-K rats as above, following abdominal aortic perfusion of cold PBS at pH 7.4, followed by 4% paraformaldehyde in PBS. In each rat, the remnant left kidney was removed and rinsed briefly in cold PBS, and 1- to 2-mm freehand sagittal sections were cut with a razor blade. These sections were placed in 4% paraformaldehyde in PBS for 3–4 h and washed in 70% alcohol. The tissues were dehydrated in a graded series of alcohols followed by xylene and embedded in paraffin.

Sections (7 μm) were cut from paraffin blocks using a microtome, mounted on glass slides, dried on a slide warmer for 3–4 h, and incubated overnight at room temperature with H⁺-ATPase primary antiserum diluted 1:200 in 0.1 M tris(hydroxymethyl)aminomethane buffer containing 0.6% carrageenan and 0.3% Triton X-100 (TCT). After two 7-min washes in buffer, sections were incubated for 30 min at room temperature with biotinylated antirabbit secondary antibody (Amersham) diluted 1:50 with TCT. After another series of washes, final labeling was performed by 2-h incubation in sheep anti-rabbit streptavidin-CY3 conjugate (Sigma-Aldrich) diluted 1:200 in TCT. Positive immunoreactivity was visualized by fluorescence microscopy at 545 ± 20 nm.

Immunocytochemical localization of the β-subunit of the H⁺-K⁺-ATPase was performed with paraffin sections as above. The primary antibody was used at a dilution of 1:250 followed by sheep anti-mouse CY3 (Sigma) at a dilution of 1:400.

Electron Microscopy

Remnant left kidneys from Nx and Nx-K rats were removed following abdominal aortic perfusion of a mixture of 1% paraformaldehyde/1% glutaraldehyde in 0.1 M phosphate buffer at pH 7.4 and cut longitudinally into 1- to 2-mm- slices. Midline slices were immersed in the above fixative mixture, and 1-mm³ pieces of cortex subjacent to the renal capsule were excised and immersed in fixative for 4 h at 4°C. After a thorough wash in buffer, the tissues were postfixed in 2% osmium tetroxide in 0.1 M phosphate buffer for 2 h at 4°C. Subsequently, specimens were dehydrated in a graded ethanol-propylene oxide series and embedded in Epon (Marivac, Halifax, NS). Gold sections were cut with a diamond knife (Reichert Ultracut Ultramicrotome; Leica Canada, Willowdale, ON), mounted on 150-mesh copper grids (Marivac), stained with uranyl acetate and lead citrate, and examined with a Philips 300 electron microscope. Images were recorded on Eastman 35-mm film (Treck Hall, Ville St. Laurent, PQ) at a magnification of ×3,300 or ×6,600.

Immunogold Labeling

Kidneys from Nx and Nx-K rats were removed following abdominal aortic perfusion of cold PBS, pH 7.4, followed by cold 4% paraformaldehyde in 0.1 M sodium phosphate buffer, pH 7.4, and 1- to 2-mm freehand sagittal sections were cut with a razor blade. Slices were immediately immersed in fixative and then 1-mm³ pieces of cortex were excised, immersed in fixative for 3–4 h, and rinsed with 0.1 M phosphate buffer. Residual aldehyde groups were quenched with 50 mM glycine in 0.1 M phosphate buffer for 30 min. The tissue was then rinsed with phosphate buffer and treated with freshly made 0.1 M NH₄Cl for 1 h. Specimens were dehydrated sequentially in 30, 50, 70, and 90% ethanol and infiltrated with 90% ethanol-LR White (Marivac), followed by 1:2 and 1:3 mixtures, and finally by pure LR White. The above procedures were done at 4°C. The tissue samples were then placed in gelatin capsules subsequently filled with LR White and made air tight with a cover. Polymerization was done at 50°C for 6 h in a vacuum oven.

