INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

THE EXPRESSION CLONING of the first Na⁺-driven bicarbonate transporter (NBC) by Romero et al. (36) led to the cloning of many other electrogenic and electroneutral Na⁺-driven HCO transporters. The original electrogenic NBC in mammals (NBCe1) is currently represented by three splice variants: NBCe1-A, which is found predominantly in kidney (8, 35); NBCe1-B, which is found in pancreas, heart, and many other tissues (1, 12); and NBCe1-C, which is found predominantly in brain (4).

The electroneutral Na⁺-driven HCO transporters in mammals share ~70% sequence identity on the amino acid level and appear to fall into two functional groups. The first group mediates Cl-independent Na⁺-HCO cotransport. The first member of this group has been cloned under various names, including NBC2 from human retina (22), NBC3 from human skeletal muscle (31), and NBCn1 from rat smooth muscle (11). However, the NH₂-terminal 90 amino acids of NBC2 are missing, and the next 28 represent a cloning artifact.

The second group of electroneutral Na⁺-driven HCO transporters mediates the Cl-dependent transport of Na⁺ and HCO, functionally described as Na⁺-driven Cl/HCO exchange. In mammals, this transporter is represented by a clone from human brain, NDCBE1 (18). A partial clone had been named NBC3 (3), representing degeneracy in the nomenclature. A related Drosophila clone (NDAE) encodes a Na⁺-driven anion exchanger (37).

NBC4, originally cloned by Pushkin et al. (32), is a new member of the HCO transporter superfamily. It is ~60% identical at the amino acid level to the three electrogenic NBCe1 splice variants and 40-50% identical to the electroneutral Na⁺-driven HCO transporters. A Northern blot (32) and a similar mirror image (33) showed high levels of NBC4 mRNA in liver, spleen, and testes and moderate levels in heart, kidney, placenta, and stomach.

We mapped the NBC4 sequence onto a recent topology model developed for the anion exchanger AE1 (Fig. 1A), which is in the same superfamily as the Na⁺-coupled HCO transporters. Pushkin et al. (32, 33) described two variants of NBC4, NBC4a and NBC4b, and have submitted two additional sequences, NBC4c and NBC4d, to GenBank. As shown in Fig. 1B, NBC4a and NBC4b each contain, beginning just after the predicted transmembrane segment (TM) 11, a putative exon not present in either NBC4c or NBC4d. NBC4b has a unique 16-bp insertion in its last TM. This insertion/frameshift introduces a premature stop codon and causes the deduced amino acid sequence to exhibit little homology with other members of the HCO transporter (BT) superfamily. NBC4d is missing two exons present in NBC4c, resulting in the elimination of all amino acids between the putative end of TM 10 and near the beginning of TM 14. To date, no functional data have been reported for any of the four NBC4 variants.

View larger version (26K): [in this window] [in a new window]   Fig. 1.   Na+-driven bicarbonate transporter (NBC)4 sequence analysis and topology. A: using sequence alignment analysis, we mapped the NBC4c amino acid sequence on a recent topology model developed for anion exchanger (AE)1 (43). The putative NH2 and COOH termini are intracellular. Gray cylinders indicate transmembrane segments (TMs) 1-14, of which TM 12 is nonhelical (stippled). Four glycosylation consensus motifs are located on the large extracellular loop connecting TMs 5 and 6. B: gray bars indicate the relative length of the amino acid sequence of the 4 NBC4 variants. The position of the putative TMs are indicated by numbered vertical light gray bars. Variant b has a 16-bp insertion (cross-hatching on dark background) in the middle of the last TM, which causes a frameshift and results in a unique COOH terminus (cross-hatching on white background). Variant c lacks a 48-bp exon (between TMs 11 and 12) present in variants a and b. Variant d lacks 2 exons in addition to the 1 missing in variant c, resulting in the loss of the region between TMs 10 and 14.

We have undertaken the present study to investigate the functional properties of NBC4. Using PCR, we obtained nine clones corresponding to the NBC4c coding region but none corresponding to the other NBC4 variants. We have expressed NBC4c in Xenopus laevis oocytes and assayed for pH regulatory function as well as membrane potential and ionic currents. The results show that NBC4c is an electrogenic Na⁺-HCO cotransporter that, at least in the oocyte, appears to operate with a HCO:Na⁺ stoichiometry of 2:1.

