Na/Bicarbonate Cotransporter NBCn1 in the Kidney Medullary Thick Ascending Limb Cell Line is Upregulated under Acidic Conditions and Enhances Ammonium Transport

Abstract

In this study, we examined the effect of bicarbonate transporters on ammonium/ammonia uptake in the medullary thick ascending limb cell line ST-1. Cells were treated with 1 mM ouabain and 0.2 mM bumetanide to minimize carrier-mediated NH₄⁺ transport, and the intracellular accumulation of ¹⁴C-methylammonium/methylammonia (MA) was determined. In ${\mathrm{CO}}_2∕{\mathrm{HCO}}_3^{-}\text{-free}$ solution, cells at normal pH briefly accumulated ¹⁴C-MA over 7 min and reached a plateau. In ${\mathrm{CO}}_2∕{\mathrm{HCO}}_3^{-}$ solution, however, cells markedly accumulated ¹⁴C-MA over the experiment period of 30 min. This ${\mathrm{CO}}_2∕{\mathrm{HCO}}_3^{-}\text{-dependent}$ accumulation was reduced by the bicarbonate transporter blocker 4,4’-diisothiocyanatostilbene-2,2’-disulfonate (DIDS; 0.5 mM). Replacing Cl^– with gluconate reduced the accumulation but the reduction was more substantial in the presence of DIDS. Incubating cells at pH 6.8 (adjusted with NaHCO₃ in 5% CO₂) for 24 h lowered the mean steady-state intracellular pH to 6.96, significantly lower than 7.28 for controls. DIDS reduced ¹⁴C-MA accumulation in controls but had no effect after acidic incubation. Immunoblot showed that NBCn1 was upregulated after acidic incubation and in NH₄Cl-containing media. The Cl/HCO₃ exchanger AE2 was present but its expression remained unaffected by acidic incubation. Expressed in Xenopus oocytes, NBCn1 increased carrier-mediated ¹⁴C-MA transport, which was abolished by replacing Na⁺. Two-electrode voltage clamp of oocytes exhibited negligible current after NH₄Cl application. These results suggest that DIDS-sensitive ${\mathrm{HCO}}_3^{-}$ extrusion normally governs NH₄⁺/NH₃ uptake in the MTAL cells. We propose that, under acidic conditions, DIDS-sensitive ${\mathrm{HCO}}_3^{-}$ extrusion is inactivated while NBCn1 is upregulated to stimulate NH₄⁺ transport.

INTRODUCTION

One of the major tasks of the kidney is to maintain acid-base homeostasis in the body. The kidney regulates blood pH by reclaiming filtered acid or base equivalents or by releasing them into urine. The major acid-base component that the kidney regulates is ${\mathrm{HCO}}_3^{-}$. In the medullary thick ascending limb (MTAL), where 10-15% of filtered ${\mathrm{HCO}}_3^{-}$ is reclaimed, ${\mathrm{HCO}}_3^{-}$ absorption is initiated by H⁺ secretion into the lumen via the apical Na/H exchanger NHE3 (Good & Watts, III 1996), accompanied by ${\mathrm{HCO}}_3^{-}$ exit to the medullary interstitium. The basolateral ${\mathrm{HCO}}_3^{-}$ exit is probably mediated by the Cl/HCO₃ exchanger AE2 (Brosius et al. 1995;Sun 1998). However, Na/HCO₃ transporters (Kikeri et al. 1990;Vorum et al. 2000;Xu et al. 2003) and KCl cotransporter-mediated HCO₃ exit (Bourgeois et al. 2002) have also been reported.

The MTAL also handles NH₄⁺/NH₃ that serve as major non-bicarbonate buffer components in the kidney (for review, see ref. (Weiner & Hamm 2007). NH₄⁺, which is produced in the proximal tubules and secreted to the lumen, is absorbed via the apical Na/K/2Cl cotransporter NKCC2 (Kinne et al. 1986) and dissociates intracellularly into NH₃ and H⁺. NH₃ then moves to the medullary interstitium and re-associates with H⁺ to form NH₄⁺. This process provides a concentration gradient of NH₃ across the cells, which ultimately enables NH₃ to diffuse into the collecting ducts for urinary excretion or to the renal vein for recycling. The blood type Rh glycoprotein family proteins are mammalian ammonium transporters (Marini et al. 2000;Liu et al. 2001;Liu et al. 2000) and are found in the proximal tubules and collecting ducts, but not in the MTAL (Verlander et al. 2003).

${\mathrm{HCO}}_3^{-}$ and NH₄⁺ movements are associated with each other in many nephron segments including the MTAL (Good 1994;Wagner 2007;Kraut & Kurtz 2005). The physiological coupling between these two ion transport processes is particularly evident during cellular and systemic pH changes. The absorptive capacity for ${\mathrm{HCO}}_3^{-}$ and NH₄⁺ is increased in response to chronic metabolic acidosis, and this change typically involves up- or down-regulation of membrane proteins responsible for transporting acid/base equivalents (for review, see ref. (Good 1994;Wagner 2007). Among these proteins, NBCn1 is of particular interest. This transporter normally moves Na⁺ and ${\mathrm{HCO}}_3^{-}$ into the cell, but it is localized to the basolateral membrane of the MTAL. In vivo animal studies show a nearly 10-fold increase in NBCn1 protein expression in chronic metabolic acidosis induced by NH₄Cl load (Kwon et al. 2002;Odgaard et al. 2004). The upregulation of NBCn1 occurs probably to counter H⁺ overload caused by an increase in NH₄⁺ uptake. Consistent with this compensatory function, NBCn1 in neurons is upregulated under acidic conditions (Park et al. 2010).

