NBCn1 is a basolateral cotransporter in rat kidney inner medullary collecting ducts

【摘要】  Primary cultures of rat inner medullary collecting duct (IMCD) cells Na + dependently import across the basolateral membrane through an undefined transport protein. We used RT-PCR, immunoblotting, and immunohistochemistry to identify candidate proteins for this basolateral cotransport. The mRNA encoding the electroneutral cotransporter NBCn1 was detected as the only cotransporter in the rat inner medulla (IM) among the five characterized Na + -dependent transporters. The mRNA of a yet uncharacterized transporter-like protein, BTR1, was also present in the IM, but its expression in microdissected tubules seemed restricted to the thin limbs of Henle's loop. Immunoblotting confirmed the presence of NBCn1 as an 180-kDa protein of the rat IM. Anti-NBCn1 immunolabeling was confined to the basolateral plasma membrane domain of IMCD cells in the papillary two-thirds of the IM. Consistent with the presence of NBCn1, IMCD cells possessed stilbene-insensitive, Na + - and -dependent pH recovery after acidification, as assessed by fluorescence microscopy using a pH-sensitive intracellular dye. In furosemide-induced alkalotic rats, NBCn1 protein abundance was decreased in both the IM and inner stripe of outer medulla (ISOM) as determined by immunoblotting and immunohistochemistry. In contrast, NBCn1 abundance in the IM and ISOM was unchanged in NaHCO 3 -loaded animals, and the NBCn1 abundance increased only in the ISOM after NH 4 Cl loading. In conclusion, NBCn1 is a basolateral cotransporter of IMCD cells and is differentially regulated in IMCD and medullary thick ascending limb.

acid-base balance; hydrogen-ATPase; bicarbonate transport; bicarbonate metabolism; immunohistochemistry; intracellular pH; 2',7'-bis(2-carboxyethyl)-5(6)-carboxyfluorescein

【关键词】  basolateral cotransporter medullary collecting

SEVERAL KEY FUNCTIONS IN TERMS of fluid and electrolyte homeostasis are associated with the collecting duct of the mammalian kidney ( 7 ). Among these are the regulated reabsorption of water and Na + and the adjustment of urinary pH. In particular, fine regulation of the acid-base balance takes place in the collecting ducts by either net H + or secretion, and these processes are mediated by several acid-base transporters. In the cortical and outer medullary collecting duct (OMCD) and in the initial part of the inner medullary collecting duct (IMCD), transport of acid-base equivalents is thought to be mediated primarily by intercalated cells through the H + -AT-Pase, the K + -H + -ATPase, the anion exchangers AE1 and pendrin, as well as perhaps AE4 and an electroneutral cotransporter, NBC3 ( 12, 13, 16, 18, 2 ). However, the distal part of the IMCD is devoid of intercalated cells ( 3 ), and the acid-base transport proteins expressed in these IMCD cells may well differ from those of the intercalated cells in the cortex, outer medulla, and initial part of the inner medulla. Basolateral Na + -dependent transport has been reported in primary cultures of rat IMCD cells ( 5 ). cotransport was more likely to be involved than an Na + -dependent exchanger, because transport was unaffected by depletion of intracellular Cl -.

Na + -dependent cotransport is maintained by proteins that belong to one superfamily of transporters ( 1 ). The SLC4A family consists of electrogenic and electroneutral cotransport proteins (NBCs), Na + -dependent exchangers (NDCBE or NCBE), and Na + -independent exchangers (AEs). The electrogenic NBCe1 (or NBC1) is a basolateral protein expressed in the proximal tubules ( 14 ). The electroneutral transporter NBCn1 is found basolaterally in the medullary thick ascending limbs (mTAL) of Henle's loop and in intercalated cells ( 17 ), most likely type A. Interestingly, NBC3, a variant of NBCn1, seems to be expressed apically in type A intercalated cells and basolaterally in type B intercalated cells ( 12 ). Recently, an electrogenic NBCe2 (or NBC4) has been detected in kidney by Northern blotting ( 11 ), and RT-PCR analysis suggests its expression in thick ascending limbs ( 19 ). BTR1 is the most recent member of the gene family, and Northern blot analysis revealed its expression in the human kidney ( 9 ). Finally, the renal NDCBE1 and NCBE expression patterns have not yet been established. From the mouse, rat, and human genomes it seems that the discovery of new members of the transporter superfamily SLC4A has been exhausted. Accordingly, the aim of the present study was to identify the Na + -dependent transporter in the IMCD cells among known and newly discovered proteins of this gene family. Furthermore, we aimed preliminarily to characterize the renal regulation of NBCn1 protein expression. RT-PCR, immunoblotting, immunohistochemistry, and recording of intracellular pH (pH i ) changes were applied to address these tasks.

MATERIALS AND METHODS

Experimental animals. Adult male Munich-Wistar rats (250-300 g) from the Møllegaard Breeding Centre (Denmark) had free access to water and pelleted food (Altromin, Lage, Germany) until use for routine RNA/protein isolation and immunohistochemistry, or until the inclusion in an experimental model for acid loading or alkaline loading ( 4 ). These experimental rats were adjusted to metabolic cages for 3 days and given a fixed amount of ground rat food (0.068 g/g body wt) mixed with water (0.168 g/g body wt). A total of 0.033 mmol/g body wt or NH 4 Cl or NaHCO 3 was included in the food for the experimental group ( n = 6, for each) for 7 days, whereas the control group ( n = 6) received the same diet without NH 4 Cl and NaHCO 3. Alternatively, rats were anesthesized with halothane, and osmotic minipumps (Alzet, 2ML1 ) were implanted subcutaneously in the back of the rat, releasing 12 mg furosemide/day. The control rats obtained minipumps containing the vehicle solution (1.7% ethanolamine, pH 7.5). The furosemide-treated and control rats had free access to food and tap water. To prevent a pronounced difference in body weight between the groups, the control rats obtained an amount of food corresponding to the average food intake of the furosemidetreated rats on the preceding day.