Silver-gold sections were cut with a diamond knife (Leica) and collected on formvar-coated 150-mesh nickel grids (Marivac). Sections were floated on 1% casein/5% normal goat serum in PBS for 30 min at room temperature to block the nonspecific binding sites, then rinsed with PBS, and treated overnight at 4°C with either1) H⁺-ATPase antiserum diluted 1:150 with PBS, 2) preimmune serum diluted 1:150 with PBS, 3) PBS, or4) H⁺-ATPase antiserum preabsorbed with the decapeptide antigen (100 μg/ml peptide) diluted to 1:150 with PBS. Grids were then thoroughly washed with PBS and incubated for 1 h at room temperature with diluted (1:100) goat anti-rabbit IgG coated with 18-nm gold particles (Bio/Ca Scientific, Mississauga, ON). Sections were lightly stained with 0.25% uranyl acetate and lead citrate (18) and examined with a Philips 300 electron microscope. Images were recorded on Eastman 35-mm film at an original magnification of ×3,300 or ×6,600.

Morphometry

Morphometric analysis of electron microscope images was performed on a Power Macintosh 7100/80 computer, using the public domain NIH Image program (developed at National Institutes of Health and available athttp://rsb.info.nih.gov/nih-image). The 35-mm negatives were digitized with a Polaroid PrintScan 35 at 675 dpi, and the reversed image was displayed on the computer screen. Measurements of cell surface area were determined by passing the cursor around the boundary of the cell at a final magnification of ×8,200. The cell surface area was measured on a total of 60 A-type intercalated cells, representing 20 cells from three Nx rats or 20 cells from three Nx-K rats.

Measurements of H⁺-ATPase immunogold labeling were done at a final magnification of ×40,300 by passing the cursor along the boundary of the apical cell membrane, including the surface of those microplicae continuous with the cell surface. The number of immunogold particles lying along that distance of membrane was expressed as gold particles per micrometer. Immunogold particle deposition was determined on a total of 30 A-type intercalated cells, representing 10 cells from three Nx rats or 10 cells from three Nx-K rats.

Statistics

All data are expressed as means ± SE. Balance, blood, and urine data summarized in Table 2 represent pooled data from all Nx rats ( groups 1 and1B) and all Nx-K rats ( groups 2, and2B–2D). Comparisons between two groups were done by two-tailed unpaired Student’st-test, whereas comparisons among more than two groups were carried out by ANOVA with posttesting by either Dunnett’s test (when comparisons were made within all possible pairings) or the Newman-Keuls test (when comparing each group vs. a single control). When raw data failed statistical tests for normality and/or homoscedasticity, the corresponding nonparametric test was employed. P < 0.05 indicated a statistically significant difference between groups (2).

RESULTS

Balance, Blood, and Urine Data

K depletion in Nx rats stimulated significant renal growth beyond that resulting from nephrectomy alone (left kidney wt as % total body wt, 0.63 ± 0.01 vs. 0.54 ± 0.01%,P < 0.05; Table 2). Nx-K rats had significantly lower plasma K concentration (2.3 ± 0.1 vs. 4.5 ± 0.1 mM, P < 0.05) and excreted a virtually K-free urine (52 ± 4 vs. 3,990 ± 159 μeq/16 h, P < 0.05). Plasma creatinine concentration and plasma protein concentration were also significantly elevated in the Nx-K rats (69 ± 2 vs. 60 ± 2 μM and 5.6 ± 0.1 vs. 5.1 ± 0.1 g/dl, respectively;P < 0.05).

Microperfusion

Effect of K depletion in Nx rats. There was a 60% increase in DT$J_{{\text{tCO}}_2}$in the Nx-K rats (102 ± 8 vs. 64 ± 7 pmol ⋅ min⁻¹ ⋅ mm⁻¹,P < 0.05; Table3 and Fig.1), which was not associated with any significant change inJ_v orJ_Cl. Perfusable tubule lengths in Nx rats tend to be longer than those found in sham-operated controls (11), but there was no further increase noted in Nx-K rats, despite greater kidney mass and increased cellular hypertrophy (see below).

Luminal perfusion of Sch-28080 in Nx and Nx-K rats. Although DT$J_{{\text{tCO}}_2}$was enhanced in Nx rats vs. sham-operated rats (16), it was not decreased by Sch-28080 inhibition of H⁺-K⁺-ATPase [64 ± 7 vs. 69 ± 12 pmol ⋅ min⁻¹ ⋅ mm⁻¹,P = not significant (NS)]. Furthermore, there were no significant effects of luminal Sch-28080 perfusion on$J_{{\text{tCO}}_2}$(102 ± 8 vs. 121 ± 13 pmol ⋅ min⁻¹ ⋅ mm⁻¹,P = NS),J_Cl, orJ_v in Nx-K rats.

