Abstract

The purpose of the present studies was to examine the renal distribution and functional properties of Na+-HCO 3 cotransporter type 4 (NBC4), the latest NBC isoform to be identified. Zonal distribution studies in rat kidney by Northern blot hybridization and RT-PCR demonstrated that NBC4 is highly abundant in the outer medulla and cortex but is low in the inner medulla. Nephron segment distribution studies indicated that NBC4 is predominantly expressed in the medullary and cortical thick ascending limb of the loop of Henle. Using specific primers on the basis of the published sequence (GenBank accession no. AF-207661), a full-length NBC4 variant was cloned from human liver and examined. The sequence of this variant (called NBC4e) is shorter by 86 amino acids vs. the published sequence. Xenopus laevisoocytes injected with the full-length NBC4e cRNA were compared with NBC1-expressing oocytes. Although exposure of NBC1-expressing oocytes to CO2/HCO 3 resulted in immediate hyperpolarization, the NBC4-expressing oocytes did not show any alteration in membrane potential. NBC activity in oocytes, assayed as the Na+-dependent, HCO 3 -mediated intracellular pH recovery from acidosis, indicated that NBC4 is a DIDS-inhibitable NBC. We propose that NBC4 is expressed in the thick ascending limb of the loop of Henle and mediates cellular HCO 3 uptake in this segment.

  • sodium-bicarbonate cotransporter
  • NBC4
  • NBC1
  • acid-base transporters
  • thick limb of the loop of Henle

first described in the kidney, the Na+-HCO 3 cotransporter (NBC) is responsible for the bulk of HCO 3 reabsorption across the basolateral membrane of kidney proximal tubule (3, 6, 9, 32, 33). Subsequent studies have shown the presence of NBC in numerous tissues, including brain, liver, cornea, heart, and lung, suggesting that the process of Na+-dependent HCO 3 flux has an important role in overall HCO 3 transport in both epithelial and nonepithelial cells (7, 15-17, 21, 23, 24, 27).

On the basis of functional data, the presence of more than one NBC isoform was suggested (3, 6, 8, 9, 12, 32-34, 39). Recent molecular cloning experiments have confirmed this prediction by identifying four NBC isoforms (NBC1–4) and two related proteins (NCBE and AE4) (1, 2, 4, 10, 11, 13, 20, 25, 26, 28, 29, 33, 34,36, 39). For each NBC isoform, a number of splice variants have been identified. NBC1 has four distinct splice variants, with kidney (kNBC1) and pancreatic (pNBC1) variants being the two most abundant ones (1, 2, 10, 11, 28, 29, 33). With respect to NBC2, three splice variants have been identified. These are retinal NBC2, skeletal muscle NBC3 (mNBC3), and vascular NBCn1, which is a rat ortholog of NBC2 (13, 20, 25). NBC3 has three identified splice variants, with kNBC3 (also referred to as NDCBE) being the most common (4, 19, 40). NBC4 has three variants that have been reported so far (26, 31). These are referred to as NBC4a-c (26, 31). NBC1–3 mediate Na+-HCO 3 cotransport, and NCBE works as a Na+-dependent Cl/HCO 3 exchanger (reviewed in Refs. 32 and 33). Despite structural homology to NBCs, AE4 does not transport Na+ and works only in Cl/HCO 3 exchange mode (36). NBC4 is the latest member of the NBC family to be cloned (26, 31); however, little is known about its functional properties and renal distribution. A dendrogram of the NBC/AE family demonstrates that NBC4 cDNA has the highest homology to NBC1, whereas NBC2 is closely related to NBC3 (33).

The purpose of the present studies was to study the renal distribution and functional properties of NBC4. Specifically, we were interested in comparing the electrogenicity of this transporter vs. NBC1, as both show the highest degree of homology to each other.

EXPERIMENTAL PROCEDURES

RNA Isolation

RNA was extracted from various rat tissues by using TriReagent (Molecular Research, Cincinnati, OH) according to the manufacturer's instructions. The extracted RNA was quantitated spectrophotometrically and stored at −80°C.

Northern Blot Hybridization

RNA samples (30 μg/lane) were fractionated on 1.2% agarose-formaldehyde gels and transferred to nylon membranes. RNA was covalently bound to the nylon membranes by ultraviolet cross-linking. Hybridization was performed according to established methods and as used previously (4), with [32P]dCTP (NEN, Boston, MA)-labeled cDNA probes. Radiolabeled blots were exposed to PhosphorImager storage screens for 24–72 h and imaged by using ImageQuant software (Molecular Dynamics). For NBC4, a PCR fragment encoded by the rat NBC4-specific primers (ATGGTTGACCGATCCTTG and GCTGGCTCTTAATAATGATGGC for the sense and antisense directions, respectively) was purified from the rat kidney cortex RNA and used as a specific probe for Northern blot hybridization. For NBC1, an ∼700-bp PCR fragment corresponding to nucleotides 1910–2650 of the rat kidney NBC1 cDNA was used as a specific probe.