Dissection of inner medullary regions. The rats were anesthetized by halothane inhalation, and the kidneys were excised and rinsed in a 4° C saline solution after application of an abdominal longitudinal incision. Samples of outer medulla, cortex, and whole kidney were prepared for RT-PCR and immunoblotting. The inner medulla was further divided where indicated into three regions: one-third of the inner medulla closest to the outer medulla (IM1), an intermediate one-third region (IM2), and a one-third region consisting of the tip of the papilla (IM3). IMCD and thin limbs of Henle's loop were isolated after mild enzymatic digestion (see Measurements of cotransport ) of inner medullary slices at 4° C in Tris-buffered saline, pH 7.4, under x 10-25 magnification and rinsed before RNA isolation.

RT-PCR and sequence analysis. Total RNA was extracted using an RNeasy Mini- or Midi-Kit (Qiagen, Germantown, MD) and for microdissected tissue, mRNA was isolated using a Dynabeads mRNA Direct Micro Kit (Dynal, Oslo, Norway). After DNase treatment (DNaseI, Promega, Madison, WI), the RNA was reverse transcribed by 2 U/µl reverse transcriptase (Superscript II, Invitrogen, Taastrup, Denmark) in the presence of either poly-T primers or reverse primers for specific NBC gene products (transcript-specific RT). The reverse transcriptase was replaced by water in negative control samples. The resulting cDNA product was amplified by PCR. One to two microliters of cDNA were added to a 5-µl Taq polymerase mixture with deoxyribonucleotides (HotStarTaq Master Mix, Qiagen) and 0.5 µl of each of the two primers, in a final volume of 10 µl. Specific primers for various transporters were derived from published rat cDNA sequences or by homology between human and mouse sequences ( Table 1 ). The PCR products were analyzed by agarose gel electrophoresis. PCR products of predicted molecular sizes were excised from the gel and purified using a QIAquick Gel Extraction Kit (Qiagen) for nucleotide sequencing (Lark Technologies, Essex, UK). All RT-PCR reactions were performed on at least two separate RNA isolates.

Table 1. Primers used for detection of cotransporters in rat inner medulla

Antibodies. Two antibodies were used to detect NBCn1 by immunoblotting and immunohistochemistry. Both were raised against a common peptide of the COOH-terminal domain of rat NBCn1 and have previously been described and validated ( 17 ). An antibody against NBCe1 ( 14 ) was also applied for immunoblotting and -labeling.

Immunoblotting. The protein contents of 4,000- g centrifugation supernatants were determined using a bicinchoninic acid protein assay reagent kit (Pierce, Rockford, IL). Protein samples were adjusted to 1.5% (wt/vol) SDS, 40.0 mM DTT, 6% (vol/vol) glycerol, 10 mM Tris, pH 6.8, and added bromophenol blue. About 10 µg of proteins were separated on 9% polyacrylamide gels and electrotransferred onto nitrocellulose membranes, which were then blocked by incubation in 5% nonfat dry milk in a phosphate-buffered salt solution (PBS-T; containing 80 mM Na 2 HPO 4, 20 mM NaH 2 PO 4, 100 mM NaCl, and 0.1% vol/vol Tween 20, pH 7.5). The membranes were incubated with primary antibody overnight at 5° C in PBS-T. After being washed, the membranes were incubated with horseradish peroxidase-conjugated goat anti-rabbit IgG secondary antibody (Dako, Glostrup, Denmark) for 2 h in PBS-T. Excess antibody was then removed by extensive washing, and bound antibody was detected by an ECL chemiluminiscence kit (Amersham, Little Chalfont, UK). Semiquantification of the immunoreactive proteins was performed using standard equipment for densitometry. The band intensities were measured within the linear range and corrected for differences in sample loading using Coomassie-stained control gels.

Immunohistochemistry. The kidneys of halothane-anesthetized male Wistar rats were fixed by perfusion via the abdominal aorta, with 4% paraformaldehyde in 0.1 M cacodylate buffer, pH 7.4. The tissue was dehydrated, embedded in paraffin, and 2-µm sections were cut using a rotary microtome (Leica, Heidelberg, Germany). The sections were dewaxed, rehydrated, and endogenous peroxidase was blocked by 0.5% H 2 O 2 in absolute methanol. The sections were boiled in 10 mM Tris, pH 9, supplemented with 0.5 mM EGTA, and then incubated with 50 mM NH 4 Cl and blocked in PBS supplemented with 1% BSA, 0.05% saponin, and 0.2% gelatin. The sections were incubated overnight at 4° C with the primary antibodies diluted in PBS supplemented with 0.1% BSA and 0.3% Triton X-100.

For brightfield microscopy, the sections were incubated with horseradish peroxidase-conjugated goat anti-rabbit Ig (Dako P448) in PBS with BSA and Triton X-100. The staining was visualized by 0.05% 3,3'diaminobenzidine tetrahydrochloride dissolved in PBS with 0.1% H 2 O 2. Mayer's hematoxylin was used for counterstaining, and the sections were dehydrated in graded alcohol and xylene and mounted in hydrophobic Eukitt mounting medium (O. Kindler, Freiburg, Germany). Microscopy was performed on a Leica DMRE brightfield microscope equipped with PL Fluotar x 25 (0.75 numerical aperture) or PlApo x 63 (1.32 numerical aperture) objectives and a Leica DM300 digital camera.

