INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

ACTIVE PROTON SECRETION by the collecting duct (CD) is coupled, in part, to active K⁺ absorption via a membrane-bound adenosinetriphosphatase (H⁺-K⁺-ATPase) (24, 32). The H⁺-K⁺-ATPase (HKA) that is expressed in renal CD under normal conditions shows striking molecular, biochemical, and physiological similarities to the HKA found in gastric parietal cells, which is responsible for the secretion of acid into the gastric lumen (21, 24, 32). This exchanger has been referred to as gastric HKA (HKAg) and is sensitive to inhibition by SCH-28080 (24, 32). A distinct but structurally related HKA is expressed in the distal colon, which mediates active K⁺ reabsorption (6, 15). This transporter is called colonic HKA (HKAc) or nongastric HKA and is sensitive to inhibition by ouabain (24). Molecular studies indicate expression of both HKAg and HKAc in CD (24, 32). Functionally they are thought to be involved in H⁺ secretion and/or K⁺ reabsorption but appear to be regulated differentially (24, 32). Furthermore, HKAc may also mediate the exchange of extracellular K⁺ for intracellular Na⁺ or NH (5, 11), consistent with this transporter working as a cation/K⁺ exchange mode.

To better understand the role of HKAg in acid-base regulation, transgenic mice deficient in this gene were examined. HKAg null mice have severe achlorhydria (25), consistent with the role of this transporter as the major acid-secreting process in the parietal cells of the stomach. The HKAg-deficient mice did not show any significant abnormality in systemic acid-base balance or serum potassium (25), despite the fact that this transporter plays an important role in net bicarbonate reabsorption in the collecting ducts of mouse kidney (12, 13). These results suggest that another acid-secreting transporter(s) is likely upregulated in the kidneys of HKAg null mice. To explore this issue further, renal cortical collecting ducts (CCDs) of wild-type or HKAg knockout mice were isolated and perfused, and their - and -intercalated cells (ICs) were examined. The results indicate that a novel exchanger, which is distinct from HKAc, is upregulated in HKAg null mice and maintains the K⁺-dependent H⁺ secretion at a comparable level to wild-type animals.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animals. Wild-type and HKAg knockout mice were used for these experiments. HKAg knockouts were described recently (25). Animals were allowed free access to food and water and were studied at 3-6 mo of age. For potassium depletion studies, animals were placed on a potassium-free diet (11) for 19 days.

Isolation of CCDs. Animals were killed by intraperitoneal injection of 50 mg/kg pentobarbital sodium. Kidneys were quickly removed and placed in ice-cold dissection medium (solution 1, Table 1). Thin (~1 mm) slices were obtained and transferred to the dissection chamber. The temperature of the dissection medium in the chamber was kept at 4°C. Bovine serum albumin, 0.1%, was added to the dissection medium to prevent sticking of dissected tubules to the glass and forceps. CCDs were obtained by free-hand dissection out of cortical medullary rays. Tubules were measured using an eyepiece micrometer and generally were 0.3-0.5 mm in length. CCDs were distinguished from nearby proximal straight tubules by their approximately one-third smaller diameter and more turbid appearance compared with the "ground-glass" appearance of the proximal straight tubules. Thick ascending limbs, on the other hand, are thinner than CCDs and have a fragile, homogeneous, and slightly shiny appearance in reflected light.

In vitro microperfusion. Dissected tubules were quickly transferred to a 1.5-ml temperature-controlled specimen chamber mounted on an inverted Zeiss Axiovert S-100 microscope (Carl Zeiss, Thornwood, NY). Tubules were perfused using concentric glass pipettes according to the method of Burg and Green (1) with modifications (19, 27) at 4.5 cmH₂O pressure. Solutions that were used to perfuse and bathe the tubules are listed in Table 1. All solutions were delivered to the specimen chamber in CO₂- and O₂-impermeable tubing (Cole Palmer, Chicago, IL) by a peristaltic pump (Peristar, WPI, Sarasota, FL) at a rate of 1 ml/min. The chamber had a lid to minimize evaporation and heat loss and maintain constant gas pressure and pH. Chamber pH was frequently checked on a model B 213 Horiba pH meter that allows measurement of pH in small samples.