In this study, we examined the effect of ${\mathrm{HCO}}_3^{-}$ transporters on NH₄⁺/NH₃ uptake in the MTAL cell line ST-1. This cell line is a non-polarized cell and exhibits many features characteristic for MTAL cells (Kone et al. 1995;Kone & Higham 1999). We measured intracellular accumulation of ¹⁴C-MA, and evaluated the effects of DIDS, as well as ion replacement, on ¹⁴C-MA accumulation. The effect of acidic pH on NH₄⁺/NH₃ uptake was evaluated by measuring 14C-MA accumulation after acidic incubation of the cells. NBCn1 was then expressed in Xenopus oocytes and examined for its ability to accumulate ¹⁴C-MA and affect pH changes mediated by NH₄⁺ transport. Our data suggest that NBCn1 plays dynamic roles in regulating ${\mathrm{HCO}}_3^{-}$ and NH₄⁺ transport in the MTAL cells.

MATERIALS AND METHODS

ST-1 cells

The ST-1 cells (provided by Bruce Kone, University of Florida) were cultured in Dulbecco's Modified Eagle Medium supplemented with 10% fetal bovine serum, 50 U/ml penicillin and 50 μg/ml streptomycin in 5% CO₂-equilibrated 37°C incubator (Kone & Higham 1999). For ¹⁴C-MA accumulation assay, cells were placed on 24-well plates at the density of 0.2–3.5 × 10⁵ cells/well and incubated for 1–4 days. For acidification experiments, cells were placed on 6-well plates at the density of 2.2–8.8 × 10⁴ cells/well until cells were > 90% confluent. The pH in the medium was adjusted by varying ${\mathrm{HCO}}_3^{-}$ concentration according to the Henderson-Hasselbalch equation with the solubility coefficient of 0.03 mmol CO₂/mmHg and P_CO2 of 35.65 mmHg.

Measurements of pH_i in ST-1 cells

Steady-state pH_i was determined according to the protocol (Cooper et al. 2009) with a slight modification. Briefly, cells grown on a coverslip (>60% confluent) were loaded with 6.5 μM of 2',7'-bis-(2-carboxyethyl)-5(6)-carboxyfluorescein acetoxymethyl ester (BCECF-AM) for 10 min and then mounted to a closed chamber RC-30 (Warner Instruments; Hamden, CT) affixed on the stage of a Zeiss Axiovert 135 inverted microscope. The microscope was equipped with a Lambda 10-2 filter wheel controller and a multi wavelength filter set (Sutter Instruments, Novato, CA). The dye was alternately excited with 490 nm and 440 nm light, and the emission light at 535 nm (i.e., I₄₉₀ and I₄₄₀) was captured. The ratio of I₄₉₀/I₄₄₀ was calculated after background subtraction. Dye calibration was done using nigericin. Steady-state pH_i in ${\mathrm{CO}}_2∕{\mathrm{HCO}}_3^{-}$ was calculated by determining linear least square analysis over a minimum of 20 seconds. Data were acquired using Nikon NIS Elements AR 3.0 (Nikon; Melville, NY).

Immunoblot

Cells were scraped in ice-cold homogenization buffer containing 300 mM mannitol, 5 mM HEPES, pH 7.2, 0.1 mg/ml phenylmethanesulphonyl fluoride and 1 × protease inhibitor cocktail I (Calbiochem; San Diego, CA, USA). Cells were homogenized with a 26G needle and centrifuged at 810 × g for 10 min at 4°C, and supernatants were ultracentrifuged at 100,000 × g for 30 min at 4°C. Membrane pellets were collected and dissolved in phosphate buffered saline (PBS), and protein concentration was determined using the Bradford reagents (Sigma-Aldrich). The equal amounts of protein samples were separated on a 7.5% SDS polyacrylamide gel and blotted to a polyvinylidene fluoride membrane. The blot was incubated in PBS containing 0.05% Tween 20 and 5% nonfat dry milk for 1 h and then treated with antibodies to rat NBCn1 (Cooper et al. 2009) and AE2 (Frische et al. 2004). The blot was washed with PBS containing 0.05% Tween 20 and then incubated with a horseradish peroxidase-conjugated secondary antibody (1:2500) for 1 h (Millipore; Billerica, MA). The blot was washed and visualized by ECL chemiluminescence (GE Healthcare; Chicago, IL). For β-actin, the blot was striped with 62.6 mM Tris-HCl (pH 6.7), 2% SDS, and 0.7% β-mercaptoethanol at 50°C for 5 min, and reprobed with the mouse β-actin antibody (Millipore). For quantitation, mean pixel intensities of immunoreactive signals were measured by positioning boxes around protein bands using ImageJ image analysis software (NIH; Bethesda, MD, USA). The intensity values for NBCn1 and AE2 were normalized to the intensity value for β-actin after background subtraction.

¹⁴C-MA accumulation

Cells grown on a 24-well plate were treated with the assay solution (mM; 140 NaCl, 1 KCl, 2 CaCl₂, 1 MgSO₄, 2.5 NaH₂PO₄, 5.5 glucose, 10 HEPES, pH 7.4, 1 ouabain and 0.2 bumetanide; 2 ml/well) for 20–30 min, and then incubated in the fresh ¹⁴C-MA assay solution containing 0.5 mM CH₃-NH₃Cl and 1 μCi/ml ¹⁴CH₃-NH₃Cl (ICN; Costa Mesa, CA, USA). For ${\mathrm{CO}}_2∕{\mathrm{HCO}}_3^{-}\text{-bufferred}$ solution, 33 mM NaHCO₃ replaced the equimolar concentration of NaCl and the solution was equilibrated with 5% CO₂ for 30 min. For ion replacement experiments, Na⁺ was substituted with N-methyl-D-glucamine (NMDG⁺) and Cl^– with gluconate. To block ${\mathrm{HCO}}_3^{-}$ transporters, 0.5 mM 4,4’-diisothiocyanato-2,2’-disulfonate stilbene DIDS (Sigma-Aldrich; St. Louis, MO) was included in the ¹⁴C-MA assay solution. Experiments were terminated at different time points by washing cells 4 times with ice-cold non-radioactive assay solution containing 1 mM MA. Cells were solubilized in 2% SDS/0.1 N NaOH, and the radioactivity was determined using a Packard Tricarb scintillation counter (PerkinElmer; Waltham, MA). Each time point was an independent measurement. The counts (count per minute; cpm) were divided by the mean specific activity of ¹⁴C-MA assay solution and presented as pmole per μg total protein. For ¹⁴C-MA accumulation experiments with oocytes, 3-5 oocytes were incubated with the ¹⁴C-MA assay solution in the presence/absence of 5% CO₂, 25 mM ${\mathrm{HCO}}_3^{-}$ (pH 7.4) without bumetanide and ouabain.