For fluorescence microscopy, the sections were incubated with Alexa 488-conjugated goat anti-rabbit secondary antibodies (Molecular Probes, Eugene, OR) in PBS supplemented with BSA and Triton X-100. After being washed, sections were mounted on a coverslip in Glycergel Antifade Medium (Dako) and inspected on a Leica DMRS confocal microscope using an HCX PlApo x 64 (1.32 numerical aperture) objective. The immunofluorescence images were merged with differential interference contrast images to reveal the spatial relationship between the tissue structures and the fluorescence labeling.

Measurements of cotransport. Rat kidneys were removed, and the papillary two-thirds of the inner medulla was immediately isolated, segmented into <1-mm 3 pieces, and digested by 1 mg/ml collagenase A (Sigma, St. Louis, MO) and 1 mg/ml hyaluronidase (Roche Diagnostics, Mannheim, Germany) in oxygenated K + gluconate solution ( Table 2 ) for 45 min at 37° C during agitation at 120 rpm, as modified from Shaw and Marples ( 15 ). After being washed by centrifugation for 1 min at 800 g, the tubules were resuspended in K + gluconate solution and allowed to sediment and adhere to Cell-Tak (BD Biosciences, Bedford, MA)-coated coverslips. The tubules were loaded with the pH-sensitive dye BCECF by exposure to 16 µM membrane permeant form BCECF-AM (Molecular Probes) for 30 min.

Table 2. Solutions for tubule isolation and pH measurements

Coverslips were mounted with flow chambers on an Olympus CK40 microscope equipped with a UApo/240 ( x 40, 1.35 numerical aperture) oil-immersion objective, a monochromator, IC-200 CCD camera, and control unit from PTI (Lawrenceville, NJ). For each experiment, the 510-nm emission from nine individual tubular cells was recorded every 10 s using both 440- and 495-nm light excitation. The recordings were performed with a constant flow of 3 ml/min at 37° C. Cells were acidified by a prepulse of 25 mM NH + 4 and then superfused with a Na + -free solution containing 600 µM amiloride (all Na + -dependent "acid extruders" blocked). Thereafter, Na + was reintroduced in the continued presence of amiloride (only Na + /H + exchangers inhibited). Amiloride was then removed to allow acid extruders to function. Four variations of this protocol were applied: 1 ) all solutions contained; 2 ) all solutions were buffered by HEPES only; 3 ) all solutions contained buffer, which after acidification were supplemented with 200 µM DIDS (Sigma), an inhibitor of several Cl - and transporters; and 4 ) all solutions were Cl - free and contained. Cl - -free media were used throughout BCECF loading, the NH + 4 prepulse [(NH 4 ) 2 SO 4 ], and during the experimental recording to deplete the cells of Cl -. All experimental solutions are listed in Table 2.

The pH-dependent cellular fluorescence ratios (excitation 495/440) were calibrated into pH values by clamping pH i to values of extracellular pH from 8 to 6 in a high-K + medium with nigericin (Sigma). Calibration data from seven experiments were pooled to obtain a common calibration equation from a linear fit. The slopes of the pH recovery traces (dpH i /d t ) were determined, corresponding to a 1-min period after full change of bathing solution. The rate of recovery was converted into values of base influx through multiplication of dpH i /d t by the total buffering capacity, tot, which is the sum of the measured intrinsic buffering capacity, int, and the calculated contribution by the buffer system, CO2 ( 10 ). The int was obtained by stepwise decreasing intracellular NH + 4 concentrations and was calculated as the pH i change divided by the induced d[H + ] i, assuming the intracellular NH + 4 concentration d[NH + 4 ] i = d[H + ] i and a permeability for NH 3 NH + 4, and an equal distribution of NH 3 across the plasma membrane ( 10 ), where [H + ] i and are intracellular H + concentration and intracellular H +, respectively. A linear fit of the int data points was used for the subsequent calculation of base influx. Where was present, int was added to CO2, which was calculated as 2.3 x (9.6 x 10 -7 )/10 -pHi. The mean base influx of nine single cells was calculated for each experiment and used for the computation of the mean ± SE values from all three to five experiments for each of the four protocols.

Statistics. Mean values ± SE are reported. A two-tailed Mann-Whitney rank sum test was used for statistical analysis of data from two groups (immunoblot), and ANOVA with Tukey's posttest was used for four-group data comparison (net OH - influx). Values of P < 0.05 were considered statistically significant.

RESULTS

Expression of cotransporter mRNA in rat inner medulla. The presence of mRNA encoding cotransporters in the inner medulla was investigated by qualitative PCR. Figure 1 A shows that a 665-bp PCR product was detected using mRNA from the tip of the inner medulla (IM3) as a template and primers specific to rat NBCn1. Similar bands were observed with IM1 and IM2 mRNA (not shown) as well as with the control tissue, the inner stripe of outer medulla (ISOM). The presence of NBCn1 mRNA in the inner medulla was verified by sequence analysis of the PCR products. The 616-base sequence flanked by the primers showed 100% nucleotide identity with the published sequence for rat NBCn1 (AF069511 ).

