NBCn1 Increases NH₄⁺ Reabsorption Across Thick Ascending Limbs, the Capacity for Urinary NH₄⁺ Excretion, and Early Recovery from Metabolic Acidosis

Significance Statement

Recovery from metabolic acidosis necessitates increased renal net acid excretion through urinary elimination of NH₄⁺. Renal thick ascending limbs (TALs) contribute to a medullary shortcut, where NH₄⁺ originating from proximal tubules is ultimately secreted in collecting ducts. NH₄⁺ transfer across TALs requires a basolateral exit pathway for H⁺ to avoid intracellular accumulation. Experiments with knockout mice show that the Na⁺/HCO₃⁻ cotransporter NBCn1 mediates basolateral HCO₃⁻ uptake and increases NH₄⁺ reabsorption in TAL, amplifies the corticomedullary NH₄⁺ gradient, elevates the capacity for urinary NH₄⁺ excretion, and accelerates recovery of arterial blood pH and [HCO₃⁻] during metabolic acidosis. NBCn1 is crucial for acid-base handling in TALs, and for early renal compensation of systemic acid-base disturbances.

Abstract

Background

The electroneutral Na⁺/HCO₃⁻ cotransporter NBCn1 (Slc4a7) is expressed in basolateral membranes of renal medullary thick ascending limbs (mTALs). However, direct evidence that NBCn1 contributes to acid-base handling in mTALs, urinary net acid excretion, and systemic acid-base homeostasis has been lacking.

Methods

Metabolic acidosis was induced in wild-type and NBCn1 knockout mice. Fluorescence-based intracellular pH recordings were performed and NH₄⁺ transport measured in isolated perfused mTALs. Quantitative RT-PCR and immunoblotting were used to evaluate NBCn1 expression. Tissue [NH₄⁺] was measured in renal biopsies, NH₄⁺ excretion and titratable acid quantified in spot urine, and arterial blood gasses evaluated in normoventilated mice.

Results

Basolateral Na⁺/HCO₃⁻ cotransport activity was similar in isolated perfused mTALs from wild-type and NBCn1 knockout mice under control conditions. During metabolic acidosis, basolateral Na⁺/HCO₃⁻ cotransport activity increased four-fold in mTALs from wild-type mice, but remained unchanged in mTALs from NBCn1 knockout mice. Correspondingly, NBCn1 protein expression in wild-type mice increased ten-fold in the inner stripe of renal outer medulla during metabolic acidosis. During systemic acid loading, knockout of NBCn1 inhibited the net NH₄⁺ reabsorption across mTALs by approximately 60%, abolished the renal corticomedullary NH₄⁺ gradient, reduced the capacity for urinary NH₄⁺ excretion by approximately 50%, and delayed recovery of arterial blood pH and standard [HCO₃⁻] from their initial decline.

Conclusions

During metabolic acidosis, NBCn1 is required for the upregulated basolateral HCO₃⁻ uptake and transepithelial NH₄⁺ reabsorption in mTALs, renal medullary NH₄⁺ accumulation, urinary NH₄⁺ excretion, and early recovery of arterial blood pH and standard [HCO₃⁻]. These findings support that NBCn1 facilitates urinary net acid excretion by neutralizing intracellular H⁺ released during NH₄⁺ reabsorption across mTALs.

Protein-rich diets challenge systemic acid-base homeostasis as they necessitate renal excretion of nonvolatile acid derived from metabolism of sulfur- and phosphorus-containing amino acids.¹ Additional acid loading occurs under pathophysiologic conditions, for instance, because of reduced respiratory function or metabolic derangements.