Nephron Segment Distribution of NBC4 in Rat

To examine the tissue and nephron segment distribution of NBC4 in rat, a rat NBC4 cDNA fragment was cloned by RT-PCR. The following primers were designed on the basis of the human NBC4 cDNA sequence: GCC AGC TAT GCA TGA AAT TG (sense) and ATG GGT CCT GTG CTG CTG AG (antisense). Using a gradient-based RT-PCR cycler, a cDNA fragment of 1.1 kb, which is the expected size for the fragment, was amplified from rat kidney. The cDNA sequence of this fragment is 76% homologous to human NBC4. Rat-specific primers (ATGGTTGACCGATCCTTG and GCTGGCTCTTAATAATGATGGC for the sense and antisense directions, respectively) were designed on the basis of the sequence of the amplified fragment and used for NBC4 RT-PCR. The rat PCR fragment corresponds to nucleotides 1434–1751 of NBC4a (AF-243499), 1781–2098 of NBC4b (NM-033323), 1781–2098 of NBC4c (AF-293337), and 1615–1932 of NBC4e (AF-452248). This fragment is therefore common to all NBC4 variants that have been cloned to date.

Single nephron segments from the medullary thick ascending limb (mTAL), cortical thick ascending limb (cTAL), proximal straight tubule, or cortical collecting duct (CCD) were dissected from freshly killed rat kidney at 4–6°C. The dissection media comprised the following (in mM): 140 NaCl, 2.5 K2HPO4, 2 CaCl2, 1.2 MgSO4, 5.5 d-glucose, 1 Na-citrate, 4 Na-lactate, and 6 l-alanine, pH 7.4, bubbled with 100% O2. Tubule lengths were ∼0.5–0.7 mm. The mTAL, cTAL, proximal straight tubule, or CCD segments were pooled in a small volume (5–10 μl) of ice-cold PBS, three or four segments per pool. The tubules were centrifuged at 12,000 g for 1 min at room temperature, and then PBS was removed and replaced with 10 μl of a tubule lysis solution consisting of 0.9% Triton X-100, 5 mmol/l DTT, and 1 U/μl rRNasin (Promega, Madison, WI). After 5 min on ice, the tubules were gently agitated by tapping the tube, and (in μl) 1 (0.5 μg) oligo(dT) primer, 1 H2O, 4 5× reverse transcription buffer, 2 DTT (0.1 mol/l), and 1 dNTPs (10 mmol/l each) were added. The reaction was equilibrated to 42°C for 2 min, and 1 μl SuperScript II RT (Life Technologies, Grand Island, NY) was added, mixed, and incubated for 1 h at 42°C. After reverse transcription, 30 μl of TE (10 mmol/l Tris · Cl and 1 mmol/l EDTA, pH 8.0) were added, and the combined mixture was heated to 95°C for 5 min and then placed on ice. Amplification of the NBC4 cDNA by PCR was performed with rat NBC4-specific primers (ATGGTTGACCGATCCTTG and GCTGGCTCTTAATAATGATGGC for the sense and antisense directions, respectively). Cycling parameters were 95°C for 32 s, 50°C for 37 s, and 68°C for 4 min.

Cloning of Human NBC4

Total RNA (2 μg) from human liver was reverse transcribed with an oligo(dT) RT and then subjected to PCR by using human NBC4-specific primers. The primers were designed on the basis of the published sequence (AF-207661) (26). One NH2-terminal and two distinct COOH-terminal fragments from human liver were identified by RT-PCR. To obtain the COOH-terminal fragment, the following primers were used: 5′-TCA GTC TCC ACC ACA AAT CGC AGT C (sense) and 5′-TTGTCA ATC CAG GCC AGG TCG TGC (antisense). This fragment encodes nucleotides 2869–5397 of the published cDNA. To obtain the NH2-terminal end, the following primers were used: 5′-GGA CAT GGT TCA GAA ACA GCT AAA GCA GAC (sense) and 5′-CGG TGA AGC GGG TGA TAT ATT TGA TG (antisense). This fragment corresponds to nucleotides 1989–4078 of the published cDNA. To obtain the full-length cDNAs, the following primers were used: GGA CAT GGT TCA GAA ACA GCT AAA GCA GAC (1989, sense) and TTG TCA ATC CAGGCCAGG TCG TGC (5397, antisense) for RT-PCR on RNA isolated from rat liver. Two full-length NBC4 variants were isolated that are referred to as NBC4e and NBC4f and contain 3345 and 3067 nucleotides, respectively. The NBC4f cDNA is shorter than NBC4e by only 278 bp; however, it shows a truncation of 377 amino acids on its COOH-terminal end due to a stop codon at position 2464. The hydropathy plot analysis of this truncated variant shows only six transmembrane-spanning domains (data not shown), which most likely makes it nonfunctional. NBC4e, on the other hand, has 10 transmembrane-spanning domains (as judged by the hydropathy plot analysis) and encodes a 1,051-amino acid protein (results). For the purpose of these studies, we focused on NBC4e characterization. Cycling parameters were as 30 cycles of 94°C for 30 s, 57°C for 30 s, and 72°C for 3 min. All fragments were subcloned into pGEM-T vector and sequenced.