Expression of NBCn1 in Xenopus oocytes

Frogs were purchased from Xenopus Express (Brooksville, FL). A frog was anesthetized with fresh 0.1% 3-aminobenzoic acid ethyl ester (tricaine) in 5 mM HEPES (pH 7.5) for 20 min. The frog was then placed on ice and surgery was done to collect oocytes. After suture, the frog was placed in a recovery tank containing 0.1 M NaCl until the animal was fully recovered. Oocytes were washed with Ca²⁺-free solution containing (mM) 96 NaCl, 2 KCl, 1 MgCl₂, 10 HEPES (pH 7.4) five times for 20 min each, and then agitated with 2 mg/ml type IA collagenase (Sigma-Aldrich) twice for 20 min each. Oocytes without follicles (stage V-VI) were manually sorted and stored in OR3 medium (Chang et al. 2008) at 18°C overnight. For cRNA injection, the plasmid containing rat NBCn1-E (Cooper et al. 2005) was linearized with Nhe I and in vitro transcribed using the mMessage/mMachine transcription kit (Ambion; Austin, TX). Injection was done with 46 nl of either NBCn1-E cRNA (25 ng) or sterile water. Oocytes were maintained at 18°C for 3 days. Experiments requiring the use of frogs were conducted under the NIH guideline for research on animals, and the protocols were approved by the Institutional Animal Care and Use Committee at Emory University and University of Alabama at Birmingham.

Electrophysiology

An oocyte was impaled with current and voltage electrodes filled with 3M KCl. The electrode resistance was < 1.5 MΩ for both. The oocyte was superfused with the recording solution (mM; 80 NaCl, 20 NMDG-Cl, 2 KCl, 1 MgCl₂, 1.8 CaCl₂, 5 HEPES, pH 7.5) and clamped at -60 mV using the oocyte clamp OC-725C (Warner). Step-voltage commands from -120 mV to +40 mV (100 msec, 20 mV increments) were applied before and ~1 min after the switch to the recording solution containing 20 mM NH₄Cl. NMDG⁺ was replaced with NH₄⁺. Current and voltage signals were collected using a Digidata 1322 interface (Molecular Devices; Sunnyvale, CA, USA) and data were acquired and analyzed using pClamp 8 (Molecular Devices). Experiments were done at room temperature.

Statistical analysis

Data were reported as mean ± SE. For level of significance, a one-way ANOVA with the Bonferroni post-test was used to compare the ratio of NBCn1/β-actin expression at different time points. A two-way ANOVA with the Bonferroni post-test was used to compare ¹⁴C-MA accumulation at different time points. An unpaired, 1-tailed Student t-test was used to analyze 1) ¹⁴C-MA accumulation and pH_i after acidic incubation, 2) quantitation of immunoblot signals, and 3) conductance and zero-current voltage in oocytes. The p values less than 0.05 were considered significant. The n value referred to the number of independent experiments for ¹⁴C-MA accumulation assay and immunoblot, and the number of oocytes that were independently prepared from 4 experiments. Data were analyzed using Microsoft Excel and Origin 8.1 software (OriginLab Corporation; Northampton, MA).

RESULTS

¹⁴C-MA accumulation in ST-1 cells is stimulated by $HC{O}_3^{-}$ transporters

To examine whether ${\mathrm{HCO}}_3^{-}$ transporters affect NH₄⁺/NH₃ uptake, we treated cells with the ¹⁴C-MA assay solution (pH 7.4) and measured the radioisotope accumulated in cells. Both CH₃-NH₂ diffusion and carrier-mediated CH₃-NH₃⁺ transport affect ¹⁴C-MA accumulation in the non-polarized ST-1 cells. This complicates our study examining the effect of ${\mathrm{HCO}}_3^{-}$ transporters on NH₄⁺/NH₃ uptake in the cells because ${\mathrm{HCO}}_3^{-}$ transporters change pH_i, to which NH₄⁺ transporters such as NKCC2 (Gamba & Friedman 2009) and Na,K-ATPase (Kaplan 2002) are sensitive. For this and subsequent experiments, 0.2 mM bumetanide and 1 mM ouabain (Kone & Higham 1999) were included in the assay solution to minimize carrier-mediated CH₃-NH₃⁺ transport. Thus, the ¹⁴C-MA accumulation value in our experiments primarily reflects the amount of CH₃-NH₃⁺ that is converted from CH₃-NH₂ and trapped in the cell.