Intracellular pH measurements. After 20-min equilibration in solution 2, each tubule was perfused with 6 µM of the fluorescent pH indicator bis(carboxyethyl)-5(6)-carboxyfluorescein acetoxymethyl ester (BCECF-AM) for 15 min. When BCECF-AM is introduced from the luminal side of a CCD, intercalated, but not principal, cells take up the dye (29). Charged and fluorescent molecules of the dye are trapped in the cell after cleavage of the ester moiety by cytoplasmic esterases. Another 10-15 min were allowed for the dye washout. Fluorescent measurements were done on a Zeiss Axiovert S-100 inverted microscope equipped with the Attofluor RatioVision Digital Imaging System (Attofluor, Rockville, MD). Achroplan ×40/0.8 water objective with 3.6-mm working distance was used. Excitation wavelengths were alternated between 488 and 440 nm, and emission was measured at 520 nm. Background fluorescence was measured before the dye was introduced to the tubule and automatically subtracted from all subsequent measurements. The Attofluor Digital Imaging System allows for the control of light source intensity. Balance between the light source intensity and the camera gain optimizes fluorescent imaging while at the same time applies the smallest light intensity that is sufficient to excite the dye. This helps to minimize photobleaching and photodamage of the tubule cells, which can be substantial (29).

Experimental procedures. After a stable baseline cell pH reading in bicarbonate-buffered solution (solution 2) was achieved, the bath solution was switched to a chloride-free solution (solution 3) to distinguish between the IC types. As shown previously (14, 18, 20, 30), removal of bath Cl causes alkalinization of -ICs due to blocking and/or reversing the basolateral Cl/HCO exchanger. At the same time, -ICs acidify, because this maneuver stimulates HCO exit via the apical Cl/HCO exchanger. Basolateral Cl conductance (14, 18, 20, 30) of the -ICs allows for intracellular Cl to leave the cell on basolateral Cl removal. This adds to the apical membrane Cl gradient and stimulates apical Cl/HCO exchange and HCO exit. Both the alkalinization of -ICs as well as the acidification of -ICs is reversed on addition of Cl to the bath.

RNA isolation and Northern hybridization. Total cellular RNA was extracted from whole kidney of wild-type or HKAg null mice (on normal or K⁺-free diet for 19 days) by the method of Chomczynski and Sacchi (2). In brief, 0.5-1 g of tissue was homogenized at room temperature in 10 ml Tri Reagent (Molecular Research Center, Cincinnati, OH). RNA was quantitated by spectrophotometry and stored at 80°C. RNA samples were fractionated on a 1.2% agarose-formaldeyde gel. The samples were transferred to a nylon membrane, cross-linked by ultraviolet light, and baked for 1 h. Hybridization was performed according to Church and Gilbert (3). The cDNA probes were labeled with ³²P-deoxynucleotide using the RadPrime DNA labeling kit (Life Technologies). After hybridization, the membranes were washed, blotted dry, exposed to a PhosphorImager cassette at room temperature for 24-48 h, and read by PhosphorImager (Molecular Dynamics). The K⁺-free diet protocol was similar to previous studies reported from our laboratories (11). For HKAc, three PCR products from the rat -subunit cDNA (nucleotides 135-515, 2369-2998, and 3098-3678; Ref. 6) were pooled and used as an isoform-specific probe.

Chemicals. All chemicals were obtained from Sigma unless specified otherwise. EIPA was dissolved in methanol as a 40 mM stock on the day of an experiment and diluted 1:1,000 for the final concentration of 40 µM. A stock solution of bafilomycin A₁ was made fresh for each experiment in methanol, and 1:1,000 dilution was used to make the final concentration of 170 nM. Ouabain was directly added and dissolved in appropriate solutions at a final concentration of 1.5 mM. This concentration was used on the basis of studies indicating that HKAc activity was inhibited partially by 1 mM ouabain (24). Nigericin was dissolved in ethanol as a 10 mM stock and diluted 1:1,000 for the final concentration of 10 µM. BCECF-AM was obtained from Molecular Probes (Eugene, OR) and kept frozen in small aliquots of stock in DMSO. It was diluted 1:1,000 for the final concentration of 6 µM. SCH-28080 was kept frozen in small aliquots of 100 mM stock in methanol and was used with a dilution of 1:10,000 in the final concentration of 10 µM.