Luminal perfusion of concanamycin A and losartan in Nx-K rats. The H⁺-ATPase inhibitor concanamycin A completely eliminated the enhanced$J_{{\text{tCO}}_2}$in Nx-K rats (65 ± 8 vs. 102 ± 8 pmol ⋅ min⁻¹ ⋅ mm⁻¹,P < 0.05; Table 4 and Fig.2). Associated with this reduction in$J_{{\text{tCO}}_2}$, there was a significant decrease inJ_v (2.46 ± 0.33 vs. 4.00 ± 0.37 nl ⋅ min⁻¹ ⋅ mm⁻¹,P < 0.05).

Western Analysis

Laser scanning densitometry of Western blotting results was performed to determine whether changes in levels of 33-kDa H⁺-ATPase could be detected in total cortical tissue (Table 5), as was reported in Nx vs. sham-operated rats (16). As seen in Fig.3, no obvious difference in the level of the 33-kDa immunoreactive band was apparent between samples prepared from Nx and Nx-K tissues. All density values were normalized for comparison by constructing a calibration curve utilizing density values from different loadings of Nx-K sample 8. Results indicate that there was no significant difference in cortical tissue H⁺-ATPase immunoreactive blotting between Nx and Nx-K rats (1.17 ± 0.24 vs. 1.06 ± 0.13,P = NS).

Immunocytochemistry

Compared with Nx rats, Nx-K rats showed no further enhancement of immunofluorescence of H⁺-ATPase in the apical plasma membrane of A-type intercalated cells. This is in contrast to our previous observations of Nx vs. sham-operated rats (16).

Electron Microscopy and Immunogold Labeling

Morphometric ultrastructural analyses of A-type intercalated cells of Nx-K rats revealed a significant increase in cross-sectional cell area compared with Nx rats (68.4 ± 2.6 vs. 51.5 ± 6.8 μm²,P < 0.05) (Fig. 4). H⁺-ATPase immunogold labeling of A-type intercalated cell apical plasma membrane was also enhanced in Nx-K compared with Nx rats (2.25 ± 0.13 vs. 1.61 ± 0.14 gold particles/μm, P < 0.05) (Fig.5).

Immunochemical Analysis for the β-Subunit of the Gastric H⁺-K⁺-ATPase

Western blotting for the β-subunit of the H⁺-K⁺-ATPase in samples of stomach revealed intense immunoreactive bands at 55 and 73–118 kDa, which represent the anticipated β-subunit precursor and its glycosylated form (4). Western blots of rat transverse and distal colon and kidney revealed no H⁺-K⁺-ATPase immunoreactive bands at 55 or 73–118 kDa. These immunoblotting results were confirmed with our immunocytochemical studies, which revealed intense immunostaining in parietal cells of the stomach but failed to detect any β-subunit immunoreactivity in colon or kidney (data not shown).

DISCUSSION

Our present studies on K-depleted nephrectomized rats extend our previous observations (16) that surviving DT markedly enhance bicarbonate reabsorption when compared with sham-operated controls. Our demonstration (16) that this effect is concanamycin A sensitive and dependent on the AT₁ receptor stimulated us to explore other mediators of this adaptive response. Although we recognize that K depletion has been reported to stimulate components of the renin-angiotensin system (1, 9, 19, 22) and may involve recruitment of apical H⁺-K⁺-ATPase to further enhance$J_{{\text{tCO}}_2}$(28), we were intrigued by our own observations showing a striking increase in number of “active” (defined by the presence of H⁺-ATPase-containing apical membrane) A-type intercalated cells in K-depleted rats (21). Accordingly, notwithstanding other possible mechanisms by which K depletion may increase the already augmented$J_{{\text{tCO}}_2}$in surviving nephrons (see below), we hypothesized that this effect would be based, at least in part, on enhanced H⁺-ATPase activity. Indeed, our results show that, after 7 days of dietary K depletion,1) surviving DT significantly increase$J_{{\text{tCO}}_2}$and 2) this effect is H⁺-ATPase dependent, as demonstrated by the sensitivity to concanamycin A inhibition and by increased H⁺-ATPase immunogold deposition on apical membranes of A-type intercalated cells.