Fig. 1A shows the time course of mean ¹⁴C-MA accumulation in the absence of 5% CO₂, 33 mM ${\mathrm{HCO}}_3^{-}$ (pH 7.4). The accumulation increased over ~7 min and then reached a plateau, reflecting no net movement of CH₃-NH₃⁺/CH₃-NH₂ across the cell membrane. DIDS caused negligible effect under this condition (n = 3; p > 0.05; two-way ANOVA with Bonferroni post-test). In contrast, parallel experiments with sister plates showed that ¹⁴C-MA accumulation was greater in the presence of 5% CO₂, 33 mM ${\mathrm{HCO}}_3^{-}$ (Fig. 1B). The dotted line in the figure represents the time course of the accumulation measured in ${\mathrm{CO}}_2∕{\mathrm{HCO}}_3^{-}\text{-free}$ solution. The ${\mathrm{CO}}_2∕{\mathrm{HCO}}_3^{-}\text{-dependent}$ ¹⁴C-MA accumulation was markedly reduced by 0.5 mM DIDS (n = 3; p < 0.05; two-way ANOVA with Bonferroni post-test). The reduction was particularly substantial over 12 min and then less effective afterward. Thus, at 22 min, ~37% of the accumulation was inhibited by DIDS.

Fig. 1.

¹⁴C-MA accumulation in cells. A) ¹⁴C-MA accumulation in ${\mathrm{CO}}_2∕{\mathrm{HCO}}_3^{-}\text{-free}$ solution (pH 7.4). Cells grown on a 24-well plate were treated with ${\mathrm{CO}}_2∕{\mathrm{HCO}}_3^{-}\text{-free}$ ¹⁴C-MA assay solution. Cells were then rapidly rinsed with ice-cold wash solution containing unlabeled MA at different time points. The ¹⁴C-MA assay solution in this and subsequent experiments with ST-1 cells contained 0.2 mM bumetanide and 1 mM ouabain to minimize carrier-mediated CH₃-NH₃⁺ transport. Experiments were done in the presence or absence of 0.5 mM 4,4’-diisothiocyanatostilbene-2,2’-disulfonate (DIDS). The accumulation was presented as pmole/μg total protein. Data were averaged from 3 experiments. B) ¹⁴C-MA accumulation in ${\mathrm{CO}}_2∕{\mathrm{HCO}}_3^{-}$ solution. The experimental procedure was identical to that in A except that the assay solution contained 5% CO₂, 33 mM ${\mathrm{HCO}}_3^{-}$ (pH 7.4). The dotted line was derived from the line in A. Data were averaged from 3 experiments.

The above results indicate that DIDS-sensitive ${\mathrm{HCO}}_3^{-}$ transporters assist in providing H⁺ and influence intracellular NH₄⁺ trapping. Such a process is expected in an acid-loading transporter that removes ${\mathrm{HCO}}_3^{-}$ from the cell. The DIDS-sensitive ${\mathrm{HCO}}_3^{-}$ extrusion in the MTAL is governed by the Cl/HCO₃ exchanger (Quentin et al. 2004). To test whether the Cl/HCO₃ exchanger is involved in the process of NH₄⁺ trapping in our experiments, we replaced Cl^– in the assay solution with gluconate and measured ¹⁴C-MA accumulation over 10-30 min after treatment, during which the accumulation reached a plateau in ${\mathrm{CO}}_2∕{\mathrm{HCO}}_3^{-}\text{-free}$ solution. Replacing Cl^– progressively reduced the accumulation over time compared to controls (Fig 2A), consistent with the idea that the absence of external Cl^– decreases ${\mathrm{HCO}}_3^{-}$ exit and the concomitant H⁺ needed to trap ¹⁴C-MA. Nonetheless, adding 0.5 mM DIDS to the Cl^–-replaced assay solution caused a more profound decrease, comparable to the inhibition by DIDS in Cl^–-containing assay solution (Fig. 1B). Thus, Cl^– removal partially inhibited ${\mathrm{CO}}_2∕{\mathrm{HCO}}_3^{-}\text{-dependent}$ ¹⁴C-MA accumulation. In the MTAL, DIDS-sensitive ${\mathrm{HCO}}_3^{-}$ exit can occur via the KCl cotransporter (Bourgeois et al. 2002) in addition to the Cl/HCO₃ exchanger. We raised the K⁺ concentration by 10 fold (from 2 mM to 20 mM) to test whether the KCl transporter contributes to ${\mathrm{CO}}_2∕{\mathrm{HCO}}_3^{-}\text{-dependent}$ ¹⁴C-MA accumulation. Raising the K⁺ concentration produced negligible effect (p > 0.05, n = 4; Fig. 2B), implying that the KCl cotransporter is unlikely involved. In parallel experiments with sister plates, cells were treated with the assay solution that contained NMDG⁺ instead of Na⁺. Replacing Na⁺ caused a small reduction (Fig. 2C). Adding the NHE blocker dimethylamiloride (0.1 mM) had negligible effect on this reduction (data not shown).

Fig. 2.

Effects of Cl^– & Na⁺ replacement and K⁺ concentration on ¹⁴C-MA accumulation. All assay solutions contained 5% CO₂, 33 mM ${\mathrm{HCO}}_3^{-}$ (pH 7.4). A) Effect of Cl^– replacement. Cl^– in the ¹⁴C-MA assay solution was replaced by gluconate in the absence or presence of 0.5 mM DIDS (n = 4 for each). B) Effect of raising K⁺ concentration. The K⁺ concentration in the assay solution was raised from 2 mM to 20 mM (n = 4). C) Effect of Na⁺ replacement. Na⁺ in the assay solution was replaced by NMDG⁺ (n = 4).