Statistics. Results are give as means ± SE. Statistical comparisons between the groups were performed according to Student's t-test for nonpaired data. The data were considered significant if P < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

K+-dependent intracellular pH recovery in -ICs and -ICs in wild-type mouse CCDs. The first set of experiments was designed to examine K⁺-dependent luminal proton secretion in each of the IC types ( and ) in CCDs in wild-type animals. A representative tracing is shown in Fig. 1. At the beginning of each experiment, IC type was determined by basolateral Cl removal in the presence of HCO. The same cells were then followed throughout the experiment. Nine of 19 cells in 5 CCDs of wild-type mice alkalinized on basolateral Cl removal [change in intracellular pH (pH_i) = 0.21 ± 0.01] and were therefore considered -ICs. Ten of 19 cells in 5 CCDs acidified on basolateral Cl removal (pH_i = 0.26 ± 0.02) and were considered -ICs.

View larger version (21K): [in this window] [in a new window]   Fig. 1.   Representative tracings demonstrate the effect of 5 mM extracellular K+ on intracellular pH (pHi) recovery from an acute acid load in intercalated cells (ICs) in wild-type (WT) mouse. Upon identification of IC type (see METHODS and first part of tracing), HCO was removed from lumen and bath for the rest of the experiment to render both luminal and basolateral HCO transporters inactive. In the presence of bafilomycin A1 on both sides (an inhibitor of H+- ATPase) and ethylisopropylamiloride (EIPA) in the bath (an inhibitor of basolateral Na+/H+ exchange), removal of K+ from lumen blocked pHi recovery from an acid load. The return of 5 mM K+ to the lumen resulted in a significant pHi recovery from acidosis in both cell types. The rate of K+-dependent pHi recovery was similar in both cells: 0.069 pH units/min in -ICs and 0.065 pH units/min in -ICs. Removal of all inhibitors allowed for faster recovery at the rate of 0.142 pH units/min in -IC and 0.129 pH units/min in -ICs (to 7.27 and 7.28, respectively).

Effect of SCH-28080 on K+-dependent pH_i recovery in -ICs and -ICs in wild-type mouse CCD. Using the same protocol as described above, we also examined the effect of HKAg inhibitor SCH-28080 on K⁺-dependent pH_i recovery in -ICs and -ICs of wild-type mouse CCD. A representative tracing is shown in Fig. 2, and the results are summarized in Fig. 3. In this group of experiments, 7 of 16 cells in 6 CCDs alkalinized on basolateral Cl removal (pH_i = 0.25 ± 0.02) and were therefore considered -ICs. Nine of 16 cells in 6 CCDs acidified on basolateral Cl removal (pH_i = 0.24 ± 0.02) and were considered -ICs. In these experiments, the NH₄Cl prepulse acidified -ICs from a baseline pH_i of 7.26 ± 0.035 to a nadir pH_i of 6.64 ± 0.047 (n = 7 cells in 6 CCDs) and -ICs from a baseline pH_i of 7.33 ± 0.027 to a nadir pH_i of 6.49 ± 0.05 (n = 9 cells in 6 CCDs). In the absence of K⁺ in the lumen, no pH_i recovery was observed in either cell type. As demonstrated in Figs. 2 and 3, the presence of 10 µM SCH-28080 in the lumen blocked the K⁺-dependent pH_i recovery by >80% in both IC cell types. Rates of K⁺-dependent pH_i recovery were 0.019 ± 0.005 and 0.012 ± 0.004 pH units/min in -ICs and -ICs, respectively (P < 0.001, compared with the rates of recovery in the absence of SCH-28080). The HKAg is highly sensitive to SCH-28080, with the ATPase activity inhibited by >90% in the presence of 1 µM SCH-28080 (reviewed in Ref. 24). It has been shown that at concentrations >10 µM, SCH-28080 can inhibit other transporters such as Na⁺-independent H⁺-ATPase (16). Furthermore, recent data demonstrated that prolonged exposure of kidney CD cells to SCH-28080 as well as high concentrations of SCH-28080 cause ATP depletion (4), which in turn can have nonspecific inhibition of HKA activity. To avoid these complications, we did not try higher concentrations of SCH-28080. On removal of SCH-28080 and other inhibitors, all cells recovered to their baseline pH_i at the rate of 0.275 ± 0.02 pH units/min in -ICs and 0.253 ± 0.022 pH units/min in -ICs. In summary, the results in Figs. 1-3 demonstrate comparable K⁺-dependent, SCH-28080-sensitive H⁺-secreting activity in the apical membrane of both -ICs and -ICs and are consistent with the activity of luminal HKAg in both cell types.