Portions of this work have been published in abstract form (44).

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

cDNA Cloning

Expression in Oocytes

Stage V-VI oocytes from Xenopus laevis were isolated as described previously (35). One day after isolation, the oocytes were injected with 50 nl of a solution containing 0.5 ng/nl of mRNA encoding NBC4c. Control oocytes were injected with 50 nl of sterile water. The oocytes were used in experiments 2-5 days after injection. All experiments were performed at room temperature (~22°C).

Solutions

Electrophysiological Measurements

Measurement of intracellular pH. We assayed pH regulatory function of oocytes expressing NBC4c by measuring intracellular pH (pH_i) with pH-sensitive microelectrodes. The electrodes were fabricated and used as described previously (36, 40). Briefly, the oocyte was impaled with two microelectrodes, one for measuring the membrane potential (V_m) and the other for measuring pH_i. The tip of the pH electrode contained a liquid membrane across which a pH-dependent voltage was generated. pH_i was obtained by subtracting the signal of the V_m electrode from that of the pH electrode. A calomel electrode was used as the reference in the bath. Voltages were measured with an FD 223 electrometer (World Precision Instruments, Sarasota, FL), and data were acquired with software written in-house. The system was calibrated with buffered pH standards at pH 6.0 and 8.0. An additional single-point calibration was performed with the standard ND96 solution of pH 7.50 in the bath before the oocyte was impaled.

Two-electrode voltage clamp. We used two-electrode voltage clamp to measure whole cell ionic currents in oocytes expressing NBC4c or injected with water (control). Oocyte currents and voltages were recorded with a model OC-725C oocyte clamp (Warner Instruments, Hamden, CT) controlled by the Clampex module of pCLAMP software (Version 8; Axon Instruments, Foster City, CA). Electrodes were pulled from thin-walled borosilicate glass and had resistances of 0.4-1.0 M when filled with 3 M KCl. Oocytes were held at a potential close to the spontaneous V_m until the initiation of the voltage-clamp protocol. Current-voltage (I-V) relationships were generated by stepping the holding potential from 140 mV to +40 mV in 20-mV increments, each of which lasted 200 ms. Data were analyzed with the Clampfit module of pCLAMP.

Stoichiometry

The reason for returning to Na⁺-free solutions for 2-3 min between I-V curves was to minimize changes in intracellular Na⁺ concentration ([Na⁺]_i) and [HCO]_i due to NBC4 transport activity; these Na⁺-free periods prompted NBC4 to run backwards, presumably depleting the Na⁺ and HCO that had built up as NBC4 was running in the forward direction in the presence of Na⁺. We obtained HCO-dependent E_rev from the above I-V curves by subtracting the HCO curve measured in ND96 (CO₂/HCO free) from each of the I-V curves measured in the presence of CO₂/HCO at different Na⁺ concentrations.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Nine Consecutive PCR Products Represented NBC4c

NBC4 is Electrogenic, Na+ Dependent, and Cl Independent

pH_i data. Figure 2A shows a recording of pH_i and V_m for an oocyte injected with cRNA encoding NBC4c, and Fig. 2B shows a similar recording for a water-injected (control) oocyte. The initial pH_i of the NBC4c-expressing oocytes was slightly higher (7.34 ± 0.01; n = 18) than in the controls (7.22 ± 0.03; n = 9), suggesting that the ambient HCO in the incubation medium had provided enough substrate for at least some NBC4c activity since the time of the cRNA injection. We then switched the superfusate from ND96 to 5% CO₂/33 mM HCO, which caused a dramatic hyperpolarization of the NBC4c-expressing oocyte, followed by a slower relaxation of V_m toward more positive values. The rapid hyperpolarization is consistent with the hypothesis that NBC4c mediates the uptake of Na⁺ and the equivalent of two or more HCO ions. In addition, pH_i began to fall rapidly, reflecting the entry of CO₂ into the oocyte, the subsequent hydration to form H₂CO₃, and dissociation to liberate HCO and H⁺. In the continued presence of CO₂/HCO, pH_i eventually started to recover slowly (Table 1), consistent with the NBC4c-mediated uptake of HCO. When we removed extracellular Na⁺ (replacing it with NMDG⁺), the cell rapidly depolarized and began to acidify, indicating that the direction of transport had been reversed. Reintroducing Na⁺ caused the opposite set of V_m and pH_i changes. At the end of the experiments, removing CO₂/HCO caused the cell to depolarize and alkalinize rapidly. The rapid depolarization probably reflects the temporary reversal of NBC4c caused by the removal of extracellular HCO while [HCO]_i is still high. The rise in pH_i reflects the exit of CO₂, which produces the depletion of HCO that is reflected by the slower negative drift of V_m.