Acid-loading $HC{O}_3^{-}$ extrusion is found in ST-1 cells

We monitored pH_i changes from an alkali load using the pH-sensitive fluorescence dye BCECF. Upon the application of 20 mM NH₄Cl in ${\mathrm{CO}}_2∕{\mathrm{HCO}}_3^{-}\text{-free}$ solution (Fig. 3A), pH_i was rapidly increased and then nearly reached plateau (n = 5). Lack of intracellular acidification reflects inhibition of carrier-mediated CH₃-NH₃⁺ transport by bumetanide and ouabain in our experiments. The recovery was slightly enhanced by DIDS, the nature of which is unclear. Thus, DIDS had negligible effect on pH_i alone. In the presence of 5% CO₂, 33 mM ${\mathrm{HCO}}_3^{-}$ (Fig. 3B), however, the pH_i recovery toward the resting pH was significantly faster (dpHi/dt of -27.8 ± × 10^-4 pH/sec; n =5), indicating acid-loading ${\mathrm{HCO}}_3^{-}$ transport in the cells. This acid-loading process was substantially reduced by DIDS (-6.9 × 10^-4; p < 0.05). Fig. 3C summarizes the pH_i recovery rates from an alkali load in the absence and presence of ${\mathrm{CO}}_2∕{\mathrm{HCO}}_3^{-}$. The dpH_i/dt in the presence of ${\mathrm{CO}}_2∕{\mathrm{HCO}}_3^{-}$ was higher and inhibited by DIDS (75%).

Fig. 3.

${\mathrm{CO}}_2∕{\mathrm{HCO}}_3^{-}\text{-induced}$ stimulation of the pH_i recovery from an alkali load. A) pH_i recovery from an alkali load in the absence of ${\mathrm{CO}}_2∕{\mathrm{HCO}}_3^{-}$. Cells were treated with 20 mM NH₄Cl, and 0.5 mM DIDS was applied during the pH_i recovery from an alkali load (n =5). B) pH_i recovery from an alkali load in the presence of 5%CO₂, 33 mM ${\mathrm{HCO}}_3^{-}$. Cells were subjected to the protocol as described in A except that the experiment was done in the presence of ${\mathrm{CO}}_2∕{\mathrm{HCO}}_3^{-}$ (n =5). C) Summary of the pH_i recovery rate. Data were obtained from the experiments in A and B.

DIDS-sensitive ¹⁴C-MA accumulation is abolished at low pH_i

Fig. 4A shows ¹⁴C-MA accumulation rates measured in the absence and presence of 5% CO₂, 33 mM ${\mathrm{HCO}}_3^{-}$ (pH 7.4). The measurements without DIDS (open boxes in the figure) revealed that the accumulation rate was higher in the presence of ${\mathrm{CO}}_2∕{\mathrm{HCO}}_3^{-}$ (2.5 ± 0.4 pmole/μg/min in HEPES versus 6.1 ± 0.4 pmole/μg/min in ${\mathrm{CO}}_2∕{\mathrm{HCO}}_3^{-}$, n = 3 for each; the unit is pmole of ¹⁴C-MA per μg of total protein per min), corresponding to a 2.4-fold increase (p = 0.001; a in the figure). Adding 0.5 mM DIDS to the assay solution had negligible effect in the absence of ${\mathrm{CO}}_2∕{\mathrm{HCO}}_3^{-}$, but reduced the accumulation (4.5 ± 0.1 pmole/μg/min, n = 3, p = 0.002; c in the figure). The remaining value after subtracting the value in HEPES represents the DIDS-insensitive accumulation (p = 0.01; d in the figure).

Fig. 4.

¹⁴C-MA accumulation in cells after acidic incubation. A) Effect of DIDS on ¹⁴C-MA accumulation. Cells were treated with ¹⁴C-MA assay solution in the absence and presence of 5% CO₂, 33 mM ${\mathrm{HCO}}_3^{-}$ (pH 7.4) for 30 min. The concentration of DIDS was 0.5 mM. The p values for comparison (a–d) are described in the text. Data were averaged from 3 experiments. B) Effect of acidic pH_i on ¹⁴C-MA accumulation. Cells were incubated in the medium of pH 7.4 or 6.8 for 24 hours. Experiments were done by treating cells with ${\mathrm{CO}}_2∕{\mathrm{HCO}}_3^{-}\text{-buffered}$ ¹⁴C-MA assay solution (pH 7.4) for 20 min. Data were averaged from 5 experiments. C) Steady-state pH_i. Cells were incubated in the medium of pH 7.4 or 6.8 for 24 h, and their pH_i was measured in ${\mathrm{CO}}_2∕{\mathrm{HCO}}_3^{-}$ at the corresponding pH (n = 4 for each). The measurements were done using the ratiometric pH-sensitive dye BCECF-AM.

We then incubated cells at pH 7.4 and 6.8 (adjusted with NaHCO₃ in 5% CO₂) for 24 h and measured ¹⁴C-MA accumulation in the presence of 5% CO₂, 33 mM ${\mathrm{HCO}}_3^{-}$ (pH 7.4) (Fig. 4B). Acidic Incubation at pH 6.8-6.9 has been used for the in vitro study of metabolic acidosis in other systems (Tsuruoka et al. 2006;O'Hayre et al. 2006). Adding 0.5 mM DIDS to the assay solution reduced ¹⁴C-MA accumulation in control cells (p = 0.007; n = 5; a in the figure). However, DIDS showed negligible inhibition in cells incubated at pH 6.8. The ¹⁴C-MA accumulation was similar in the two groups of cells treated and untreated with DIDS (p > 0.05; b in the figure), indicating that DIDS-sensitive ${\mathrm{HCO}}_3^{-}$ transport was inactivated after acidic incubation. This inactivation caused a substantial decrease in the accumulation.