Fig. 1. Detection of mRNA encoding Na + -dependent transporters in the rat renal inner medulla (IM). PCR was performed using reverse-transcribed mRNA from rat IM, whole kidney (WK), inner stripe of outer medulla (ISOM), and cortex (Ctx) as a template. In cases where the IM was further separated, IM1 denotes the one-third closest to the ISOM, IM3 is the one-third closest to the tip of the papilla, and IM2 was the intermediate one-third of IM. Negative controls included lack of a template (H 2 O) or the omission reverse transcriptase (-RT). A : specific primers were used to probe for a fragment of 665 bases from rat electroneutral cotransport (NBCn1) using renal ISOM mRNA as a positive control. Other primers were used to probe IM specifically for BTR1 ( B ), for NBCe1 and NBC4/NBCe2 ( D ), or for Na + -dependent exchangers (NCBE and NDCBE; E ). Expected size of RT-PCR products is indicated by arrowheads. C : microdissected inner medullary collecting ducts (CD) and thin limbs of Henle's loop (TnL) were analyzed for BTR1 mRNA using aquaporin-1 (AQP1) and -3 (AQP3) primers to verify the dissection.

The expression of mRNA encoding the putative transporter BTR1 was also examined in rat kidney, because a SLC4A11 transcript was recently found in human kidney ( 9 ). Figure 1 B shows the resulting RT-PCR analysis, where BTR1 mRNA was readily amplified from the IM1, IM2, and IM3. The renal cortex and whole kidney homogenates were also positive, although the bands appeared weaker (not shown). Sequencing revealed 100% nucleotide identity between the primer-flanked 302-base product and the rat genomic sequence. RT-PCR of the microdissected inner medullary thin limb fractions and collecting ducts revealed that BTR1 mRNA was found in the thin limbs and not in collecting ducts ( Fig. 1 C ). The quality of the tubular preparations was verified by the detection of AQP1 in medullary thin limbs and AQP3 in collecting ducts. RT-PCR analyses for additional Na + -dependent transporters, NBCe1, NBCe2/NBC4, NCBE, and NDBCE1, were all negative using inner medullary mRNA as a template, whereas products of the predicted sizes were obtained using cDNA from the whole kidney, kidney cortex, or ISOM ( Fig. 1, D and E ). The NBCe2/NBC4, NDCBE1, and NCBE PCR products were validated by nucleotide sequencing of control tissue PCR products, whereas the NBCe1 primers were validated previously ( 10 ).

Actin controls yielded RT-PCR products of the expected size with all RNA samples only when RT was performed. Contamination with chromosomal DNA would have increased the PCR product size, because the sequence flanked by the actin primers contains an intron at the DNA level. Thus the PCR products encoding fragments of the cotransporters were formed from reverse-transcribed mRNA from the respective kidney fractions.

Verification of NBCn1 expression in inner medulla by immunoblotting. Antibodies against cotransporters were applied for immunoblotting using proteins from IM3. Figure 2 A illustrates the reaction of the anti-NBCn1 antibody with an 180-kDa protein of the postnuclear protein fractions from both rat inner medulla and outer medulla. The antibody binding to inner medullary proteins was prevented by preabsorbing the anti-NBCn1 antibody with the immunizing peptide. An anti-NBCe1 antibody, which reacted with an 140-kDa protein from renal cortex, failed to detect the protein in distal inner medulla isolate, as illustrated in Fig. 2 B. The antibody binding to cortical proteins was prevented by peptide preabsorbtion of the anti-NBC1 antibody.

Fig. 2. Immunoblot analysis of NBC proteins in the distal IM from rats using peptide-derived antibodies. A : a protein of 180 kDa from the distal one-third of the IM (IM3) and ISOM was recognized by an anti-NBCn1 antibody after SDS-PAGE and electrotransferred onto nitrocellulose membranes. Antibody preabsorption by the immunizing peptide prevented antibody binding of IM proteins (IM3-p). B : an anti-NBCe1 antibody recognized a 140-kDa renal cortex (Ctx) protein but did not label IM3. Peptide preabsorption prevented labeling in renal cortex protein samples (Ctx-p).

Immunolocalization of NBCn1 in the papillar two-thirds of IMCD. An antibody against the COOH terminus of the NBCn1 peptide was used for immunohistochemical analysis. In addition to the known labeling of mTAL and intercalated cells, the antibody stained the basolateral membrane domains of IMCD cells of the middle and terminal part of the tubule, IMCD2 and IMCD3, respectively ( Fig. 3, A and B ). Preabsorption of the antibody by the immunizing peptide completely prevented NBCn1 labeling of the IMCD cells, as shown in Fig. 3 C. In contrast, an irrelevant peptide [derived from aquaporin-1 (AQP1)] had no effect on NBCn1 labeling in the IMCD (not shown). The basolateral localization of NBCn1 to IMCD2 cells was confirmed by laser-scanning confocal fluorescence microscopy ( Fig. 3 E ). The figure illustrates two collecting ducts merging into a common collecting duct in the inner medulla and reveals selective NBCn1 labeling of the basolateral plasma membrane domains with little or no labeling of intracellular or apical structures. The principal cells of the initial part of IMCD (IMCD1) did not bind the anti-NBCn1 antibody, whereas the intercalated cells and the surface epithelium lining the renal papilla were stained by the antibody ( Fig. 3 D ). This immunoreactivity was confined to the basolateral domain of these cells, as illustrated in Fig. 3 D (arrowheads). Figure 3 F shows the previously reported basolateral domain labeling of mTAL and basolateral staining of intercalated cells of the OMCD. The NBCe1 antibody was also applied for immunohistochemistry to verify the absence of this electrogenic NBC found by PCR and immunoblotting. Figure 4 A shows that the distal part of the inner medulla did not label with the anti-NBCe1 antibody. The basolateral staining of proximal tubules was used as a positive control ( Fig. 4 B ). Thus novel NBCn1 immunolabeling was detected in cells from the IMCD.