Expression Studies in Xenopus laevis Oocytes

Synthesis of NBC1 and NBC4 cRNAs.

The capped NBC1 (kidney variant) and NBC4e cRNAs were generated by using the mCAP RNA Capping Kit (Stratagene, La Jolla, CA) according to the manufacturer's instruction. Briefly, the plasmids containing full-length human NBC1 or NBC4 were linearized by restriction enzyme digestion. The products were then in vitro transcribed to cRNAs. The purified linear template, at 1 μg, was mixed with transcription buffer, rNTP mix, mCAP analog, and T7 polymerase and incubated at 37°C for 30 min. The template was removed by adding RNAse-free DNAse I to the reaction, and the product was purified by phenol-chloroform extraction and ethanol precipitation. The size of the in vitro transcription product, its quantity, and its quality were evaluated by denaturing agarose gel electrophoresis. The cRNAs were stored in RNase-free water at −80°C.

Expression of NBC transporters in X. laevis oocytes.

X. laevis oocytes were injected with NBC1 or NBC4e cRNA. Stage IV and V oocytes were isolated as previously described (14) and used for expression studies according to established methods (35). Fifty nanoliters cRNA (0.2–1.3 μg/μl) were injected with a Drummond 510 microdispenser via a sterile glass pipette with a tip of 20–30 μm. After injection, the oocytes were maintained in a solution of the following composition (in mM): 96 NaCl, 2.0 KCl, 1.0 MgCl2, 1.8 CaCl2, 5 HEPES, 2.5 Na pyruvate, and 0.5 theophylline, as well as 100 U/ml penicillin and 100 μg/ml streptomycin, pH 7.5. Injected oocytes were stored in an incubator at 19°C and were used for electrophysiological experiments after 2–4 days.

Electrophysiology.

Oocytes were placed on nylon mesh in a perfusion chamber and continuously perfused (3 ml/min perfusion rate). The perfusion solution had the following composition (in mM): 96 NaCl, 2 KCl, 1 MgCl2, 1.8 CaCl2, and 15 HEPES, pH 7.5. After a stabilization period, when the membrane potential (V m) was constant, the perfusion solution was switched to a CO2/HCO 3 -containing solution of the following composition (in mM): 10 NaHCO3, 86 NaCl, 2 KCl, 1 MgCl2, 1.8 CaCl2, and 15 HEPES, pH 7.5 (gassed with 1.5% CO2). Experiments were performed at room temperature (22–25°C). TheV m values were recorded with conventional microelectrode techniques by glass microelectrodes (resistance 3–5 mΩ) filled with 3 M KCl and connected to an Axoclamp 2A amplifier (Axon Instruments, Foster City, CA) (14, 40). The digitized signals were stored and analyzed on a personal computer with Axotape (Axon Instruments).

Intracellular pH studies.

Intracellular pH (pHi) in oocytes was measured with the pH-sensitive fluorescent probe BCECF (Molecular Probes, Eugene, OR) as described (30, 40). Briefly, oocytes were incubated with 10 μM BCECF-AM for 8–10 min at room temperature and then transferred onto nylon mesh attached to the bottom of a 1-ml perfusion chamber. Fluid was delivered to the chamber by a peristaltic pump in the CO2-impermeable tubing (Cole Palmer) and exchanged at the rate of 4 ml/min. The chamber was closed by a custom-made tightly fitting lid and constantly superfused with the gas mixture of 1.5% CO2-98.5% O2 to prevent CO2 loss and keep the pH constant. pH of the fluid in the chamber was frequently checked with a Horiba pH meter (model B213). Ratiometric fluorescence measurements were performed 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. Ratiometric fluorescence measurements were performed by using the Attofluor digital imaging system. Excitation wavelengths were 450 and 488 nm, and fluorescence emission intensity was recorded at 520 nm. Digitized images were analyzed by using the Attograph and Attoview software packages provided by Attofluor. Intracellular calibration studies were performed by the high-K+-nigericin method. Accordingly, at the end of the experiment, each oocyte was perfused with at least three different pH calibration solutions (6.4, 7, and 7.5). Nigericin (40 μM) was added to each calibration solution, and pH was constantly monitored until it reached a stable value, usually in ∼10 min or less. Calibration points were fitted to a linear regression curve, which was then used for conversion of calculated F488/F450 ratios to cell pH (see Fig. 6 B, right, a representative calibration tracing) and was performed at the end of the experiment depicted in Fig. 6 B, left. The dynamic range of our measurements, the range of acquired ratios vs. the range of pH values used for calibration, was comparable to those published previously for oocytes (30). Rather than performing calibration studies in oocytes separate from the experimental oocytes (30), we obtained a calibration curve at the end of each experiment in the same oocyte that was used for the actual experiment. This modification may account for the differences in the absolute pHi values in our present study compared with the published results (30).