Renal proximal tubules reabsorb most filtered HCO₃⁻, and renal elimination of elevated acid loads requires de novo synthesis of HCO₃⁻ coupled with net urinary acid excretion. During systemic acidosis, the kidneys secrete NH₄⁺/NH₃ in proximal tubules—increasing the urinary buffering power—and acidify the preurine, particularly in distal tubular segments with low paracellular permeability for acid-base equivalents.² En route to urinary excretion, NH₄⁺ follows a medullary shortcut: it is absorbed across medullary thick ascending limbs (mTALs), and subsequently secreted by diffusion trapping in luminally acidified collecting ducts.^2,3

Transcellular vectorial transport of NH₄⁺ across mTALs involves luminal uptake via the Na⁺/K⁺/2Cl⁻ cotransporter NKCC2 and renal outer medullary K⁺ channels, where NH₄⁺ substitutes for K⁺ (Figure 1).⁴ The basolateral exit pathway, however, remains controversial. NH₃ traverses the lipid bilayer either directly, owing to its hydrophobicity, or through gas channels such as rhesus proteins.^5,6 As cellular efflux of NH₃ leads to net removal of base from the cytosol, intracellular acidification would soon be limiting if mechanisms were not in place to neutralize H⁺ liberated during intracellular dissociation of NH₄⁺ (Figure 1). It has been hypothesized that basolateral Na⁺/HCO₃⁻ cotransport via NBCn1 and Na⁺/H⁺ exchange via NHE1 and NHE4 (Figure 1) mediate the required net acid extrusion.^7,8

Figure 1.

Acid-base transport across mTALs and collecting ducts act in concert to facilitate NH₄⁺ excretion in the urine. The schematic model shows the proposed acid-base handling mechanisms in mTALs, and their contribution to NH₄⁺ secretion in collecting ducts. In this study, we show that NBCn1 protein expression and function are dramatically upregulated during metabolic acidosis; under these conditions, NBCn1 contributes to urinary NH₄⁺ excretion and early recovery of arterial blood pH and standard [HCO₃⁻].

The Slc4a7 promoter, which controls the transcriptional activity of NBCn1,^9,10 is active in mTALs,¹¹ where the NBCn1 protein immunolocalizes to basolateral cell membranes.¹² NBCn1 protein expression is upregulated in the inner stripe of the renal outer medulla of rats, showing enhanced urinary NH₄⁺ excretion in response to systemic NH₄Cl loading or K⁺ depletion.^8,13,14 Na⁺/HCO₃⁻ cotransport activity is also elevated under these conditions, but the functional contribution from NBCn1, and its consequences for urinary NH₄⁺ excretion and systemic acid-base balance, have not previously been established because of a lack of specific pharmacologic tools.^15–17 NBCn1 is usually reported as insensitive to 4,4'-diisothiocyano-2,2'-stilbenedisulfonic acid (DIDS), when expressed in epithelia.¹⁵ Conflicting reports show that DIDS inhibits the Na⁺/HCO₃⁻ cotransport in mTALs of kidney slices,⁸ but not in isolated perfused mTALs.¹⁴

Here, we used NBCn1 knockout (KO) mice to test the hypothesis that NBCn1 facilitates basolateral HCO₃⁻ uptake in mTALs, contributes to the renal medullary NH₄⁺ shortcut that promotes urinary NH₄⁺ excretion, and accelerates renal compensation of systemic metabolic acidosis.

Methods

We studied isolated renal thick ascending limbs (TALs) and blood acid-base parameters in male Slc4a7^Gt(40G1)Cmhd mice with genetically disrupted NBCn1 expression.¹⁸ Wild-type (WT) mice and age-matched NBCn1 KO mice (backcrossed into the C57BL/6j genetic background for at least ten generations) were bred in the animal facility at Department of Biomedicine, Aarhus University, and fed a standard laboratory diet (#1310; Altromin). The mice had free access to regular tap water (control) or received tap water containing 196 mM NH₄Cl (metabolic acidosis) for up to 4 days before experiments.⁸ The Danish Animal Experiments Inspectorate approved all experimental procedures (approval number 2016–15–0201–00982) that adhered to the National Institutes of Health Guide for the Care and Use of Laboratory Animals.