K Depletion Enhances DT H⁺-ATPase Activity in Nx Rats

We have previously shown in the normal rat cortical collecting duct that dietary K depletion dramatically increases the proportion of A-type intercalated cells with studded double plasma membranes on their apical surfaces (21). These studs are the morphological correlate of proton pumps and presumably derive from insertion of subapical vesicles into the proliferated microplicae. Although K depletion in Nx rats elicited neither increased H⁺-ATPase expression (Table 5, Fig. 3) nor enhanced H⁺-ATPase antibody immunofluorescence as seen by light microscopy, K depletion was associated with important changes in A-type intercalated cells. Figure 4 illustrates the enhanced cellular hypertrophy and increased elaboration of apical microplicae in A-type intercalated cells from Nx-K vs. Nx rats. In addition, Nx-K rats demonstrated significantly increased deposition of H⁺-ATPase immunogold particles, compared with Nx alone (Fig. 5). This increased polarization of H⁺-ATPase to the luminal surface of A-type intercalated cells is consistent with the view that the augmented$J_{{\text{tCO}}_2}$in Nx-K rats vs. Nx rats is proton pump dependent. These morphological observations are also supported by the ability of luminally perfused concanamycin A to abrogate the increment in Nx-K$J_{{\text{tCO}}_2}$above that seen in Nx alone (from 65 to 102 pmol ⋅ min⁻¹ ⋅ mm⁻¹). Moreover, this concanamycin A-dependent decrease in Nx-K$J_{{\text{tCO}}_2}$is almost twice that observed in Nx rats (16), suggesting that H⁺-ATPase sustained a much greater portion of the bicarbonate flux after K depletion.

Possible Role of Other Transporters in K-Depleted Nx Rats

Notwithstanding the substantial evidence supporting a role for H⁺-ATPase in both Nx and Nx-K$J_{{\text{tCO}}_2}$, it is clear that more than half the bicarbonate flux is not suppressible by concanamycin A, suggesting the possibility that sodium/hydrogen exchange (NHE) and/or H⁺-K⁺-ATPase activity may contribute as well. Eiam-ong et al. (8) have demonstrated an 88% increase in H⁺-K⁺-ATPase in cortical collecting ducts of unilaterally nephrectomized rats, whereas, in intact but K-depleted rats perfused in vivo, Wang et al. (27) showed a Sch-28080-suppressible increase in late DT$J_{{\text{tCO}}_2}$. We evaluated the role of H⁺-K⁺-ATPase in Nx and Nx-K rats by luminally perfusing 10⁻⁵ M Sch-28080. Although this dose of Sch-28080 was sufficient to reduce$J_{{\text{tCO}}_2}$in acid-loaded rats by half (17) and to abolish unidirectional bicarbonate reabsorption in rats with chloride-depletion metabolic alkalosis (18), in the present experiments, there was no discernible reduction in$J_{{\text{tCO}}_2}$in either Nx or Nx-K rats. In fact, although not statistically significant, there appears to be a Sch-28080-induced tendency to increase$J_{{\text{tCO}}_2}$in Nx-K rats (seeresults).