To measure NBC activity, oocytes were first incubated in a HCO 3 -free solution that was identical to the solution used for electrophysiology experiments (seeElectrophysiology). After the equilibration in this solution, oocytes were switched to a Na+- and HCO 3 -containing solution that was similar to the solution used for electrophysiology experiments (seeElectrophysiology), which had a pH of 7.5 and was gassed with 1.5% CO2-98.5% O2. Exposure to the CO2-containing solution resulted in rapid intracellular acidification. NBC activity was measured as the rate of Na+- and HCO 3 -dependent pHirecovery from acidosis. All experiments were performed at room temperature (22°C), and perfusion solutions were continuously bubbled with either 100% O2 or 1.5% CO2-98.5% O2.

Materials

A Marathon cDNA Amplification Kit was purchased from Clontech Laboratories (Palo Alto, CA). DMEM/F-12 medium was purchased from Life Technologies. All other chemicals were purchased from Sigma (St. Louis, MO).

Statistics

Results are given as means ± SE. Statistical comparisons between the groups were performed by ANOVA. The data were considered significant if P < 0.05.

RESULTS

Tissue Distribution of NBC4 in Rat

Figure 1 A is a representative Northern blot hybridization and demonstrates that NBC4 mRNA levels are abundant in proximal colon, distal colon, and kidney, moderate in duodenum and stomach, and very low in heart and liver. To determine whether NBC4 is indeed present in liver and heart, an RT-PCR experiment was performed by using rat-specific primers (see experimental procedures). As shown in Fig.1 B, an NBC4 transcript of expected size is detected in RNA isolated from liver and heart. The expression of NBC4 in the kidney is shown for comparison. The cDNA sequence of the purified PCR band verified its identity as NBC4.

Fig. 1.

A: tissue distribution of rat Na+-HCO Formula cotransporter type 4 (NBC4) by Northern blot hybridization. Rat NBC4 is highly expressed in the colon, with lower expression levels in the kidney and stomach. Expression of NBC4 in the heart and liver was very low. B: tissue distributions of rat NBC4 by RT-PCR. Ethidium bromide staining of an agarose gel demonstrates a PCR product of expected size (∼300 bp) for NBC4 in rat heart, liver, and kidney. Sequencing of the product verified its identity as NBC4. For each tissue, experiments were performed with and without RT. Lane 7, DNA ladder.

Renal Distribution of NBC4

The studies in Fig. 1 indicated that NBC4 mRNA is expressed in the kidney. To examine the zonal distribution of NBC4 in the kidney, RNA from rat kidney cortex, outer medulla, and inner medulla was utilized for RT-PCR by using rat-specific primers. As shown in Fig.2, a PCR fragment, which has abundant expression in the cortex and outer medulla but very low expression in the inner medulla, was identified. The cDNA sequence of the purified PCR fragment verified its identity as NBC4.

Fig. 2.

Distribution of NBC4 in rat kidney by RT-P7CR. Zonal distribution of NBC4 mRNA in the rat kidney by RT-PCR indicates high expression levels in the cortex and outer medulla. NBC4 expression in the inner medulla was very low. Lane 9, DNA ladder. The experiments were performed with and without RT.

To determine whether detectable NBC4 mRNA levels are expressed in various kidney zones, Northern blot hybridizations were performed on RNA isolated from cortex, outer medulla, and inner medulla. A representative blot is shown in Fig.3 A and indicates high NBC4 mRNA expression levels in the outer medulla followed by the cortex. The abundance of NBC4 in the inner medulla was very low (Fig.3 A). This pattern of expression is distinct from NBC1, which is predominantly expressed in the cortex and is very faint or absent in the outer medulla or inner medulla, respectively (Fig. 3 B).

Fig. 3.

A: renal distribution of NBC4 by Northern blot hybridization. A representative blot indicates high NBC4 mRNA expression levels in the outer medulla followed by the cortex. Abundance of NBC4 in the inner medulla was low. B: renal distribution of NBC1 by Northern blot hybridization. A representative blot indicates that NBC1 mRNA is exclusively expressed in the cortex.

Nephron Segment Distribution of NBC4

The presence of NBC4 in the outer medulla (and cortex) is interesting. To determine whether NBC4 expression in the outer medulla (and cortex) is originating in part from mTAL (and cTAL), single-nephron RT-PCR was performed on isolated tubules. Three single mTAL tubules and three single cTAL tubules were pooled separately and subjected to RT-PCR. As indicated in Fig.4 A, mTAL and cTAL tubules abundantly express NBC4. In separate experiments, the expression of NBC4 mRNA in proximal tubule and CCD was examined. Figure 4 Bshows very faint bands of expected size in proximal tubule and CCD. The tubule lengths were comparable and cycling parameters were identical in all nephron segment RT-PCR experiments.