Perfusion of Isolated Tubules

We isolated renal mTALs from 24- to 38-day-old mice, euthanized by cervical dislocation. The kidneys were excised, transferred to a cooled (approximately 4°C) dissection chamber, and sliced transversely after the capsule was removed. Then, mTALs were dissected manually, moved to a perfusion chamber mounted on an Axiovert 100 TV inverted microscope (Zeiss), and perfused through glass pipettes, as previously described in detail.^19,20

Intracellular pH Measurements

Isolated perfused mTALs were loaded for 10 minutes at room temperature, with 20 µM 2',7'-bis-(2-carboxyethyl)-5-(and-6)-carboxyfluorescein acetoxymethyl ester (BCECF-AM). Using a VisiChrome High Speed Polychromatic Illumination System controlled with VisiVIEW 2.0.1 software (Visitron Systems GmbH), the isolated perfused tubules were alternately excited at 490 nm for 25 ms and 436 nm for 50 ms. BCECF fluorescence ratios, on the basis of emission light collected at 520–560 nm with a Pursuit 1.4 MP Monochrome camera (Spot Imaging), were converted to pH values using the high [K⁺] nigericin technique.²¹

We calculated net H⁺ and HCO₃⁻ transport across plasma membranes as the product of the intracellular pH (pH_i) recovery rate and the intracellular buffering capacity. Intrinsic buffering capacity (β_i) was estimated on the basis of the magnitude of the pH_i change in response to basolateral addition or removal of 40 mM acetate. To isolate effects of intrinsic physicochemical buffering, plasmalemmal acid-base transport was inhibited in these experiments by combined basolateral and luminal application of a Na⁺- and CO₂/HCO₃⁻-free solution containing 1 mM amiloride. On average, the intrinsic buffering capacity was 58.6±6.1 mM (n=6) in mTALs from NBCn1 KO mice, which was not significantly different (P=0.25; unpaired two-tailed t test) from 50.8±2.9 mM (n=7) in mTALs from WT mice. We plotted the intrinsic buffering capacities as function of pH_i, and used the corresponding least-squares linear regression model to calculate intrinsic buffering capacities at the individual pH_i values where acid-base transport activities were evaluated. In experiments performed in the presence of CO₂/HCO₃⁻, the total intracellular buffering capacity (β_total) was calculated as follows: β_total=β_i+2.3×[HCO₃⁻]_i.²² We calculated [acetate] and [HCO₃⁻] on the basis of the Henderson-Hasselbalch equation, using a pK_a of 4.76 for acetic acid and 6.1 for the CO₂/HCO₃⁻ buffer system. Potential size differences between mTALs from NBCn1 KO and WT mice could influence the comparison of net acid extrusion activities. However, we observed no obvious differences in tubular morphology; accordingly, on the basis of luminal diameters and epithelial thicknesses measured from BCECF fluorescence images, we estimated very similar transverse, cross-sectional, epithelial tissue areas of 287±19 and 324±18 µm² in mTALs from NBCn1 KO and WT mice, respectively (n=10; P=0.18; unpaired two-tailed t test).

The physiologic saline solution used for functional experiments had the following composition (in mM): 120 NaCl, 0.4 KH₂PO₄, 1.6 K₂HPO₄, 5 glucose, 5 HEPES, 1 MgCl₂, 1.3 calcium gluconate, 22 NaHCO₃. In Na⁺-free solutions, we substituted N-methyl-D-glucammonium for Na⁺, except for NaHCO₃, which we replaced with choline-HCO₃. In HCO₃⁻-free solutions, Cl⁻ replaced HCO₃⁻. HCO₃⁻-containing solutions were aerated with 5% CO₂/balance air, whereas HCO₃⁻-free solutions were bubbled with nominally CO₂-free air; all solutions were titrated to pH 7.4 at 37°C.