Nevertheless, our present results do not exclude a role for H⁺-K⁺-ATPase- mediated bicarbonate reabsorption in Nx rats with or without K depletion. Certainly, there is evidence for both enhanced apical A-type intercalated cell H⁺-K⁺-ATPase activity in the intact K-depleted rat (28) and for the existence of different kidney H⁺-K⁺-ATPase isoforms (12, 28). Indeed, the colonic α-isoform of the H⁺-K⁺-ATPase, whose message is upregulated in kidneys in rats subjected to chronic hypokalemia (7), is Sch-28080 insensitive when expressed in a heterologous system (5, 6, 14). Accordingly, it is quite possible that the Sch-28080-resistant colonic H⁺-K⁺-ATPase α-isoform is active in surviving distal tubules of our Nx and/or Nx-K rats. Furthermore, the colonic α-subunit can form an active complex with either the gastric β-isoform or the β₁-subunit of the Na⁺-K⁺-ATPase (5). To gain further insight into isoform expression in our Nx or Nx-K rats, we carried out additional immunochemical studies that indicate that the gastric β-isoform is not upregulated to detectable levels in the Nx or Nx-K kidney. Therefore, the β₁-subunit of the Na⁺-K⁺-ATPase is a candidate partner to function with the colonic subunit and mediate the Sch-28080-resistant increase in$J_{{\text{tCO}}_2}$in Nx and Nx-K rats. In our previous study on Nx rats (16), we presumed there existed an ANG II-stimulated NHE component to$J_{{\text{tCO}}_2}$(15), and we were therefore surprised to find sensitivity to losartan combined with resistance to different amiloride analogs. The demonstration by Tse et al. (26) of an amiloride-resistant epithelial NHE isoform makes plausible the notion that in DT cells in both Nx and Nx-K rats, unrecognized transporters may emerge that exhibit resistance to various inhibitors. Finally, we have already shown (17) in rats with chronic severe metabolic acidosis, a brisk DT$J_{{\text{tCO}}_2}$, which is insensitive to amiloride, Sch-28080, and bafilomycin A₁.

Moreover, it is conceivable that when luminal H⁺-ATPase activity is blocked by concanamycin A, a known isoform of NHE may sustain a larger portion of the bicarbonate reabsorptive flux. Conversely, perhaps during perfusion with amiloride or Sch-28080, proton secretion is diverted and sustained by H⁺-ATPase. In short, it is possible that the components supporting the$J_{{\text{tCO}}_2}$of 102 pmol ⋅ min⁻¹ ⋅ mm⁻¹in Nx-K rats undergo dynamic allocation from one transporter to another. Simultaneous perfusion of multiple inhibitors in both Nx and Nx-K rats would test this proposal, and such studies are currently underway in our laboratory.

Role of the Renin-Angiotensin System

It is clear that K depletion in the rat stimulates renal renin release (19, 22). Accordingly, it appeared possible from the outset that, in Nx-K rats,$J_{{\text{tCO}}_2}$may be stimulated by an ANG II-dependent process via an enhanced intrarenal renin substrate mechanism. Indeed, in Nx rats, we clearly demonstrated that a significant portion of the enhanced bicarbonate flux was AT₁ receptor dependent, as evidenced by its suppression by both luminal and intravenous losartan (16). Surprisingly, luminal perfusion of losartan in Nx-K rats was without effect.

Assuming intrarenal ANG II levels are not decreased in Nx-K rats, could there be a losartan-sensitive component present in Nx rats that does not persist after superimposed K depletion? Linas et al. (20) have shown that, for cultured vascular smooth muscle cells, steady-state uptake of ANG II is decreased, despite an increase in the number of surface receptors. This effect may be related to changes in intracellular pH during K depletion and/or alteration in ANG II receptor internalization and recycling (20). It is perhaps more relevant that, in very young rats, after 2 wk of dietary K depletion, plasma renin activity increased sixfold, whereas there was a decrease in ANG II receptor density in both kidney cortex and medulla (22). Thus it is possible that, in Nx-K rats, changes to distal tubule luminal AT₁ receptor density and/or the signaling cascade subsequent to ANG II/AT₁ binding may account for the apparent resistance to losartan.

In summary, we evaluated the hypothesis that surviving distal tubules in K-depleted nephrectomized rats may increase bicarbonate reabsorption by enhanced H⁺- ATPase activity. Our results demonstrate a marked augmentation of$J_{{\text{tCO}}_2}$, which is inhibitable by luminal perfusion of concanamycin A, resistant to inhibition by losartan, or Sch-28080 and associated with further A-type intercalated cell hypertrophy and enhanced H⁺-ATPase immunogold labeling in apical microplicae. Accordingly, these combined in vivo, morphometric, and immunogold studies confirm that K depletion increases$J_{{\text{tCO}}_2}$in substantial part by stimulating H⁺-ATPase activity, independent of the AT₁ receptor. It is possible that other transporters, such as a Sch-28080-resistant colonic H⁺-K⁺-ATPase, may also contribute to the enhanced flux.