Fig. 4.

Nephron segment RT-PCR of NBC4 in rat kidney.A: representative ethidium bromide staining of agarose gel demonstrates a PCR product of expected size (∼300 bp) for NBC4 in rat medullary thick ascending (mTAL) and cortical thick ascending limb (cTAL). Sequencing of the product verified its identity as NBC4. Experiments were performed with and without RT (separate experiments verified that similar to mTAL, no band was detected in no-RT samples from cTAL, data not shown). Lane 4, DNA ladder.B: PCR product of expected size was detected in the proximal straight tubule (PST) and cortical collecting duct (CCD). Band intensity was vanishingly faint in the CCD and very low in the proximal tubule.

Cloning of a Novel NBC4 Variant (Nucleotide and Amino Acid Sequence)

RT-PCR of human liver RNA with primers encoding the full-length coding region of NBC4 (see experimental procedures) identified two distinct transcripts that when sequenced were confirmed to be novel variants of NBC4 (see experimental procedures). Figure 5 depicts and compares the conceptual translation of the open reading frame of NBC4e with NBC4a and kidney NBC1. As indicated, the sequence of NBC4e is shorter by 86 amino acids compared with the published sequence of NBC4a. The open reading frame of NBC4e encodes a 1,051-amino acid protein. The sequence of NBC4f, which is shorter than NBC4a by 377 amino acids, is not shown here but can be accessed from GenBank (AF-453528).

Fig. 5.

Amino acid sequence of human NBC4. The amino acid sequence of the novel NBC4e variant is compared against NBC4a and kidney NBC1. Potential NH2-linked glycosylation sites were found at amino acids 436, 688, 693, 703, and 711 of NBC4e. Casein kinase II phospholation sites were found at amino acids 78, 82, 115, 161, 227, 255, 277, 356, 428, 471, 472, 713, 721, 813, and 1046. A potential cAMP- and cGMP-dependent protein kinase phosphorylation site was found at amino acid 422. Protein kinase C phosphorylation sites were observed at amino acids 132, 220, 238, 351, 491, 676, 721, 736, 760, 801, and 905.

NBC4 Expression in Oocytes: pHi Studies

In this series of experiments, the functional properties of NBC4e were examined. Toward this end, pHi was monitored in oocytes that were injected with NBC4e cRNA. For comparison, oocytes that were injected with water were utilized. As shown in Fig.6, switching to a CO2/HCO 3 -containing solution resulted in intracellular acidification. The pHi recovered in the presence of Na+ and HCO 3 in NBC4e-injected oocytes (Fig. 6 A, bottom) but not in water-injected oocytes (Fig. 6 A, top). This pHi recovery from acidosis in oocytes injected with NBC4e was completely abolished in the presence of 0.5 mM DIDS (Fig.6 A, bottom). In Na+-free solution, no pHi recovery from acidosis was observed in NBC4e cRNA-injected oocytes (0.000 ± 0.00 pH/min, n = 5). A representative pHi tracing indicating the absence of pHi recovery from acidosis in Na+-free (but HCO 3 -containing) solution in an NBC4e-injected oocyte is shown in Fig. 6 B, left. Figure 6 B,right, is a calibration curve tracing generated by the KCl-nigericin method at the end of the experiment depicted in Fig.6 B, left, using the same oocyte.

Fig. 6.

A: NBC4 expression in oocytes [representative intracellular pH (pHi) tracings]. pHi was monitored in oocytes injected with NBC4e cRNA or water. NBC activity was measured as the Na+-dependent, HCO Formula -mediated pHi recovery from acidosis-induced by exposure to a CO2/HCO Formula -containing solution.Top: control oocytes showed no pHi recovery from acidosis. Bottom: NBC4e-injected oocytes demonstrated significant Na+- and HCO Formula -dependent pHi recovery from acidosis. DIDS, at 0.5 mM, completely inhibited NBC4e-mediated pHi recovery from acidosis.B, left: Na+-dependency of NBC4e (representative pHi tracing). pHi was monitored as above. In the absence of Na+ in the perfusate, no pHirecovery was observed in NBC4e-injected oocytes. Right: KCl-nigericin calibration curve (representative tracing). A pHi calibration tracing generated by KCl-nigericin method at the end of the above experiment is depicted. The 488/450 ratios (right) were utilized to acquire pHi values (left).