View larger version (12K): [in this window] [in a new window]   Fig. 2.   Recording of intracellular pH (pHi) and membrane potential (Vm). A: representative pHi and Vm trace of an oocyte expressing NBC4c. The oocyte was initially superfused with ND96. During the indicated period, the oocyte was superfused with the 5% CO2/33 mM HCO solution. CO2/HCO-induced hyperpolarization averaged 90 ± 6 mV in 6 experiments. While in the CO2/HCO solution, extracellular Na+ was removed as indicated. This resulted in a marked depolarization, averaging +95 ± 10 mV in the same experiments. B: data obtained with a similar protocol on a water-injected oocyte. Arrowheads indicate when Na+ was removed or reintroduced. The experiment was performed on 5 water-injected oocytes.

Voltage-clamp data. Using a two-electrode voltage clamp, we further explored the electrogenic nature of NBC4c. Figure 3A shows representative current records obtained in an NBC4c-expressing oocyte under three conditions: the nominal absence of CO₂/HCO, the presence of 5% CO₂/33 mM HCO, and the absence of Na⁺ in the continued presence of CO₂/HCO. Introducing CO₂/HCO elicits a large increase in the whole cell current, whereas removing Na⁺ virtually abolishes the NBC4c-dependent outward currents.

View larger version (22K): [in this window] [in a new window]   Fig. 3.   Voltage-clamp experiments. A: representative current recordings obtained from an NBC4c-expressing oocyte. The top tracing was acquired in ND96, and the middle and bottom tracings were acquired in 5% CO2/33 mM HCO. In the bottom tracing, external Na+ was removed. B: representative current-voltage (I-V) curves for NBC4c-expressing oocytes generated from current tracings similar to those in A. Data were acquired in ND96 solution (), CO2/HCO solution (), Cl-free CO2/HCO solution (), and Na+-free CO2/HCO solution (). C: representative I-V curves generated from water-injected oocytes (dashed lines) and NBC4c-expressing oocytes (solid lines), all in the nominal absence of CO2/HCO. Data from water-injected oocytes were acquired in ND96 solution () and CO2/HCO solution (). Data from NBC4c-expressing oocytes were acquired in ND96 solution (), Na+-free ND96 solution (), and Cl-free ND96 solution (). We performed a total of 8 such experiments for water-injected oocytes and 7 for NBC4c-injected oocytes. Results were similar to those shown.

Electrical Gradient Can Drive Substantial Acid-Base Transport Through NBC4c

To test this hypothesis, we performed an experiment similar to that shown in Fig. 2A except that, at times, we not only monitored pH_i but voltage clamped as well. The oocyte had a spontaneous V_m of about 50 mV when impaled only with the voltage electrode. V_m shifted to about 45 mV when we introduced the current electrode and initially fell to about 25 when we introduced the pH electrode. By the beginning of the record in Fig. 4, V_m had recovered somewhat, but this oocyte was still rather "leaky." Introducing CO₂/HCO caused V_m to shift only to 100 mV (compared with about 155 mV for the tighter oocyte in Fig. 2A). At the times indicated in Fig. 4, we acquired I-V curves by using the same voltage protocol as in Fig. 3. We calculated E_rev for the NBC4c-dependent currents by subtracting the I-V curve acquired in ND96 from the I-V curves acquired in the CO₂/HCO solutions. For the first I-V curve, E_rev was 110 mV, compared with a spontaneous V_m of about 95 mV. This ~15-mV difference between E_rev and V_m would explain why the spontaneous pH_i recovery is faster in Fig. 4 than in Fig. 2. When we then voltage-clamped the oocyte in Fig. 4 to 30 mV, pH_i increased approximately six times more rapidly than when V_m was free-floating at about 95 mV. This large increase in the pH_i recovery rate is consistent with the large increase in the difference between E_rev (90 mV) and the clamped V_m of 30 mV. When we reversed the transporter by removing extracellular Na⁺ in Fig. 4 (V_m = 30 mV), pH_i initially decreased more rapidly than in Fig. 2 (V_m = +2 mV). Thus a large electrical gradient can substantially increase acid-base transport through NBC4c. The rates of pH_i change in oocytes clamped to 30 mV are compared with those of unclamped oocytes in Table 1.