View larger version (21K): [in this window] [in a new window]   Fig. 2.   Representative tracings demonstrating the effect of Schering (SCH)-28080 on K+-dependent pHi recovery in ICs of WT mouse. The identification of IC type is demonstrated in the first part of the tracings. HCO was removed from lumen and bath for the rest of the experiment to render HCO-dependent transporters inactive. In the presence of bafilomycin A1 on both sides and EIPA in the bath, removal of K+ from lumen blocked pHi recovery from an acid load. SCH-28080 blocked the K+-dependent pHi recovery in IC cells. Removal of the inhibitors allowed for fast recovery at the rate of 0.211 pH units/min in -IC and 0.223 pH units/min in -IC.

View larger version (12K): [in this window] [in a new window]   Fig. 3.   SCH-28080 inhibition of K+-dependent pHi recovery from acid load of ICs in WT mice CCD (summary of results). The effect of luminal K+ and SCH-28080 on the rate of pHi recovery in - or -IC in wild-type mice is shown.

K+-dependent pH_i recovery in -ICs and -ICs in HKAg knockout mouse CCD. In the next series of experiments, we examined K⁺-dependent pH_i recovery in both IC cell types in HKAg knockout mice CCDs. On the basis of their pH_i response to basolateral Cl removal, 12 of 25 cells (in 6 CCDs) were identified as -ICs, and the remaining 13 were identified as -ICs. In the presence of HCO, baseline pH_i was comparable in -ICs of HKAg knockout (7.353 ± 0.004, n = 12) and wild-type animals (7.346 ± 0.002, n = 9) (P > 0.05). Interestingly, basal pH_i in -ICs was lower in HKAg knockout animals (7.265 ± 0.005, n = 13) compared with wild-type animals (7.326 ± 0.003, n = 10) (P < 0.05). Absence of HCO did not significantly alter the baseline pH_i levels (data not shown).

View larger version (22K): [in this window] [in a new window]   Fig. 4.   Representative tracings demonstrating the effect of 5 mM K+ on pHi recovery from acid load in ICs of gastric H+-K+-ATPase (HKAg) null mice. After the identification of - and -cells (as demonstrated in the first part of the tracing), HCO was removed from the experiments to render HCO transporters inactive. In the presence of bafilomycin A1 on both sides and EIPA in the bath, removal of K+ from lumen prevented pHi recovery from acidosis. The return of 5 mM K+ to the lumen resulted in significant pHi recovery (to ~6.88) in both cells. Rate of K+-dependent recovery was 0.083 pH units/min in -IC and 0.066 pH units/min in -ICs. Removal of the inhibitors allowed for additional recovery at the rate of 0.142 pH units/min in -IC and 0.143 pH units/min in -IC.

View larger version (12K): [in this window] [in a new window]   Fig. 5.   Rate of K+-dependent pHi recovery from acute acid load in ICs of WT and HKAg knockout (KO) mice (summary of results). The effect of luminal K+ on the rate of pHi recovery in - or -IC of WT and KO mice is shown.