The results of Na+-dependent, HCO 3 -mediated pHi recovery studies in water-injected oocytes (control) and in oocytes injected with NBC4 cRNA were compiled. As shown in Fig. 7, the Na+-dependent HCO 3 cotransport activity was 0.08 ± 0.007 pH/min in oocytes injected with NBC4 cRNA (n = 6) and was completely inhibited in the presence of 0.5 mM DIDS. The pHi recovery rate in control oocytes was negligible (0.006 ± 0.001 pH/min,n = 5). In HCO 3 -free solutions, NBC4e-injected oocytes did not demonstrate any significant Na+-dependent pHi recovery from acidosis (0.004 ± 0.00 pH/min, n = 4 oocytes), which was generated by switching from a Na+-containing to a Na+-free solution (the recovery from acidosis was assayed in the presence of 50 μM EIPA to block the endogenous Na+/H+ exchange).

Fig. 7.

NBC4 expression in oocytes (summary of pHitracings). The Na+-dependent HCO Formula cotransport activity was 0.08 ± 0.007 pH/min in oocytes injected with NBC4e cRNA (n = 6) and was completely inhibited in the presence of 0.5 mM DIDS. The pHi recovery rate in control oocytes was negligible (0.006 ± 0.001 pH/min,n = 5).

Baseline pHi was 7.37 ± 0.029 in control oocytes (n = 5) and 7.41 ± 0.025 in oocytes injected with NBC4 cRNA (n = 6; P > 0.05 for both). The rate of Na+-dependent pHi recovery in NBC4-injected oocytes was similar in normal (Cl-containing solutions) and Cl-depleted oocytes (perfused with Cl-free, gluconate-containing solutions for 45 min before pHi recording), indicating that NBC4 does not function in Na+-dependent Cl/HCO 3 exchange mode (the pHi recovery rate from acidosis was 0.09 ± 0.008 in the absence of Cl and 0.08 ± 0.007 in the presence of Cl,P > 0.05, n = 6 for each group).

NBC4 Expression in Oocytes: Electrogenecity

In this series of experiments, we monitoredV m in oocytes injected with NBC4e cRNA. For comparison, oocytes injected with NBC1 cRNA were utilized. Exposure to CO2/HCO 3 in oocytes expressing NBC1 (2–4 days after cRNA injection, 1 μg/μl) resulted in an immediate membrane hyperpolarization, which reached a peak within 3 min (Fig. 8 B, representative tracing). The hyperpolarization showed mild decay during the time of exposure to HCO 3 and was reversible on returning to the HCO 3 -free perfusion solution. This is consistent with published reports and indicates that NBC1 is highly electrogenic (6, 28, 29, 33, 40).

Fig. 8.

NBC4 expression in oocytes: electrophysiological properties (representative tracings). A: water-injected oocyte. Membrane potential (V m) was measured by conventional intracellular microelectrodes in Xenopus laevisoocytes 2–4 days after injection with NBC1 (50 ng/50 nl;B) and NBC4e (10–65 ng/50 nl; C) cRNAs. After an equilibrating period, the perfusion solution was switched to a solution containing 10 mM HCO Formula and gassed with 1.5% CO2-98.5% O2 at pH 7.5. A continuous line indicates the time of exposure to CO2/HCO Formula . B: exposure of NBC1-injected oocytes to CO2/HCO Formula resulted in an immediate and sustained hyperpolarization. The hyperpolarization was reversible on removal of CO2/HCO Formula from the perfusion medium.C: exposure of NBC4e-injected oocytes to CO2/HCO Formula did not alter theV m.

Interestingly, exposure to CO2/HCO 3 in oocytes expressing NBC4e (2–4 days after cRNA injection, 0.2–1.3 μg/μl) did not affect the V m(Fig. 8 C, representative tracing). Figure 8 A is from a control oocyte and indicates that exposure to CO2/HCO 3 did not affect theV m in water-injected oocytes.

The summary of the effects of CO2/HCO 3 in water-injected oocytes and oocytes injected with NBC1 or NBC4e are depicted in Fig. 9. There was no significant difference in the resting V m(measured within 1 min before perfusing with HCO 3 ) of the oocytes in these three groups: −59 ± 2 mV in control oocytes (n = 7), −61 ± 1 mV in NBC1 (n = 11), and −62 ± 1 mV in NBC4 (n = 12). The average V m measured after 3-min exposure to CO2/HCO 3 was significantly altered only in NBC1-injected, but not in NBC4e-injected, oocytes.

Fig. 9.

Summary of expression studies in oocytes injected with NBC1 or NBC4 cRNA. V m was measured by standard intracellular recording techniques. The oocytes were exposed to a control medium (HCO Formula -free) followed by a medium that contained CO2/ HCO Formula (same protocol as in Fig. 8). The values of V m were measured within 1 min before exposure to CO2/HCO Formula (filled bars) and 3 min after CO2/ HCO Formula exposure (open bars). Values are means ± SE of number of oocytes [n = 12 control (CTRL), n = 9 NBC4, and n = 21 NBC1]. A statistically significant hyperpolarization on exposure to CO2/ HCO Formula was observed in NBC1- but not NBC4-injected oocytes (* P < 0.001).