View larger version (17K): [in this window] [in a new window]   Fig. 4.   Measurement of pHi in a voltage-clamped oocyte. The oocyte was initially superfused with ND96 without clamping the voltage. During the indicated period, the oocyte was superfused with the 5% CO2/33 mM HCO solution. While in the CO2/HCO solution, first extracellular Na+ and then also extracellular Cl were removed as indicated. Gray area indicates the period during which the oocyte was voltage-clamped to 30 mV. Asterisks indicate the time points at which an I-V curve was acquired. We performed a total of 3 such experiments on different NBC4c-expressing oocytes, obtaining results similar to those shown.

While continuing to voltage clamp the oocyte in Fig. 4 at 30 mV, we removed extracellular Cl in the absence of Na⁺ to revisit the question of whether acid-base transport by NBC4c requires extracellular Cl. If NBC4c were a Na⁺-driven Cl-HCO exchanger, then it should require extracellular Cl when running in reverse. However, even at a relatively high rate of pH_i decrease in Fig. 4, Cl removal had no effect on the pH_i trajectory. Thus NBC4c does not mediate Na⁺-driven Cl-HCO exchange.

NBC4c is HCO Dependent

View larger version (13K): [in this window] [in a new window]   Fig. 5.   HCO dependence of NBC4. A: representative pHi and Vm traces of an oocyte expressing NBC4c. The oocyte was initially superfused with ND96. During the indicated time period, the oocyte was superfused with 30 mM butyrate solution at a constant extracellular pH of 7.50. After washout of the butyrate-containing solution, the oocyte was reacidified with 5% CO2-33 mM HCO solution. B: data obtained with a similar protocol on a water-injected oocyte. We performed a total of 3 such experiments on different water-injected oocytes and 3 on NBC4c-expressing oocytes. Results were similar to those shown.

NBC4c Is Blocked by DIDS

pH_i data. To determine whether NBC4c is sensitive to 4,4'-diisothiocyanostilbene-2,2'-disulfonic acid (DIDS), we exposed NBC4c-expressing oocytes and control oocytes to 200 µM DIDS during the plateau phase of a CO₂/HCO pulse. Figure 6A shows that in an NBC4c-expressing oocyte, application of DIDS markedly reduced the rate of pH_i recovery and also caused a modest depolarization. DIDS reduced the mean rate of pH_i recovery from 21.5 ± 4.2 to 3 ± 1.2 × 10⁵ pH units/s (n = 5), an inhibition of ~80%. Thus DIDS strongly inhibits the transport function of NBC4c.² Although the inhibition of some HCO transporters by DIDS is poorly reversible, the pH_i recovery in oocytes expressing NBC4c resumed on washout of DIDS. In the control oocyte (Fig. 6B), pH_i did not recover after the initial acidification and application of DIDS had no effect on pH_i. The mean rate of pH_i change was 0.7 ± 2.2 × 10⁵ pH units/s before application of DIDS and 1.7 ± 0.9 × 10⁵ pH units/s after application of DIDS (n = 3).

View larger version (13K): [in this window] [in a new window]   Fig. 6.   Effect of 4,4'-diisothiocyanostilbene-2,2'-disulfonic acid (DIDS) on pHi recovery from an acid load. A: representative pHi and Vm traces of an oocyte expressing NBC4c. The oocyte was initially superfused with ND96. During the indicated time period, the oocyte was superfused with the 5% CO2/33 mM HCO solution. DIDS (200 µM) was applied during the CO2/HCO exposure. B: data obtained with a similar protocol on a water-injected oocyte. We performed a total of 3 such experiments on different water-injected oocytes and 5 on different NBC4c-expressing oocytes. Results were similar to those shown.

Voltage-clamp data. Using two-electrode voltage clamp, we investigated the effect of 200 µM DIDS on the current carried by NBC4c. Because work by Diakov et al. (14) has shown that DIDS (at a concentration of 250 µM) can activate endogenous ion channels in the oocyte over a period of 15-60 s, we exposed the oocytes to DIDS for no more than ~20 s, which we found to be sufficient for maximal inhibition.