Effect of SCH-28080 on K+-dependent pH_i recovery in -ICs and -ICs in HKAg knockout mouse CCD. To characterize the K⁺-dependent H⁺ secretion in the CCDs of HKAg knockout mice, we sought to examine its sensitivity to SCH-28080 or ouabain. Data from the experiments examining the effect of SCH-28080 are shown as a representative tracing in Fig. 6 and are summarized in Fig. 7. Based on their pH_i response to bath Cl removal, 11 of 22 cells in 5 CCDs were identified as -ICs, and 11 cells were considered -ICs. The NH₄Cl prepulse acidified -ICs from a baseline pH_i of 7.29 ± 0.027 to a nadir pH_i of 6.64 ± 0.019 (n = 11 cells). -ICs acidified from a baseline pH_i of 7.27 ± 0.032 to a nadir pH_i of 6.67 ± 0.027 (n = 11 cells). In the absence of K⁺ in luminal perfusate, no pH_i recovery was observed in either cell type. However, and contrary to wild-type animals, SCH-28080 did not block the pH_i recovery in the presence of 5 mM potassium in the lumen. This phenomenon was observed in both IC cell types. The rate of K⁺-dependent H⁺ secretion in the lumen of HKAg knockout mice CCDs was 0.097 ± 0.005 pH units/min in -ICs (n = 11) and 0.094 ± 0.007 pH units/min in -ICs (11 cells in 5 CCDs). Compared with the cells in tubules of knockout mice perfused in the absence of SCH-28080, no significant difference was noted (P > 0.05 with or without SCH-28080), indicating that SCH-28080 did not affect the K⁺-dependent pH_i recovery in ICs of knockout mice CCDs. pH_i recovered to 6.99 ± 0.03 and 7.05 ± 0.03 in -ICs and -ICs, respectively. Removal of the inhibitors resulted in additional recovery to the baseline pH at the rate of 0.251 ± 0.012 pH units/min in -ICs and 0.243 ± 0.006 pH units/min in -ICs.

View larger version (21K): [in this window] [in a new window]   Fig. 6.   Representative tracings demonstrating the effect of SCH-28080 on K+-dependent pHi recovery from acid load in ICs from a HKAg KO mice. H+-ATPase was blocked with bafilomycin A1 on both sides, and basolateral Na+/H+ exchange was blocked with EIPA. Interestingly, the presence of SCH-28080 did not prevent the K+-dependent pHi recovery. The cell recovered to pHi ~6.8. Rate of K+-dependent recovery was 0.078 pH units/min. Removal of the inhibitors allowed for additional recovery at the rate of 0.301 pH units/min to pHi 7.34.

View larger version (12K): [in this window] [in a new window]   Fig. 7.   SCH-28080 does not inhibit K+-dependent pHi recovery from acid load in ICs of HKAg KO mice (summary of results). The effect of luminal K+ and SCH-28080 on the rate of pHi recovery in - or -IC of HKAg KO mice is shown.

Effect of ouabain on K+-dependent pH_i recovery in -ICs and -ICs in HKAg null mouse CCD. The purpose of the next series of experiments was to examine the effect of ouabain on K⁺-dependent pH_i recovery in CCDs of HKAg knockout mice. Figure 8 is a representative tracing, and Fig. 9 summarizes the results. In this group of experiments, 4 of 11 cells (from 2 CCDs) were identified as -ICs, and 7 of 11 cells were considered -ICs. The NH₄Cl prepulse acidified -ICs from a baseline pH_i of 7.33 ± 0.031 to a nadir pH_i of 6.79 ± 0.005 (n = 4). -ICs acidified from a baseline pH_i of 7.22 ± 0.03 to a nadir pH_i of 6.71 ± 0.03 (n = 7). Similar to previous experiments, in the absence of K⁺ in the lumen, no pH_i recovery was observed in either cell type. In the presence of 1.5 mM ouabain in the lumen, the rate of K⁺-dependent pH_i recovery in IC cells of HKAg knockout mice remained unchanged vs. no ouabain. As indicated in Fig. 9, the pH_i recovery rate in the presence of ouabain was 0.099 ± 0.007 pH units/min in -ICs (n = 4) and 0.097 ± 0.005 pH units/min in -ICs (n = 7) of the HKAg knockouts. These were not statistically different vs. no ouabain in -ICs and -ICs, respectively (P > 0.05) (Fig. 9). Cells recovered to 7.06 ± 0.02 in -ICs and 6.99 ± 0.03 in -ICs. Removal of inhibitors resulted in additional recovery to baseline pH_i levels at the rate of 0.236 ± 0.013 pH units/min in -ICs and 0.234 ± 0.01 pH units/min in -ICs.