Fig. 3. Immunohistochemical and confocal immunflourescence microscopic localization of NBCn1 in renal IM. Semithin sections from perfusion-fixed and paraffin-embedded rat kidneys were labeled with a peptide-derived antibody against NBCn1. A : anti-NBCn1 antibody stained the basolateral aspect of IM collecting duct (IMCD) cells in the intermediate one-third of IM (IM2). B : labeling was similarly found in the more terminal IMCD segments in IM3. C : staining was completely prevented by antibody preabsorption with the immunizing peptide. D : surface epithelial cells lining the renal papilla (arrowheads) also showed immunoreactivity with the anti-NBCn1 antibody. Labeling was confined to the basolateral domain of these cells. E : confocal immunofluorescence microscopy combined with differential interference contrast microscopy confirms the selective basolateral localization of NBCn1 in the distal IM. F : intercalated cells of the outer medullary collecting duct (OMCD) and the medullary thick ascending limb (mTAL) of Henle's loop labeled with the anti-NBCn1 antibody as previously reported. Bars = 10 µm.

Fig. 4. Immunohistochemical analysis of the distal IM using antibodies against NBCe1. A : structures in the distal IM were not stained by the anti-NBC1 antibody, whereas proximal convoluted tubules (Prox.tub.) of the same section ( B ) showed strong labeling. Bars = 10 µm.

Demonstration of DIDS-insensitive cotransport in IMCD cells. The recovery of pH i was studied in isolated, acidified IMCD segments to establish whether the presence of NBCn1 protein was accompanied by measurable cotransport. The isolated tubules displayed good viability after isolation as they adhered to the coverslips, loaded BCECF, and cleaved and retained the fluorescence probe for at least 1 h. Figure 5 A shows the BCECF fluorescence image of a single IMCD tubule during an NH 4 Cl prepulse used to acidify the cells. The net base influx was calculated from the slope of the calibrated fluorescence excitation ratio trace ( Fig. 5 B ) and the total buffering capacity ( Fig. 5 C ). This base influx was highly dependent on extracellular Na + both in the presence and in the absence of the buffer system, as illustrated in Fig. 5, B and D ( P < 0.05 in both HEPES and buffer, n = 3 and 5, respectively).

Fig. 5. Demonstration of DIDS-insensitive cotransport in isolated IMCD. A : image of the 510-nm light emission during 495-nm excitation from a BCECF-loaded tubule preparation. B : cells were acidified by an NH 4 Cl prepulse and the pH recovery was measured both in the absence and in the presence of Na + as indicated. Amiloride (600 µM) was used to inhibit Na + /H + exchangers. DIDS (200 µM) was added the buffer in separate experiments. The slope of the pH trace was used to determine the base influx rate. C : a mean calibration curve was obtained with the high-K + -nigericin method ( left ), and measurements of intrinsic buffering capacity were performed for the computation of base influx ( right ). D : the mean base-influx ± SE during pH recovery from 3 (HEPES), 5 (HCO 3, HCO 3 -DIDS), and 4 (HCO 3, Cl-free) experiments. Open bars, 144 mM Na +; closed bars, Na + -free. Values were obtained in the continued presence of amiloride. The Na + -dependent pH recovery is the difference between the open and closed bars, whereas the (and Na + )-dependent pH recovery (NBC activity) is the difference between HEPES and columns. * Significantly different, P < 0.05.

The Na + -dependent base influx was 2.5 times greater in the presence of the buffer than in HEPES buffer ( P < 0.05, n = 5 and 3, respectively). The Na + -dependent component in HEPES buffer likely reflects amiloride-resistant Na + /H + exchange. Importantly, the additional Na + -dependent base influx observed using buffer can be ascribed to cotransport. Addition of 200 µM DIDS or intracellular Cl - depletion had no effect on this Na + - and -dependent base influx (not significant, n = 5 and 4, respectively). The net base influx was calculated in a narrow range of pH i (pH 5.93-6.10), which allows direct comparison of the mean flux values. 1 When amiloride was removed, the base influx increased by 8-10 times ( Fig. 5 B ). This large net flux reflects amiloride-sensitive Na + /H + exchange.

NBCn1 expression after 7 days of NH 4 Cl, NaHCO 3, or furosemide administration. One-week dietary NH 4 Cl administration is a frequently used model for acid loading. It is known to increase the expression of NBCn1 protein ( 6 ) and to augment the cellular DIDS-insensitive Na + -dependent uptake in mTAL ( 8 ). The signaling pathways leading to this response are unknown. Furthermore, the expression of other transporters, such as pendrin, has been shown to change after NaHCO 3 loading ( 4 ). Therefore, we investigated whether NBCn1 expression in the IMCD2-3 cells changes in response to whole body NH 4 Cl or NaHCO 3 loading. The analysis of urine and blood samples was reported previously ( 4 ). In brief, urinary [H + ] ( x 10 -8 M) was 175 ± 122 (pH 5.76) in NH 4 Cl-loaded rats and 1.14 ± 1.01 (pH 7.94) in controls, and 0.017 ± 0.003 (pH 8.77) in NaHCO 3 -loaded rats and 3.99 ± 1.40 (pH 7.40) in controls. However, blood P CO 2,, and pH levels were normal in all groups, indicating full renal compensation for the experimental acid or base loading. Figure 6 A illustrates that NBCn1 protein expression levels in the inner medulla were not altered by NH 4 Cl administration (not significant, n = 6), whereas NBCn1 expression was increased 1.8-fold in ISOM of the same NH 4 Cl-treated animals ( P < 0.05, n = 6, Fig. 6 B ). Thus NBCn1 protein level in the inner medulla was not regulated in parallel with the level in the ISOM. However, NBCn1 abundance in both the ISOM and the IM was not affected by an equimolar administration of NaHCO 3 ( Fig. 6, C and D, not significant, n = 6 for each).