Measurement of NH₄⁺ Reabsorption across Isolated Perfused mTALs

We isolated mTALs that were at least 200-µm long, from WT and NBCn1 KO mice that had received drinking water with 196 mM NH₄Cl for 1 day. The tubules were perfused as described above, except that their distal end was placed inside a collection pipette filled with mineral oil, and 5 mM NH₄Cl was added to the physiologic saline solution used in the bath and for tubule perfusion. We collected fluid twice during each perfusion protocol lasting 30–35 minutes, with mean perfusion flow rates of 2.9±0.4 and 2.7±0.4 nl/min for mTALs from WT and NBCn1 KO mice, respectively (P=0.68). The tubules were assumed to be water-impermeable, and the volume of the collected fluid samples was calculated on the basis of the geometry of a transfer pipette. The collected samples were transferred to small glass capillaries, and [NH₄⁺] was determined using a fluorometric NH₄⁺ kit (MAK310; Sigma-Aldrich), with fluorescence signals (excitation: 360 nm, emission: 450 nm) acquired using a Princeton Instruments MicroMAX 5 MHz System mounted on a Zeiss Axiovert 100 TV inverted microscope. From the fluid collection rate and [NH₄⁺] measured in perfusate and collected fluid, we calculated the net rate of NH₄⁺ reabsorption across mTALs, and expressed it normalized to the individual tubule lengths.

Reverse Transcription and PCR

We analyzed expression of mRNA in TALs using reverse transcription and PCR (RT-PCR) analyses. Anesthetized mice were perfused with 10 ml physiologic saline solution (see composition above) through a cannula in the left ventricle, to drain the blood from the circulation. The kidneys were quickly removed, released from their capsule, and sliced transversely. The kidney slices were digested with 1 g/L collagenase type II at 37°C, and placed on a shaking table at 850 rpm for 10 minutes. The sample, apart from any precipitate, was then transferred to ice-cold sorting solution containing 500 mg/L albumin, and left to sediment for 5 minutes, before the supernatant was removed and an equivalent amount of new sorting solution was added. This procedure was repeated twice, and the enzymatically dispersed tubules were then manually sorted under a dissection microscope. Judged by visual inspection, equal numbers of TALs were included in each sample, which was then homogenized in RLT lysis buffer, using a Qiagen TissueLyser II at 30 Hz for 3 minutes. RNA was isolated using an RNeasy Micro kit (Qiagen) according to the manufacturer’s protocol, and reverse transcribed using random decamer primers (Eurofins Genomics) and Superscript III reverse transcriptase (Invitrogen). For each sample, we performed parallel procedures without reverse transcriptase added, to control for genomic amplification in subsequent PCR analyses.

For quantitative evaluation of NBCn1 mRNA levels, we performed TaqMan quantitative RT-PCR using the following primers: forward: 5′-ACA GAA GGC AGA ATA AGT GCA ATA GA-3′; reverse: 5′-AGG TTG CCC AGC AAA CAA TG-3′; and probe: 5′-AGG CAA TCC CAG TTA ATG ATG CTC CAA AA-3′. The C_T values were related to those of the reference genes glyceraldehyde-3-phosphate dehydrogenase (Mm99999915_g1; Applied Biosystems) and transferrin receptor protein 1 (either forward: 5′-TTT GGG CAC TAG ATT GGA TAC CT-3′; reverse: 5′-AGT AGT CTC CAC GAG CGG AAT AC-3′; and probe: 5′-CAG CGG AAG TGG CTG GTC AGC TCA TTA TTA AA-3′, or Mm01344478_m1 from Applied Biosystems). Template complementary DNA was mixed with 2 mM deoxyribonucleotide triphosphates and Maxima Hot Start Taq DNA polymerase in 1× polymerase buffer. The following reaction sequence was run: 5 minutes at 95°C, then 50 cycles at 95°C for 10 seconds, 58°C for 20 seconds, and 72°C for 30 seconds.