View larger version (20K): [in this window] [in a new window]   Fig. 8.   Representative tracings demonstrate the effect of ouabain on K+-dependent pHi recovery in ICs from HKAg KO mice. H+-ATPase was blocked with bafilomycin A1 on both sides, and basolateral Na+/H+ exchange was blocked with EIPA. Under these conditions, K+ removal from lumen prevented pHi recovery from cell acidosis. Presence of ouabain at 1.5 mM did not inhibit the K+-dependent pHi recovery from acidosis. pHi recovered to ~7.2 in the presence of ouabain. Rate of K+-dependent recovery was 0.13 ph units/min in -IC and 0.105 pH units/min in -ICs. Removal of the inhibitors allowed for additional recovery at a rate of 0.178 pH units/min in -IC and 0.16 pH units/min in -ICs to pHi of 7.34 and 7.35, respectively.

View larger version (13K): [in this window] [in a new window]   Fig. 9.   Ouabain does not inhibit K+-dependent pHi recovery from acid load in ICs of HKAg KO mice (summary of results). The effect of ouabain and luminal K+ on the rate of pHi recovery in HKAg KO mouse is shown.

RT-PCR and Northern hybridization studies in wild-type and HKAg null mouse. The gene-targeting strategy eliminated sequences encoding the catalytic phosphorylation site, which is essential for enzyme activity. In the stomach, the wild-type HKAg mRNA was eliminated (25), although there were trace levels of an ~1-kb mRNA encoding some of the NH₂-terminal sequences. To test for the presence of HKAg mRNAs in kidneys of wild-type and mutant mice, RT-PCR analysis was performed using primers from exons 6 and 10. As shown in Fig. 10, a PCR product of the appropriate size (653 bp) was identified in wild-type kidney, but not in knockout kidney. These results confirm that these critical sequences, which span the major catalytic domain and the region required for apical sorting (10) of the pump, are absent in RNA from knockout kidney.

View larger version (40K): [in this window] [in a new window]   Fig. 10.   RT-PCR analysis of RNA for the presence of HKAg. RT-PCR analysis was performed on total RNA isolated from kidneys of WT and HKAg KO animals using primers from exons 6 and 10. PCR products were size fractionated on 1.2% agarose gel and visualized by ethidium bromide staining. The results demonstrate a PCR product of expected size (653 bp) for mouse HKAg in the kidneys of WT but not HKAg KO animals. The specificity of the reaction is demonstrated by the absence of an RT product in the absence of reverse transcriptase (RT). Bottom panel shows constitutive control -actin.

View larger version (43K): [in this window] [in a new window]   Fig. 11.   Northern hybridizations of colonic HKA (HKAc) in whole kidney of WT and HKAg KO mice. As indicated, HKAc mRNA levels remained undetectable in WT and HKAg KO animals on a normal diet (control groups). Both WT and HKAg KO mice upregulated their HKAc mRNA in potassium depletion. The 28S rRNA is shown in the bottom of the blot as an index of RNA loading. RNA (30 µg/lane) was loaded in each lane.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The present studies examine the K⁺-dependent luminal acid extrusion in CCDs of wild-type and HKAg null mice. HKAg null mice demonstrate an H⁺/K⁺ exchange activity that is comparable to wild-type animals (Figs. 1-5). Whereas SCH-28080 inhibits H⁺/K⁺ exchange activity in wild-type animals, H⁺/K⁺ exchange activity in the CCDs of null mice is resistant to inhibition by SCH-28080 (Figs. 6 and 7). H⁺/K⁺ exchange activity in null mice was also insensitive to inhibition by ouabain, a known inhibitor of HKAc (Figs. 8 and 9). Northern hybridizations did not demonstrate any upregulation of HKAc in HKAg null mice (Fig. 11).

HKAg is expressed in the CDs of mammalian kidneys, as supported by molecular, biochemical, and functional studies (24, 32). Studies on the inhibitor profile of HKAg demonstrate that this transporter is inhibited by SCH-28080 but not ouabain (24). Microperfusion studies in rabbit and rat kidney indicate that a significant portion of K⁺ and HCO reabsorption in CCD is mediated via the SCH-28080-sensitive HKAg (24, 32). Studies in microperfused mouse CCD demonstrate that ~50% of HCO reabsorption is mediated via a K⁺-dependent process that is inhibited by SCH-28080 (12, 13). Taken together, these studies indicate that HKAg is responsible for a significant portion of HCO reabsorption in the mouse kidney CD.