Fig. 6. Effect of 7-day NH 4 Cl, NaHCO 3, and furosemide treatment on NBCn1 protein expression in IM and ISOM. A and B : NBCn1 protein expression levels in IM and ISOM, respectively, from NH 4 Cl-treated rats (0.033 mmol/g) as a percentage of the expression levels in control animals. C and D : NBCn1 protein expression levels in IM and ISOM, respectively, from NaHCO 3 -treated rats (0.033 mmol/g) as a percentage of the expression levels in control animals. E and F : NBCn1 protein expression levels in IM and ISOM, respectively, from rats treated with furosemide (12 mg/day) as a percentage of the expression levels in control animals. P values are indicated where statistical significant differences were obtained.

Alkalosis was induced in a third set of animals by furosemide administration, which most likely also changed the renal interstitial osmolarity. Blood [H + ] ( x 10 -8 M) was 3.69 ± 0.07 (pH 7.43) in furosemide-treated rats and 4.48 ± 0.09 (pH 7.35) in controls ( P < 0.05, n = 5 in both groups). Urinary [H + ] ( x 10 -8 M) was 7.68 ± 0.87 (pH 7.13) in furosemide-treated and 1.21 ± 0.19 (pH 7.94, P < 0.05) in control rats. Body weight, blood P CO 2, and plasma osmolarity were unchanged by furosemide treatment (not significant). Furosemide treatment decreased NBCn1 protein abundance in both the ISOM and IM, as shown in Fig. 6, E and F, respectively. The most pronounced effect was observed in ISOM, where NBCn1 abundance was reduced about threefold by furosemide treatment. Hence, the diverse experimental conditions produced either similar or differential regulation of NBCn1 abundance at the two sites of expression. These changes are also observed at the tubular level by the decrease in NBCn1 immunolabeling after furosemide treatment in IMCD3 ( Fig. 7, A vs. B ) and in mTAL ( Fig. 7, C vs. D ).

Fig. 7. Renal NBCn1 immunolabeling in furosemide-treated rats. NBCn1 staining was more marked in late IMCD from representative sections of kidneys from control rats ( A ) compared with furosemide-treated rats ( B ). NBCn1 staining was also more marked in late mTAL from control rats ( C ) compared with furosemide-treated rats ( D ). mTAL images are from the ISOM. Bars = 25 µm.

DISCUSSION

Na + -dependent uptake into primary cultures of IMCD cells is reportedly confined to the basolateral plasma membrane ( 5 ) and is found to be insensitive to deprivation of intracellular Cl -. Thus we hypothesized that a cotransporter was likely to be responsible for these observations. Reliable antibodies have only been raised against a few Na + -dependent transporters. Therefore, RT-PCR was used as a first step in identifying candidate proteins within the SLC4A gene family, from which all known Na + -dependent transporters originate. This approach is likely to be both comprehensive and precise, because the discovery of new members of this gene superfamily is unlikely.

Among the five Na + -dependent transporters, only NBCn1 mRNA was detected in the distal inner medulla by RT-PCR and sequence analysis. In addition, mRNA encoding the uncharacterized SLC4A11, called BTR1, was also detected in the inner medulla. Interestingly, BTR1 expression seemed to be confined to the thin limbs of the inner medulla, whereas isolated medullary collecting ducts were negative for BTR1 mRNA. Even though the applied technique is very sensitive, the medullar tubular BTR1 expression pattern awaits confirmation by immunolocalization and the relevance of the observation will follow the functional characterization of BTR1.

There is always a risk of overlooking low-level mRNA expression by RT-PCR. Transcript-specific RT can to some extent compensate for a low expression level, because the transcripts of interest are not competing with high-abundance mRNA for RT. This technique did not, however, reveal expression of additional transporters in rat inner medulla. Therefore, it seems that NBCn1 mRNA is most likely the only transcript encoding an Na + -dependent transporter in the IMCD cells.

Immunoblotting and preabsorption controls demonstrated that NBCn1 mRNA was translated into protein in the inner medulla. NBCn1 has previously been localized to the mTAL and to the basolateral membrane domain of intercalated cells in the outer medulla and initial inner medulla by immunohistochemical analysis of cryostat sections ( 17 ). The finding that NBCn1 is localized to the basolateral domain of IMCD2 and IMCD3 cells required paraffin embedding and target retrieval procedures of the kidney sections. The specificity of this additional labeling was ensured by a preabsorption test using the immunizing peptide. Thus RT-PCR- and the antibody-based approaches propose NBCn1 as a most likely candidate for mediating the previously described basolateral Na + -dependent transport in IMCD cells ( 5 ).

Immunohistochemistry and blotting confirmed the absence of this transporter in the distal inner medulla. Although these data robustly supported the molecular absence of NBCe1, the absence of other cotransporters cannot be confirmed in cases where antibodies have not yet been developed. A model for the localization of acid-base cotransporters in IMCD2-3 cells is presented in Fig. 8. Interestingly, the cubic surface epithelial cells lining the papilla are also labeled, suggesting a basolateral uptake of in these cells similar to that of distal IMCD cells. Although interesting per se, the presence of NBCn1 in surface epithelial cells was not investigated further.

Fig. 8. Model of acid-base cotransporters in rat IMCD (IMCD2-3). The basolateral plasma membrane contains NBCn1 and Na + /H + exchanger (NHE1), which, driven by the inward Na + gradient, tend to increase pH i by taking up or extruding H +, respectively. The anion exchanger AE2, however, extrudes in exchange for Cl - and tends to lower pH i. Controversy still exists regarding the molecular identity of the apical proton pump, because the expression of both the H + -ATPase and the K + -H + -ATPase has been proposed.