Fig. 4C shows mean steady-state pH_i in the presence of ${\mathrm{CO}}_2∕{\mathrm{HCO}}_3^{-}$ determined using the ratiometric pH-sensitive dye BCECF-AM. The steady-state pH_i in cells incubated at pH 6.8 was 6.96 ± 0.03 (n = 4), significantly lower than 7.28 ± 0.04 (n = 4) for control cells (p = 0.001). Thus, acidic incubation lowered steady-state pH_i. Incubating cells at pH 6.8 for 24 h caused 20% cell death compared to 8% for control, determined by Trypan Blue staining (data not shown). The unhealthy or dead cells are swollen, exhibit blebbing, and morphologically distinguishable from health cells. We used healthy cells for pH measurements.

NBCn1 in the MTAL cells is upregulated by acidic incubation

The above data show that the ¹⁴C-MA accumulation after acidic incubation is lower than that in control incubation, when compared in the absence of DIDS. This is probably due to inactivation of DIDS-insensitive ${\mathrm{HCO}}_3^{-}$ extruders at low pH_i, but it is also possible that acid-extruding ${\mathrm{HCO}}_3^{-}$ loaders are stimulated under acidic conditions and reduce the conversion of CH₃-NH₃ to CH₃-NH₃⁺. To test whether NBCn1 protein expression is affected, we incubated cells at pH 6.8 for 24 h and performed immunoblot using the anti-NBCn1 antibody (Cooper et al. 2009). Control cells were incubated at pH 7.4 for 24 h.

Fig. 5A shows a representative immunoblot analysis of membrane preparations from cells at 0 and 24 h after acidic incubation. An immunoreactive band of 135 kDa was detected. A band of 250 kDa, which is a multimeric form of the protein as shown in neurons (Cooper et al. 2009), was additionally recognized. The detection of this higher molecular weight band varies depending upon experiments (Han Soo Yang, unpublished observation). NBCn1 expression was increased after acidic incubation. Analyzed by quantitative measurements of NBCn1 relative to β-actin (Fig. 5B), the increase was 48% (p = 0.04; n= 3). The upregulation was progressively larger at longer incubation periods (Fig. 5C and D) and thus, at 72 h, the NBCn1 expression level after acidic incubation was nearly 3 fold higher than controls. We also performed immunoblot of cells that were incubated in the medium containing 10 mM NH₄Cl (pH 7.4) for 24 h and found a marked upregulation of NBCn1 after NH₄Cl treatment (Fig. 5E and F). This result indicates that the pH in the media is not responsible for upregulating the transporter.

Fig. 5.

Effect of acidic incubation on NBCn1 expression. A) Immunoblot for NBCn1. Cells were incubated in the medium of pH 7.4 versus 6.8 for 0 or 24 h. Plasma membranes were isolated and subjected to immunoblot with the anti-NBCn1 antibody. The blots were striped and reprobed with the β-actin antibody. One of three experiments is shown. B) Quantitative measurements of NBCn1. The pixel intensity of NBCn1 was measured and normalized to β-actin (n = 3). The p value less than 0.05 was considered significant. C) Immunoblot for NBCn1 at normal and acidic incubation for 24, 48, and 72 hours. One of three experiments is shown. D) Quantitative measurements of NBCn1. The value of NBCn1/β-actin at pH 6.8 was presented relative to the value at pH 7.4 at each time point. The data were analyzed by one-way ANOVA with Bonferroni post-test. *p value of less than 0.05. E) NBCn1 expression in cells treated with NH₄Cl. Cells were incubated in the medium (pH 7.4) containing 0 or 10 mM NH₄Cl for 24 h. NH₄Cl replaced NaCl. One of three experiments is shown. F) Quantitative measurements of NBCn1.

In other experiments, the blot was probed with the anti-AE2 antibody (Frische et al. 2004) to determine whether acidic incubation affects the Cl/HCO₃ exchanger. An immunoreactive band of 110 kDa was recognized but remained unaffected after acidic incubation (Fig. 6A). The molecular weight of AE2 in ST-1 cells was smaller than ~150 kDa in the kidney (Frische et al. 2004). Quantitative measurements of AE2 relative to β-actin revealed negligible difference between control and acidic incubation (p > 0.05; n= 4) (Fig. 6B). We also performed immunoblot with antibodies to AE1 and AE3 but did not find immunoreactive bands (data not shown).

Fig. 6.

Effect of acidic incubation on AE2 expression. A) Immunoblot for AE2. The experimental procedure was similar to that in Fig. 5 except the blot was probed with the anti-AE2 antibody. One of four experiments is shown. B) Quantitative measurements of AE2. The pixel intensity of AE2 was measured and normalized to β-actin.

NBCn1 expressed in Xenopus oocytes increases ¹⁴C-MA accumulation

The above results led us to postulate that the ${\mathrm{HCO}}_3^{-}$ loader NBCn1 might contribute to NH₄⁺/NH₃ uptake particularly under acidic conditions, in which the transporter is upregulated. To test this hypothesis, we expressed the rat renal NBCn1-E splice variant (Yang et al. 2009) in Xenopus oocytes and measured ¹⁴C-MA accumulation. Oocyte membranes are very impermeable to NH₃ (Burckhardt & Frömter 1992;Nakhoul et al. 2001), indicating that oocytes can preferably trap NH₃, opposite to our ST-1 cell system. For ¹⁴C-MA accumulation experiments with oocytes, therefore, bumetanide and ouabain were not added to the assay solution. In ${\mathrm{CO}}_2∕{\mathrm{HCO}}_3^{-}\text{-free}$ ND96 solution (Fig. 7A), oocytes expressing NBCn1-E and control oocytes similarly accumulated ¹⁴C-MA (p > 0.05, n = 16 oocytes for each). However, in a solution buffered with 5% CO₂, 25 mM ${\mathrm{HCO}}_3^{-}$ (Fig. 7B), oocytes expressing NBCn1-E showed a significant increase in ¹⁴C-MA accumulation (p < 0.05, n = 16 for each; two-way ANOVA with Bonferroni post-test). The difference between the two groups of oocytes was already substantial at 10 min after exposure to ${\mathrm{CO}}_2∕{\mathrm{HCO}}_3^{-}$ and continued afterward. Fig. 7C shows ${\mathrm{CO}}_2∕{\mathrm{HCO}}_3^{-}\text{-dependent}$ ¹⁴C-MA accumulation calculated by subtracting values in ${\mathrm{CO}}_2∕{\mathrm{HCO}}_3^{-}\text{-free}$ solution from values in ${\mathrm{CO}}_2∕{\mathrm{HCO}}_3^{-}$ solution. A profound increase was produced in oocytes expressing NBCn1-E. Na⁺ replacement abolished the difference in the accumulation (Fig. 7D).