To identify alternative molecular candidates of the Slc4 family that could contribute to Na⁺-dependent HCO₃⁻ uptake in mouse TALs, we used conventional end point RT-PCR analyses with the following primers: NBCn1/Slc4a7 forward: 5′-CCA AGT TTC TGG GAA TTC GTG AAC AG-3′, reverse: 5′-CTG GCA TGA GGT CAT CAA GCC AAC-3′; NBCe1/Slc4a4 forward: 5′-CTC ACT TCT CCT GTG CTT GCC T-3′, reverse: 5′-GTG GTT GGA AAA TAG CGG CTG G-3′; NBCe2/Slc4a5 forward: 5′-AAA GCT GCA CGT GCC TAG TGT C-3′, reverse: 5′-GCC TGT CAG GAT GAA GAC CAT G-3′; NCBE/NBCn2/Slc4a10 forward: 5′-AGG GCA CAG TAC AAC TCC CAC-3′, reverse: 5′-GGG ATG GGA GAG AGG GTT TAC-3′; NDCBE/Slc4a8 forward: 5′-GCT CAA GAA AGG CTG TGG CTA C-3′, reverse: 5′-ACG CCT TAA TGA CCC AGA GCA G-3′; BTR1/Slc4a11 forward: 5′-CAC CTG CTG TCA GAT ACC ATC C-3′, reverse: 5′-TGG TGA GCA GCT GTC TCT GAT G-3′; AE4/Slc4a9 forward: 5′-GTG TTG GGC GAC TTT TCC TC-3′, reverse: 5′-GGT GCC AGG AAG ATG GAG AC-3′. Template complementary DNA was amplified through the following sequence, using Maxima Hot Start Taq DNA polymerase (Thermo Fisher Scientific) in 1× polymerase buffer with 2 mM deoxyribonucleotide triphosphates: first, 4 minutes at 95°C; then, 35 cycles consisting of 20 seconds at 95°C, 45 seconds at 58°C, and 1 minute at 72°C; and finally, 5 minutes at 72°C. Each primer set was tested on an appropriate positive control tissue. PCR products were separated by 1% agarose gel electrophoresis and ultraviolet-visualized with Midori Green nucleic acid stain (LabGene). Specificity of the reactions was confirmed by sequencing products of expected sizes.²³

Immunoblotting

Excised kidneys from WT and age-matched NBCn1 KO mice were sliced transversely, and the inner stripe of the outer medulla was isolated under a dissection microscope. The samples were flash frozen in liquid N₂ and stored at −80°C until further analysis. Then, 50 µl lysis buffer containing 10 µl/ml Halt protease and phosphatase inhibitor cocktail (Thermo Fisher Scientific) was added to the tissue, which was then homogenized using pellet pestles (Sigma-Aldrich). Samples were sonicated for 45 seconds and centrifuged at 9500×g for 10 minutes. The total protein concentration in the supernatants was measured using a bicinchoninic acid protein assay (Bio-Rad). We loaded 10 µg total protein diluted in sample buffer in each lane of an SDS polyacrylamide gel for electrophoresis and transfer to polyvinylidene difluoride membranes blocked with 0.3% i-block (Applied Biosystems). Membranes were probed with an anti–NH₂-terminal NBCn1 antibody (a kind gift from Dr. Jeppe Praetorius, Aarhus University),²⁴ which had been affinity-purified using the corresponding immunizing peptide (21^st Century Biochemicals). The lower part of the gel was stained for pan-actin (antibody 4968s; Cell Signaling) and used as a loading control. After thorough washing, the membranes were probed with secondary goat anti-rabbit antibody (G21234; Invitrogen) conjugated to horseradish peroxidase. Bound antibody was detected by enhanced chemiluminescence (ECL Plus; GE Healthcare) using an ImageQuant LAS 4000 luminescent image analyzer (GE Healthcare), and band intensities were quantified using Image Studio Lite Version 5.2 software (LI-COR Biosciences).

Arterial Blood Gas Measurements

Mice that had received normal tap water or been given access to drinking water containing 196 mM NH₄Cl for 12 hours, 1 day, or 4 days were anesthetized with a combination of ketamine (60 mg/kg) and xylazine (10 mg/kg). The mice were then placed on a thermostatically controlled heating platform to keep body temperature at 37°C, and were endotracheally intubated and ventilated to normocapnia (arterial Pco₂ of approximately 40 mm Hg) for at least 5 minutes on a Minivent type 845 ventilator (Harvard Apparatus), before blood was drawn from the common carotid artery. The arterial blood samples were immediately analyzed for pH, Pco₂, and standard [HCO₃⁻], in addition to hematocrit, Po₂, [Na⁺], [K⁺], [Ca²⁺], and [Cl⁻], on an ABL80 FLEX (Radiometer).