The molecular detection of medullary NBCn1 was supplemented with a functional assay for cotransport in acutely isolated papillary IMCD cells. The detection of a significant Na + - and -dependent pH regulatory component in acidified IMCD cells is compatible with the presence of a cotransporter. This transport was not dependent on normal intracellular levels of Cl -, which rules out the participation of a Na + -dependent exchanger. This is fully in agreement with the reported Cl - independence of basolateral cotransport in cultured IMCD cells ( 5 ). The strongest support for the functional presence of NBCn1 was, however, that the pH recovery was insensitive to DIDS, which is a hallmark of NBCn1-mediated processes. The finding also rejects the involvement of DIDS-sensitive transporters in the observed pH recovery, i.e., all other NBCs and NDCBE/NCBE. Thus the present functional detection of NBC activity in the isolated IMCD is fully consistent with the mRNA and protein-chemical localization of NBCn1 to this renal tubular segment.

NBCn1 will, under physiologically relevant conditions, transport inward. The transport is driven by the Na + gradient and is believed to participate in cellular "base loading" or "acid extrusion" like the Na + /H + exchanger NHE1 ( 1 ). Hypothetically, NBCn1 would also be capable of sustaining an apical secretion, by loading the ion from the basolateral domain, and thereby maintaining a suitable intracellular level of. This possibility is, however, not plausible, as the distal inner medulla is not known to be involved in alkaline secretion but is capable of extruding protons into the lumen ( 7 ). Only the B intercalated cells, which are all located in the cortex, are known to secrete and the presence of NBCn1/NBC3 in these cells remains to be verified by mRNA and functional assays. Hence, regulation of pH i and cell volume remains to be the most likely function for NBCn1 in the IMCD.

The signaling pathways for long-term regulation of NBCn1 expression have not been studied, although dietary NH + 4 administration has been shown to be associated with an increase in NBCn1 protein in mTAL and intercalated cells ( 6, 8 ). In contrast to mTAL, NH 4 Cl treatment did not increase NBCn1 expression in the IMCD in the present study. The differential regulation of NBCn1 in mTAL and IMCD could rely on 1 ) differences in local pH, 2 ) varying tonicity, or 3 ) the variation in segmental distribution of receptors to circulating or local factors. These possibilities are discussed in the following section.

The renal excretion of H + /NH + 4 seemed sufficient to maintain normal blood P CO, concentratin, and pH in the applied NH 4 Cl loading model. These values were most likely equally normal in the interstitial environment near the mTAL and IMCD cells. However, luminal pH drops along the IMCD because water and salt are reabsorbed in this segment and protons are possibly secreted into the IMCD lumen ( 7 ). Nonetheless, NBCn1 expression was unchanged in rat inner medulla when urinary pH was reduced by NH 4 Cl administration, excluding luminal pH as a regulator of NBCn1 in IMCD cells. pH i could, however, be differently affected at the two sites. NH + 4 is imported from the tubular lumen by NKCC2 into the mTAL. Once inside the cells, NH + 4 is converted to H + and NH 3, of which the latter diffuses to the interstitial compartment. It can be speculated that a subsequent decrease in pH i, selectively in mTAL, triggers the increased expression of NBCn1 in this segment during NH 4 Cl loading. Thus NBCn1 is presumably involved in buffering of the protons formed intracellularly in the mTAL in this animal model as suggested by Kwon, Odgaard, and co-workers ( 6, 8 ).

From this perspective, it was not surprising that fully compensated NaHCO 3 loading failed to change NBCn1 expression in ISOM and IM. P CO 2 and pH were likely normal in the renal interstitium and luminal pH elevated (high urinary pH) but again presumably without effect on pH i and NBCn1 expression. This is also consistent with the notion that base extrusion largely occurs in the B intercalated cells of the cortical collecting ducts and not in mTAL and IMCD. In furosemide-induced alkalosis, however, the NBCn1 abundance in both ISOM and IM was significantly lowered as the blood pH was elevated (with no change in P CO 2 ). This would be in line with the notion that intracellular may be regulating NBCn1 abundance, whereas luminal pH seems irrelevant for NBCn1 regulation.

As mentioned above, interstitial tonicity could be an alternative regulator of NBCn1 expression. This would correspond well with the gradually enhanced immunolabeling for NBCn1 toward the terminal IMCD in normal rats, as the interstitial tonicity increases. The decrease in NBCn1 abundance after furosemide treatment is also in agreement with a regulatory role for interstitial tonicity. However, tonicity differences alone cannot explain the much stronger NBCn1 immunoreactivity of mTAL compared with IMCD. Further studies are warranted to uncover the regulatory significance of pH i, tonicity, and circulating factors.

In conclusion, we have demonstrated the presence of NBCn1 in IMCD cells using a combined approach with RT-PCR, immunoblotting, and immunohistochemistry. The protein is localized to the basolateral domain of the IMCD cells with increasing abundance toward the papillary IMCD. Moreover, the functional study provides further evidence that NBCn1 is the only cotransporter of rat IMCD cells, because the Na + - and -dependent base efflux from acid load was DIDS insensitive and independent of intracellular Cl -. The mRNA encoding BTR1 was also detected in the rat inner medulla, but expression seemed to be restricted to thin limbs of Henle's loop and was not detected in IMCD. Furthermore, differential regulation of NBCn1 in IMCD and mTAL was demonstrated, as dietary NH 4 Cl administration did not affect NBCn1 expression in IMCD cells in contrast to the increased NBCn1 abundance in mTAL. NaHCO 3 administration did not affect NBCn1 expression in either of the two sites, whereas furosemide-induced alkalosis decreased NBCn1 abundance in both IM and ISOM. The transporter possibly contributes to the cellular defense against acidification or volume changes in IMCD cells, and in mTAL perhaps also to NH + 4 reabsorption. Finally, the isolated IMCD preparations seem to be very a suitable model to study the acute regulation of NBCn1 expression and function.