Fig. 7.

Time course of ¹⁴C-MA accumulation in Xenopus oocytes expressing NBCn1. A & B) ¹⁴C-MA accumulation in oocytes injected with water or NBCn1-E cRNA. Oocytes were incubated with the ¹⁴C-MA assay solution (no bumetanide and ouabain) in the absence (A) or presence (B) of ${\mathrm{CO}}_2∕{\mathrm{HCO}}_3^{-}$ (pH 7.4), and then rapidly washed with ice-cold unlabeled MA solution at different time points. The accumulation was expressed as nmole/oocyte. Data were averaged from 4 independent experiments with total 16 oocytes per group. C) ${\mathrm{CO}}_2∕{\mathrm{HCO}}_3^{-}\text{-dependent}$ ¹⁴C-MA accumulation. The accumulation was calculated by subtracting values in ${\mathrm{CO}}_2∕{\mathrm{HCO}}_3^{-}\text{-free}$ ND96 solution from values in ${\mathrm{CO}}_2∕{\mathrm{HCO}}_3^{-}\text{-containing}$ solution in A & B. D) Effect of Na⁺ replacement on ¹⁴C-MA accumulation. Na⁺ was replaced with NMDG⁺ (n = 3 for each).

We then performed two-electrode voltage clamp to test whether NBCn1-E produces currents by NH₄⁺. Determined before and 1 min after applying 20 mM NH₄Cl, current-voltage relationships showed negligible change (Fig. 8A & B). The slope conductances were similar before and after NH₄Cl application (p > 0.05 for both control and NBCn1-E; n = 6 for control and n = 8 for NBCn1-E) (Fig. 8C). Oocytes expressing NBCn1-E had more positive zero-current voltage and higher slope conductance due to the ionic conductance intrinsic to NBCn1 (Cooper et al. 2005).

Fig. 8.

Two-electrode voltage clamp of oocytes. A & B) Current-voltage relationships for currents before and after NH₄Cl application. Oocytes were clamped at -60 mV and voltage commands from -120 to +40 mV were applied. Recordings were made in ${\mathrm{CO}}_2∕{{\mathrm{HCO}}^{-}}_3\text{-free}$ ND96 solution and 1 min after switching to a solution containing 20 mM NH₄Cl. Data were averaged from 6 controls and 8 NBCn1-E-expressing oocytes. C) Slope conductance. The conductance was computed near the zero-current potential in A & B.

DISCUSSION

The major findings of the present study are the following: 1) ${\mathrm{CO}}_2∕{\mathrm{HCO}}_3^{-}$ enhances ¹⁴C-MA accumulation in the non-polarized MTAL cells and this effect is inhibited by DIDS; 2) ${\mathrm{CO}}_2∕{\mathrm{HCO}}_3^{-}\text{-dependent}$ ¹⁴C-MA accumulation is affected by replacing Cl^–; 3) At low pH_i, ${\mathrm{CO}}_2∕{\mathrm{HCO}}_3^{-}\text{-dependent}$ ¹⁴C-MA accumulation is not inhibited by DIDS; 4) NBCn1, but not AE2, in ST-1 cells is upregulated under acidic conditions; and 5) NBCn1 in Xenopus oocytes stimulates ${\mathrm{CO}}_2∕{\mathrm{HCO}}_3^{-}\text{-dependent}$ ¹⁴C-MA transport. These findings are important for understanding the effect of ${\mathrm{HCO}}_3^{-}$ transporters on NH₄⁺/NH₃ uptake in the MTAL and provide evidence that cells regulate NH₄⁺/NH₃ uptake by altering the expression and activity of NBCn1 and AE2 in response to acid-base disturbance. A cell model for the effect of ${\mathrm{HCO}}_3^{-}$ transporters on NH₄⁺/NH₃ uptake in ST-1 cells is described in the Supplementary Data.

We used radiolabeled MA to analyze the effect of ${\mathrm{HCO}}_3^{-}$ transporters on NH₄⁺/NH₃ uptake. MA (pK of 10.3) has been extensively used as an NH₄⁺/NH₃ surrogate in many studies (Soupene et al. 2002;Verlander et al. 2003;Westhoff et al. 2002). When a cell is exposed to CH₃-NH₃⁺/CH₃-NH₂, membrane-permeable CH₃-NH₂ rapidly enters the cell and associates with H⁺ to form CH₃-NH₃⁺. The reaction then reaches a plateau as CH₃-NH₂ becomes equilibrated inside and outside the cell. Under this plateau phase, charged CH₃-NH₃⁺ is slowly transported into the cytosol via carrier proteins and then dissociates into CH₃-NH₂ and H⁺. If carrier-mediated CH₃-NH₃⁺ entry is inhibited, the total intracellular CH₃-NH₃⁺/CH₃-NH₂ (i.e., MA) is primarily determined by the amount of intracellular H⁺ supplied for the conversion of CH₃-NH₂ to CH₃-NH₃⁺. A high level of intracellular H⁺ supply increases the conversion and accumulates more ¹⁴C-MA, whereas a low level of intracellular H⁺ supply decreases it. Therefore, the magnitude of ¹⁴C-MA accumulation in our experiments reflects the activity of acid/base transporters in the cell.