Urinary Acid-Base Analyses

We collected spot urine on Parafilm²⁵; measured pH with a microelectrode (pH 200; Unisense); evaluated titratable acid as previously described,²⁶ by titration to an end point pH of 7.4; and analyzed [NH₄⁺] with a fluorometric kit (MAK310; Sigma-Aldrich), according to the manufacturer’s instructions. Metabolic acidosis was induced by providing mice with drinking water containing 196 mM NH₄Cl for 1 day. In a subset of experiments, mice were given an additional acute water or NH₄Cl challenge (20 µl/g mouse) by intragastric gavage, and the first subsequent spontaneously released urine was collected after an average (±SD) of 73±23 minutes, and analyzed for [NH₄⁺] as described above. To evaluate the urinary excretion rate of NH₄⁺, the average diuresis for the first 3 hours after gavage was estimated on the basis of urine volumes collected from WT and NBCn1 KO mice kept in metabolic cages. We observed no significant difference (P=0.76; unpaired two-tailed t test) in diuresis between NBCn1 KO (382±32 µl) and WT (401±49 µl) mice for the first 3 hours after intragastric gavage (n=17–18). The diuresis was also similar (P=0.94; unpaired two-tailed t test) between NBCn1 KO (78±20 µl) and WT (81±23 µl) mice when compared for the same 3 hours of the day, in experiments where no gavage load was applied (n=6).

We approximated the expected change in urinary NH₄⁺ concentration (Δ[NH₄⁺]_u) between the acid-loaded NBCn1 KO and WT mice on the basis of the assumptions that (1) the change in pH and [HCO₃⁻] of the extended extracellular fluid volume after 24 hours follows that measured in arterial blood (ΔpH_a and Δ[HCO₃⁻]_a); and (2) the extended extracellular fluid volume (V_EFV) is 20% of body wt and has an average noncarbonic buffering capacity (β_EFV) of 11 mM. We used the following equation: Δ[NH₄⁺]_u=V_EFV×(β_EFV×ΔpH_a+Δ[HCO₃⁻]_a)/D, where D is the diuresis during the first day of acid loading.

Renal Tissue [NH₄⁺] Measurements

We evaluated the renal corticomedullary NH₄⁺ gradient essentially as previously described.²⁷ We sampled biopsy tissue of the renal cortex and outer and inner medulla from coronal slices of snap-frozen kidneys from NBCn1 KO and WT mice. The mice had either been housed under control conditions or loaded with 196 mM NH₄Cl in the drinking water for 1 day. Tissue samples were homogenized in 260 µl Milli-Q water, using a Qiagen TissueLyser II. We then added 2% pH-adjusting ionic strength adjustor (#951211; Thermo Fisher Scientific), and measured [NH₄⁺] with an Orion High-Performance Ammonia Electrode (Thermo Fisher Scientific). As described by others,²⁷ we controlled for differences in tissue amount between samples by normalizing the recorded [NH₄⁺] values to the total protein concentration in the lysates, measured using a bicinchoninic acid protein assay (Thermo Fisher Scientific).

Statistical Analyses

Unless otherwise specified, data are shown as mean±SEM. The n values report the number of mice investigated (biologic replicates). A few measurements that fell more than 1.5 times the interquartile range above the third quartile, or below the first quartile of the combined group of WT and NBCn1 KO mice, were classified as outliers and excluded from further analysis. Unpaired two-tailed t tests were used to compare one variable between two independent groups. We tested the effect of two variables on the measured variable using two-way ANOVA, followed by Sidak post-tests. P<0.05 was considered statistically significant. Data analyses and statistical tests were performed using Microsoft Excel 2016 and GraphPad Prism 7.05.

Results

We investigated the role of NBCn1 for basolateral HCO₃⁻ uptake and NH₄⁺ reabsorption in mTALs, the renal corticomedullary NH₄⁺ gradient and urinary NH₄⁺ excretion, and acid-base homeostasis of arterial blood under control conditions and elevated systemic acid loads.