ACKNOWLEDGMENTS

We are grateful to Christiaan Fulton and Jens Leipziger for inspiration and fruitful discussions. We also thank Mette F. Vistisen, Lotte V. Holbech, Merete Pedersen, Inger-Merete Paulsen, Ida Maria Jalk, and Susie Mogensen for skilled technical assistance.

GRANTS

The Water and Salt Research Center at the University of Aarhus is established and supported by the Danish National (Danmarks Grundforskningsfond). Support for this study was additionally provided by The European Commission (contract QLK3-CT-2000-0078), The University of Aarhus Research Foundation, and the Human Frontier Science Program.

【参考文献】
  Boron WF. Sodium-coupled bicarbonate transporters. JOP 2: 176-181, 2001.

Brown D, Hirsch S, and Bluck S. An H + -ATPase in opposite plasma membrane domains in kidney epithelial cell subpopulations. Nature 331: 622-624, 1988.

Clapp WL, Madsen KM, Verlander JW, and Tisher CC. Intercalated cells of the rat inner medullary collecting duct. Kidney Int 31: 1080-1087, 1987.

Frische S, Kwon TH, Frøkiær J, Madsen KM, and Nielsen S. Regulated expression of pendrin in rat kidney in response to chronic NH 4 Cl or NaHCO 3 loading. Am J Physiol Renal Physiol 284: F584-F593, 2003.

Hart D and Nord EP. Polarized distribution of Na + /H + antiport and cotransport in primary cultures of renal inner medullary collecting duct cells. J Biol Chem 266: 2374-2382, 1991.

Kwon TH, Fulton C, Wang W, Kurtz I, Frøkiær J, Aalkjær C, and Nielsen S. Chronic metabolic acidosis upregulates rat kidney cotransporters NBCn1 and NBC3 but not NBC1. Am J Physiol Renal Physiol 282: F341-F351, 2002.

Morel F and Doucet A. Functional segmentation of the nephron. In: The Kidney: Physiology and Pathophysiology, edited by Seldin DW and Giebisch G. New York: Raven, 1992, p. 1049-1086.

Odgaard E, Jakobsen JK, Frische S, Praetorius J, Nielsen S, Aalkjær C, and Leipziger J. Basolateral Na + -dependent transporter NBCn1-mediated influx in rat medullary thick ascending limb. J Physiol 555: 205-218, 2004.

Parker MD, Ourmozdi EP, and Tanner MJ. Human BTR1, a new bicarbonate transporter superfamily member and human AE4 from kidney. Biochem Biophys Res Commun 282: 1103-1109, 2001.

Praetorius J, Hager H, Nielsen S, Aalkjær C, Friis UG, Ainsworth MA, and Johansen T. Molecular and functional evidence for electrogenic and electroneutral cotransporters in murine duodenum. Am J Physiol Gastrointest Liver Physiol 280: G332-G343, 2001.

Pushkin A, Abuladze N, Newman D, Lee I, Xu G, and Kurtz I. Cloning, characterization and chromosomal assignment of NBC4, a new member of the sodium bicarbonate cotransporter family. Biochim Biophys Acta 1493: 215-218, 2000.

Pushkin A, Yip KP, Clark I, Abuladze N, Kwon TH, Tsuruoka S, Schwartz GJ, Nielsen S, and Kurtz I. NBC3 expression in rabbit collecting duct: colocalization with vacuolar H + -ATPase. Am J Physiol Renal Physiol 277: F974-F981, 1999.

Royaux IE, Wall SM, Karniski LP, Everett LA, Suzuki K, Knepper MA, and Green ED. Pendrin, encoded by the Pendred syndrome gene, resides in the apical region of renal intercalated cells and mediates bicarbonate secretion. Proc Natl Acad Sci USA 98: 4221-4226, 2001.

Schmitt BM, Biemesderfer D, Romero MF, Boulpaep EL, and Boron WF. Immunolocalization of the electrogenic cotransporter in mammalian and amphibian kidney. Am J Physiol Renal Physiol 276: F27-F38, 1999.

Shaw S and Marples D. A rat kidney tubule suspension for the study of vasopressin-induced shuttling of AQP2 water channels. Am J Physiol Renal Physiol 283: F1160-F1166, 2002.

Tsuganezawa H, Kobayashi K, Iyori M, Araki T, Koizumi A, Watanabe S, Kaneko A, Fukao T, Monkawa T, Yoshida T, Kim DK, Kanai Y, Endou H, Hayashi M, and Saruta T. A new member of the transporter superfamily is an apical anion exchanger of -intercalated cells in the kidney. J Biol Chem 276: 8180-8189, 2001.

Vorum H, Kwon TH, Fulton C, Simonsen B, Choi I, Boron W, Maunsbach AB, Nielsen S, and Aalkjær C. Immunolocalization of electroneutral cotransporter in rat kidney. Am J Physiol Renal Physiol 279: F901-F909, 2000.

Wingo CS, Madsen KM, Smolka A, and Tisher CC. H-K-ATPase immunoreactivity in cortical and outer medullary collecting duct. Kidney Int 38: 985-990, 1990.

Xu J, Wang Z, Barone S, Petrovic M, Amlal H, Conforti L, Petrovic S, and Soleimani M. The expression of cotransporter NBC4 in rat kidney and characterization of a novel NBC4 variant. Am J Physiol Renal Physiol 284: F41-F50, 2003.

作者单位：The Water and Salt Research Center, University of Aarhus, DK-8000 Aarhus C, Denmark