The above V m recordings and pHitracings were obtained from separate oocytes. In the next series of experiments, we performed V m recordings followed by pHi measurement in the same oocyte. Accordingly, oocytes were subjected to electrophysiological measurement and, on the completion of V m recording, were removed from the chamber, loaded with BCECF, and subjected to pHimeasurement. Figure 10 A is a representative recording and indicates that no changes inV m were observed in response to exposure to CO 2 /HCO 3 in NBC4e-transfected oocytes. Figure 10 B is a representative pHitracing from the same oocyte and indicates a significant pHi recovery from acidosis. Control oocytes did not demonstrate any pHi recovery from acidosis. The averaged results of six separate measurements in six NBC4e-expressing oocytes indicated no significant changes in V m in response to exposure to CO 2 /HCO 3 (P > 0.05 vs. control oocytes, n = 6). The pHi recovery rate from acidosis was 0.11 ± 0.01 in NBC4e-expressing oocytes, whereas it was undetectable in control oocytes (P < 0.01, n = 6 for each group). Taken together, these results further confirm that NBC4e is an electroneutral NBC isoform.

Fig. 10.

NBC4 expression: V m and pHi monitoring in the same oocyte (representative tracings). A: V m was measured in NBC4e-injected oocytes in a manner similar to Fig. 8. B: after the completion of an electrophysiological recording, oocytes were removed from the chamber, loaded with BCECF, and monitored for NBC activity by determining the pHi recovery from acidosis in a manner similar to that in Fig. 6. A: as indicated, exposure of NBC4-injected oocytes to CO2/HCO Formula did not alter the V m. B: same oocyte shows a robust pHi recovery from acidosis. Control oocytes (noninjected or water injected) did not demonstrate any pHirecovery from acidosis.

DISCUSSION

NBC4 mRNA expression in the kidney is abundant in the outer medulla and cortex and low in the inner medulla (Figs. 2 and 3). NBC4 is predominantly expressed in the mTAL and cTAL of the loop of Henle (Fig. 4). Functional studies indicated that NBC4 mediates a DIDS-inhibitable, Na+-dependent HCO 3 cotransport (Figs. 6, 7, and 10) and, unlike NBC1, which is highly electrogenic, is electroneutral (Figs. 8-10).

In the present studies, a novel human NBC4 transcript was cloned and characterized. The NBC4 variant has a 3153-nucleotide open reading frame and encodes a 1,051-amino acid protein (Fig. 5). Similar to other NBC isoforms, NBC4e has 10 transmembrane-spanning domains. Amino acid sequence analysis demonstrated that similar to NBC1 (9), NBC4 has one cAMP-dependent protein kinase phosphorylation site and one tryosine kinase phosphorylation site (Fig. 5). NBC4e has five potential NH2-linked glycosylation sites (436, 688, 703, and 711 amino acids) and multiple casein kinase and protein kinase C phosphorylation sites.

An interesting aspect of these studies is the distinct pattern of NBC4 distribution in the kidney (Figs. 2-4). NBC4 expression is abundant in the outer medulla, with lower levels in the cortex. The inner medulla has the least amount of NBC4 among kidney zones. This is distinct from NBC1, which is highly expressed in the cortex and is very faint or absent in the outer medulla and inner medulla, respectively (Fig. 3 B and Refs. 2, 29, and33). NBC4 is predominantly expressed in mTAL and cTAL of the loop of Henle, sites that do not express NBC1 (2, 29,33), suggesting that NBC4 and NBC1 perform distinct functions. The accuracy of nephron segment isolation was verified by examining the isolated tubules for the expression of apical Na-K-2Cl cotransporter, which is exclusively expressed in mTAL and cTAL. RT-PCR studies verified the expression of Na-K-2Cl cotransporter in isolated mTAL and cTAL tubules.

NBC1 is highly electrogenic (Figs. 8 and 9) and mediates the transport of HCO 3 and Na+ from cell to blood in the kidney proximal tubule (1-3, 6, 9-11, 28, 29, 33,41). This occurs despite unfavorable chemical gradients for Na+ and HCO 3 efflux across the plasma membrane (3, 6, 9, 34, 41). The only driving force for NBC1 to work in the efflux mode is the fact that it is electrogenic and carries a net negative charge (3, 6, 9, 28, 33), consistent with published reports that it carries 3 HCO 3 /Na+ (34, 41). As such, the inside-negative V m of the proximal tubule cell can drive the net exit of HCO 3 across the basolateral membrane against the inward Na+ and HCO 3 gradients. Interestingly, expression of NBC4 in oocytes did not alter the V m (Figs. 8-10), consistent with the conclusion that this transporter is electroneutral. The electroneutrality of the NBC indicates the cotransport of 1 HCO 3 with 1 Na+ (or 2 HCO 3 with 2 Na+). We suggest that on the basis of its electroneutrality, NBC4 works in the influx mode under physiological conditions and transports HCO 3 and Na+ into cells.