View larger version (13K): [in this window] [in a new window]   Fig. 7.   Effect of DIDS on ionic currents. A: representative I-V curves for NBC4c-expressing oocytes. Data were acquired in ND96 solution (), CO2/HCO solution (),CO2/HCO with 200 µM DIDS (), and CO2/HCO solution after washout of DIDS (). B: representative I-V curves generated from control oocytes (dashed lines) and NBC4c-expressing oocytes (solid lines), all in the nominal absence of CO2/HCO. Data from water-injected oocytes were acquired in ND96 solution () and ND96 solution with DIDS (). Data from NBC4-expressing oocytes were acquired in ND96 solution (), ND96 with 200 µM DIDS (), and ND96 solution after washout of DIDS (). C: average DIDS-sensitive currents in NBC4c-expressing oocytes. DIDS-sensitive current in ND96 solutions () is the average obtained by subtracting data from 5 experiments similar to those shown in B ( and ). DIDS-sensitive current in CO2/HCO solutions () is the comparable average for 5 different experiments similar to those shown in A ( and ). Vertical bars indicate SE; bars are omitted when they would have been smaller than the symbol.

Apparent HCO:Na⁺ Stoichiometry of NBC4c is 2:1

Slope of E_rev vs. log of extracellular Na+ concentration. To determine the HCO:Na⁺ stoichiometry of NBC4c, we used a two-electrode voltage clamp to measure the reversal potentials of HCO-dependent currents in NBC4c-expressing oocytes at four different values of extracellular Na⁺ concentration ([Na⁺]_o) (see METHODS). E_rev of an electrogenic Na⁺-HCO cotransporter is given by the following equation (5)

(1)

where q is the HCO:Na⁺ stoichiometry, [Na⁺]_i, [Na⁺]_o, [HCO]_i, and [HCO]_o are the respective intracellular and extracellular concentrations, R is the gas constant, T is temperature, and F is the Faraday constant. This equation predicts that a plot of E_rev vs. log [Na⁺]_o should be a straight line, whose slope is ln(10) × RT/[F(q  1)]. Thus the higher the HCO:Na⁺ stoichiometry, the less steep the slope of the line. Thus, for a HCO:Na⁺ stoichiometry of 2:1, the expected slope at 22°C would be 58.5 mV/log [Na⁺]_o, whereas for a stoichiometry of 3:1, the slope would be only half as steep, 29.3 mV/log [Na⁺]_o. The results of five experiments (each including all 4 [Na⁺]_o values) are summarized in Fig. 8, which is a plot of E_rev against [Na⁺]_o. Our experimental data are fitted well by a straight line with a slope of 54.8 mV/log [Na⁺]_o, consistent with the hypothesis that NBC4c operates with a 2:1 stoichiometry.

View larger version (12K): [in this window] [in a new window]   Fig. 8.   Dependence of the NBC4 reversal potential (Erev) on extracellular Na+ concentration ([Na+]o). Using a 2-electrode voltage-clamp technique, we measured the Erev of NBC4c-expressing oocytes in CO2/HCO solutions at 4 different values of [Na+]o. The plot of Erev vs. log[Na+]o was fitted with a straight line (R2 = 0.9356), the slope of which is 54.8 mV/log[Na+]o. This slope is consistent with a HCO:Na+ stoichiometry of 2:1. Each symbol represents the mean ± SE. We performed a total of 5 experiments on different NBC4c-expressing oocytes.

Absolute value of E_rev. Another approach for estimating stoichiometry is to compute q directly from each E_rev value. Solving Eq. 1 for q yields

(2)

We calculated [HCO]_i in our experiments from average pH_i data and the Henderson-Hasselbalch equation, obtaining a mean value of 7.6 mM. Sciortino and Romero (39) found that oocytes expressing rat kidney NBCe1 and bathed in 5% CO₂/33 mM HCO had a mean [Na⁺]_i of 10.4 mM. Using these values for [HCO]_i and [Na⁺]_i, 33 mM for [HCO]_o, and the four different [Na⁺]_o values from our experiments, we computed the four q values summarized in Table 2. The mean value for all data, 2.1 ± 0.1, is consistent with a 2:1 stoichiometry.