The present studies demonstrate that -IC and -IC in mouse CCDs express apically oriented K⁺-dependent, SCH-28080-sensitive acid-secreting activity, strongly suggestive of functional HKAg. This is in agreement with functional studies in split-open rat and rabbit CCDs and microperfused rabbit CCDs (22, 23, 31). Interestingly, baseline pH_i in -ICs of HKAg knockout mice was significantly lower compared with wild-type animals, whereas it was comparable in -IC cells (see RESULTS). On the basis of present data, we cannot be sure about what causes this difference in -ICs or why there is not a difference in the basal pH_i in -ICs. We can only speculate that the lower basal pH_i in -ICs in HKAg knockout mice raises the possibility that these cells are either pumping HCO out at higher rates or a H⁺-translocating transporter gets downregulated. Although both - and -ICs seem to have apically oriented HKAg, it may serve different functions in these two cell types, or the way two cell types compensate for the lack of HKAg may be different.

HKAg null mice did not display any significant abnormality in systemic acid-base balance or serum K⁺ under baseline conditions (25). This suggests that acid-base transporter(s) distinct from HKAg is upregulated in the CDs of HKAg null mice. Comparable levels of K⁺-dependent pH_i recovery in the CCDs of both animal groups are consistent with this hypothesis. Interestingly, the K⁺-dependent H⁺ secretion in the CCDs of HKAg null mice is not inhibited by classical HKA inhibitors such as SCH-28080 or ouabain, indicating the existence of a novel, yet unrecognized isoform of HKA in the kidney.

With respect to nongastric HKAs, HKAc is the well-known SCH-28080-insensitive, ouabain-sensitive isoform (7, 32). HKAc expression did not increase in HKAg null mice, indicating that this transporter is not responsible for the upregulation of HKA activity in this animal. Furthermore, the ouabain insensitivity of HKA activity in HKAg null mice points to an isoform distinct from HKAc. Recent functional studies in the colon have described the presence of another HKA isoform, which is insensitive to ouabain (9). No distinct HKA molecule that is insensitive to both SCH-28080 and ouabain has been identified.

Studies on HKAs in the kidney point to the discrepancies in the inhibitor profile of "nongastric" HKAs in heterologous expression systems and native tissues and have concluded that the presence of a novel isoform may account for the discrepancies (7, 24). Studies in cultured kidney cells (4) have questioned some of the inhibitor profile studies by demonstrating nonspecific effects of high concentrations of SCH-28080 on ATPase activity. To avoid any nonspecific effect of SCH-28080 on HKA transporters, the exposure time of perfused tubules to SCH-28080 in the present experiments was kept short.

Recently, Laroche-Joubert et al. (8) examined the properties of three functional K⁺-ATPases isoforms in microdissected rat nephron segments. They concluded that type II and III K⁺-dependent ATPase activities exhibit different sensitivity profiles to SCH-28080 and ouabain compared with the type I renal K⁺-ATPase. Type I K⁺-dependent ATPase is sensitive to SCH-28080 but is insensitive to ouabain and resembles HKAg. Type III is sensitive to ouabain but is insensitive to SCH-28080. They further found that pharmacological properties and tubular localization of type III K⁺-ATPase are not compatible with that of HKAc. They suggested that a new kidney HKA isoform might exist that may fit the properties of type III K⁺-ATPase. It is worth mentioning that an HKA that is insensitive to both SCH-28080 and ouabain has not been described in kidney epithelial cells. As such, the molecule mediating the H⁺/K⁺ exchange in apical membrane of ICs in HKAg knockout mice is distinct from all HKA isoforms described so far.

In conclusion, our results suggest that a novel acid-base transporter, distinct from HKAc, is upregulated in HKAg null mice and maintains the K⁺-dependent proton secretion at a comparable level to wild-type animals. This may account for the lack of any acid-base abnormality in HKAg null mice.