The inward NBC4 flux direction should result in the transport of HCO 3 into the cell, with subsequent cell alkalinization. As such, NBC4 is likely activated in response to cell acidosis (such as in metabolic acidosis). In this regard, NBC4 can function as a housekeeping transporter and maintain the cell pH in a narrow physiological range. With respect to the mTAL, the transport of HCO 3 into the cell may prevent the cell acidification that results from the reabsorption of NH 4+ via the apical Na-K (NH 4+ )-2Cl cotransporter (22). As such, NBC4 may play an important role in systemic acid-base regulation by allowing the reabsorption of NH 4+ to continue uninterrupted in the mTAL of the loop of Henle. It should be mentioned that the presence of Na+-dependent HCO 3 cotransport in the basolateral membrane of thick limb of the loop of Henle in rats on a normal diet has been controversial; whereas some studies have found functional evidence in support of NBC, others have not (5, 38). One plausible explanation with respect to this discrepancy may be the use of different methods of cell isolation for functional studies (5,38). This latter study found that mTAL tubules express NBCn1 (which is a rat ortholog of NBC2) on their basolateral membranes (38). The expression of NBCn1 was found to be increased in acidosis (22). Our studies indicate that in addition to NBCn1 (38), mTAL tubules also express NBC4. In view of the fact that NBCn1 is resistant to DIDS (13), we propose that the DIDS-sensitive NBC activity in the basolateral membrane of the mTAL (38) may reflect NBC4 activity and not NBCn1. With respect to the liver, NBC4 can function in tandem with the electrogenic NBC and play an important role in the transport of HCO 3 into hepatocytes (17).

Two recent reports suggested that NBC4c (a variant of NBC4) is electrogenic and carries a net negative charge (31, 37). This conclusion, although at variance with the present results (Figs.8-10), may be due to the possibility that those studies (31,37) used a splice variant of NBC4 distinct from our clone. Whether the truncation of NBC4 by 86 amino acids could be responsible for the lack of electrogenecity of NBC4e in the present results (Figs.8-10) remains to be studied. It should be mentioned that the recordings of V m and pHi recovery from acidosis in response to CO2/HCO 3 exposure in the same oocyte (Fig. 10) prove beyond any doubt that NBC4e is indeed electroneutral. In support of this conclusion, it is worth mentioning that no electrogenic NBC has been identified in the rat mTAL of the loop of Henle, lending support to the conclusion that NBC4e is indeed electroneutral.

Rat NBC4 has a molecular mass of 3.5 kb (Figs. 1-3), which is much smaller than the human NBC4 molecular mass of 6 kb (26). The difference likely results from the presence of a long noncoding region in human NBC4, because the coding region of human NBC4 is only ∼3.2 kb. The mRNA of rat NBC4 in the outer medulla and cortex appears as two transcripts, raising the possibility that similar to human NBC4, rat NBC4 has more than one splice variant. Interestingly, the larger transcript is absent in the inner medulla and stomach. Additional studies are needed to clone and examine the regulation of rat NBC4 transcripts.

Although the subcellular localization of NBC4 was not determined in these studies, it is likely that this transporter is expressed on the basolateral membrane of epithelial cells. This conclusion is strongly supported by nephron segment studies indicating the expression of NBC4 in the mTAL and cTAL (Fig. 4) as well as functional studies indicating the lack of any luminal Na+-dependent HCO 3 cotransporter activity in mTAL and cTAL (18). As such, the expression of NBC4 in these two nephron segments is likely limited to the basolateral membrane domain.

In conclusion, NBC4e is a novel variant of NBC4 and mediates a DIDS-sensitive electroneutral cotransport of Na+ and HCO 3 . On the basis of its tissue distribution, we propose that NBC4 mediates the transport of HCO 3 into the cells of the thick ascending limb of the loop of Henle.

Acknowledgments

These studies were supported by National Institute of Diabetes and Digestive and Kidney Diseases Grant DK-54430, a Merit Review grant, a grant from the Cystic Fibrosis Foundation, and grants from Dialysis Clinic, Inc. (M. Soleimani). L. Conforti was supported by American Heart Association Grant 0030091N.

Footnotes

  • The nucleotide sequences reported in this paper have been submitted to GenBank/EBI Data Bank as AF-452248 (NBC4e) and AF-453528 (NBC4f).

  • Address for reprint requests and other correspondence: M. Soleimani, Division of Nephrology and Hypertension, Univ. of Cincinnati Medical Ctr., 231 Albert Sabin Way, MSB G259, Cincinnati, Ohio 45267-0585 (E-mail:Manoocher.Soleimani{at}uc.edu).

  • The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

  • August 13, 2002;10.1152/ajprenal.00055.2002

REFERENCES

View Abstract