View this table: [in this window] [in a new window]   Table 2.   Apparent HCO-to-Na+ stoichiometry ratios

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The aim of this study was to determine the functional properties of NBC4, originally cloned from human heart by Pushkin et al. (32) but uncharacterized physiologically. Currently GenBank contains the sequences for four different NBC4 variants (NBC4a, b, c, and d). Our own cloning efforts resulted in the cloning of the NBC4c variant, which we have expressed in Xenopus oocytes. We report here that NBC4c carries out electrogenic Na⁺-HCO cotransport. Furthermore, NBC4c is independent of extracellular Cl, is blocked by DIDS, and operates with an HCO:Na⁺ stoichiometry of 2:1 when expressed in oocytes. These results correlate well with sequence alignment analysis, which shows that, on the amino acid level, NBC4 is most closely related to electrogenic NBCe1. On the basis of the function of NBC4c, we propose to call it NBCe2-c.

Different NBC4 Variants

Comparing the cDNA sequence of NBC4c with the human genome database reveals the presence of 25 exons that span the length of the coding region. Variants a and b share an additional exon not present in c and d. The presence of this additional exon results in a 16-amino acid insertion into the very short region between putative TMs 11 and 12, a region that is highly conserved among all superfamily members. In addition, the third-to-last exon in variant b is longer than in the other variants because, at its 5' end, it contains 16 bp of intronic sequence (Fig. 1B). This 16-bp insertion creates a frameshift in the middle of the last TM, resulting in a truncated COOH terminus (Fig. 1B) that has no homology to other members of the BT superfamily. Variant d is lacking two additional exons between TMs 10 and 14, resulting in a 294-bp deletion in a region that is very highly conserved in all members of the BT superfamily. Thus variant c is the only one of the four NBC4 variants whose deduced amino acid sequence maps onto the superfamily consensus sequence without major gaps or insertions. Indeed, nine consecutive cDNA clones in our studyobtained by PCR using, as a template, a mixture of human cDNAs from different tissuesproved to be of the c variety. If the four original NBC4 mRNA species had been evenly distributed, and if the PCR efficiencies for the respective cDNA species were identical, then the odds of picking the c variant nine times in a row would be 1 in 4⁹. It will be informative to learn whether variants a, b, and d encode functional proteins.

DIDS

In NBC4c, the sequence at the homologous TM 5 site is KMIG. Thus DIDS blocks the function of NBC4c even though NBC4c lacks the second K of the consensus motif. However, NBC4c is not the first DIDS-sensitive BT family member with a disrupted consensus motif. NDAE1 from Drosophila (37) and human NDCBE (18) are both DIDS sensitive, even though the sequences at their respective homologous sites are NVMV and KLIH. One explanation for these observations is that DIDS may inhibit transport by NBC4c (and also by NDAE and NDCBE1) via an ionic interaction that does not require the consensus DIDS reaction motif. Alternatively, DIDS may covalently react at another site on the transporter molecule. Distinguishing among these possibilities will require a structure-function analysis of the three different aspects of the interaction of DIDS with the Na⁺-coupled HCO transporters: 1) inhibition of transport, 2) ionic or reversible binding, and 3) covalent or irreversible binding.

NBC4 in Liver

NBC4 in Kidney

The only thing that is known about the renal localization of NBC4 is that the message is present in the kidney lane of human Northern blots. Thus it is not clear whether NBC4 participates in HCO reabsorption. Our results suggest that NBC4c operates with a 2:1 stoichiometry in oocytes, so that NBC4c would be expected to mediate net HCO uptake (i.e., acid extrusion) in most cells. In this context, NBC4c could help to protect against intracellular acidification in cells throughout the kidney. If present at an apical membrane, and operating with a 2:1 stoichiometry, NBC4c could thus contribute to HCO reabsorption. If NBC4c were present at the basolateral membrane and functioned with a 3:1 stoichiometry in the proximal tubule, then it could contribute to HCO reabsorption in that nephron segment. In principle, even with a 2:1 stoichiometry, basolateral NBC4 could contribute to HCO reabsorption elsewhere in the nephron, provided that [Na⁺]_i and/or [HCO]_i were sufficiently high and/or provided that the basolateral membrane potential were sufficiently negative.

NBC4 in Heart