In analyzing the effect of ${\mathrm{HCO}}_3^{-}$ transporters on NH₄⁺/NH₃ uptake, we found that DIDS reduces ¹⁴C-MA accumulation (Fig. 1, Fig. 2, and Fig. 4). Additional experiments of Cl^– replacement show the Cl/HCO₃ exchanger influencing ¹⁴C-MA accumulation. In ${\mathrm{CO}}_2∕{\mathrm{HCO}}_3^{-}$ solution, CO₂ enters the cell and produces H⁺ and ${\mathrm{HCO}}_3^{-}$. By moving ${\mathrm{HCO}}_3^{-}$ out of the cell, the Cl/HCO₃ exchanger concomitantly provides intracellular H⁺ available for the conversion of CH₃-NH₂ to CH₃-NH₃⁺. Nonetheless, our data suggest that the Cl/HCO₃ exchanger would not be the major ${\mathrm{HCO}}_3^{-}$ extruder affecting NH₄⁺/NH₃ uptake in ST-1 cells because Cl^– replacement partially inhibits ¹⁴C-MA accumulation. DIDS in the Cl^– replacement produces more severe inhibition than Cl^– replacement alone (Fig. 2A), implying that there is an additional DIDS-sensitive ${\mathrm{HCO}}_3^{-}\text{-extruding}$ process. This ${\mathrm{HCO}}_3^{-}$ extrusion does not require K⁺ as raising the K⁺ concentration by 10 fold has no effect on ¹⁴C-MA accumulation (Fig. 2B and C). Similarly, it also dose not require Na⁺ because replacing Na⁺ has only a small effect. The molecular nature of this ${\mathrm{HCO}}_3^{-}$ extrusion is unclear.

Under acidic conditions, the activity of Cl/HCO₃ exchange appears to be minimal because DIDS has no significant effect on ¹⁴C-MA accumulation after acidic incubation (Fig. 3). Stewart et al. (Stewart et al. 2001) reported that AE2 is inactivated at low pH_i. AE2-mediated ³⁶Cl^– uptake in Xenopus oocytes is almost abolished when the oocyte pH_i is at 6.88. Furthermore, AE2 expression remains unaffected after acidic incubation (Fig. 6). This lack of AE2 upregulation is comparable to the previous report (Quentin et al. 2004) that AE2 in the MTAL of the kidney is not upregulated by chronic metabolic acidosis. Together with the lack of DIDS inhibition and the negligible change in AE2 expression, our data support the idea of Cl/HCO₃ exchange being inactivated at low pH_i.

In contrast to AE2, that NBCn1 in ST-1 cells is upregulated after acidic incubation (Fig. 5). This is in good agreement with the reports (Kwon et al. 2002;Odgaard et al. 2004) that NBCn1 expression and activity in rat MTAL are enhanced during chronic metabolic acidosis induced by NH₄Cl feeding. It is also comparable to our recent report (Cooper et al. 2009) that NBCn1 in primary cultures of rat hippocampal neurons is upregulated after incubation at pH 6.2-6.5 for 1 hour. Others have reported that NBCn1 in rat MTAL is unaffected or down-regulated during chronic metabolic acidosis induced by ureter obstruction (Wang et al. 2008) and calcineurin inhibition (Mohebbi et al. 2009). The reason for the disparity among several reports remains unclear, but we think that the severity and duration of acidosis may determine the onset of NBCn1 upregulation. NBCn1 normally moves ${\mathrm{HCO}}_3^{-}$ into the cell and extrudes acids, implying that its upregulation compensates excessive H⁺ load caused by acidification (Kwon et al. 2002;Odgaard et al. 2004). Consistent with this, we observe NBCn1 upregulation after incubating ST-1 cells with NH₄Cl (Fig. 5E), further supporting the importance of pH_i for NBC upregulation.

In Xenopus oocytes, NBCn1 increases ${\mathrm{CO}}_2∕{\mathrm{HCO}}_3^{-}\text{-dependent}$ ¹⁴C-MA accumulation (Fig. 7). Oocytes membranes are less permeable to NH₃ than to NH₄⁺ (Burckhardt & Frömter 1992;Nakhoul et al. 2001), similar to the apical membrane of the MTAL (Kikeri et al. 1989). Oocytes expressing NBCn1 show a relatively slow rise in pH_i upon NH₄Cl application compared to control oocytes (see Supplementary Data). Thus, the increased accumulation of ¹⁴C-MA in NBCn1-expressing oocytes reflects that carrier-mediated NH₄⁺ transport is increased in these oocytes. It has been documented that NH₄Cl induces a slow acidification with depolarization of oocyte membranes. However, we found no measurable change in V_m during NH₄Cl application. The reason for this disparity is unclear although we note that NH₄Cl-induced voltage changes in oocytes have been measured in ${\mathrm{CO}}_2∕{\mathrm{HCO}}_3^{-}\text{-free}$ solutions. It is also interesting to note that NH₄⁺ transport in oocytes may be inhibited by acidic pH_i (Boldt et al. 2003). Regardless, our data provide evidence that NBCn1 enhances NH₄⁺ transport as determined by ¹⁴C-MA accumulation. This might be important for regulating NH₄⁺ transport in the MTAL particularly during chronic metabolic acidosis.

In summary, our data provide in vitro evidence for NBCn1 upregulation in acidic pH environments. The upregulation may provide a constant supply of ${\mathrm{HCO}}_3^{-}$ to the cell and enhance NH₄⁺/NH₃ uptake by buffering intracellular H⁺ load. It will be interesting to examine a possible NH₄⁺/NH₃ movement via NBCn1 in future experiments.