NBCn1 Mediates the Upregulated Basolateral HCO₃⁻ Uptake in Response to Systemic Acid Loading

As illustrated in the original trace (Figure 2A), we exposed isolated perfused mTALs—first luminally and then basolaterally—to Na⁺-free solution containing 1 mM amiloride, which resulted in stepwise intracellular acidification consistent with inhibition of Na⁺-dependent net acid extrusion. Amiloride was used to pharmacologically inhibit Na⁺/H⁺ exchange activity. From the new acidic steady-state pH_i level, we evaluated Na⁺/HCO₃⁻ cotransport activity on the basis of the rate of amiloride-insensitive pH_i recovery upon return of basolateral Na⁺ (Figure 2, A and B). Subsequent application of standard physiologic saline solution—both basolaterally and luminally—brought pH_i back to the initial steady-state level (Figure 2A).

Figure 2.

NBCn1 mediates the upregulated basolateral Na⁺/HCO₃⁻ cotransport activity in isolated perfused renal mTALs from mice with metabolic acidosis. (A) Original trace illustrating the protocol for quantifying HCO₃⁻-dependent acid-base transport in mTALs. The dashed rectangle indicates the initial recovery phase, which is expanded in (B) and used for evaluating Na⁺/HCO₃⁻ cotransport activity in (C). (B) Average traces (n=5–7) of amiloride-insensitive Na⁺-dependent pH_i recovery from intracellular acidosis corresponding to the dashed rectangle in (A). The illustrated experiments were performed on mTAL tubules from mice under control conditions, or after 4 days of oral acid loading. (C) Na⁺/HCO₃⁻ cotransport activity in isolated perfused mTALs from NBCn1 KO and WT mice under control conditions, and after 1 or 4 days of oral acid loading (n=3–7). We evaluated net base uptake in the same pH_i interval for control and acid-loaded mice (P=0.17, two-way ANOVA). For mTALs from WT mice, average pH_i in the analyzed recovery phase was 6.76±0.05 in the presence of CO₂/HCO₃⁻, and 6.65±0.07 in the absence of CO₂/HCO₃⁻. Corresponding values for mTALs from NBCn1 KO mice were 6.74±0.04 in the presence of CO₂/HCO₃⁻, and 6.56±0.04 in the absence of CO₂/HCO₃⁻. Data were compared by two-way ANOVA, followed by Sidak post-test or unpaired two-tailed t tests. *P<0.05; **P<0.01; NS, not significantly different versus WT exposed to same duration of acid loading and investigated with CO₂/HCO₃⁻ present. ^##P<0.01; ^###P<0.001 versus same genotype exposed to equal duration of acid loading and investigated with CO₂/HCO₃⁻ present. BL, basolateral; Lu, luminal; MAc, metabolic acidosis.

Figure 2C shows the rate of amiloride-insensitive Na⁺-dependent net base uptake calculated from the initial recovery rates upon readdition of Na⁺ (Figure 2B), taking into account the intracellular buffering capacity. We detected robust Na⁺-dependent pH_i recovery (Figure 2B) and basolateral net acid extrusion capacity (Figure 2C) in isolated perfused mTALs from both WT and NBCn1 KO mice under control conditions. In mTALs from WT mice that had been acid-loaded through the drinking water for 1 or 4 days, we observed a substantially faster amiloride-insensitive, Na⁺-dependent pH_i recovery (Figure 2B), and a corresponding four-fold higher net acid extrusion capacity across the basolateral membrane (Figure 2C). Whether mTALs were isolated from mice under control conditions or after 4 days of acid loading through the drinking water, the amiloride-insensitive, Na⁺-dependent net acid extrusion across the basolateral membrane was fully CO₂/HCO₃⁻-dependent (Figure 2C). Together, these findings demonstrate upregulated basolateral net acid extrusion via Na⁺/HCO₃⁻ cotransport in mTALs during metabolic acidosis.

The molecular mechanism for basolateral HCO₃⁻ uptake into mTALs under control conditions is not yet clear, but is unlikely related to NBCn1 because Na⁺/HCO₃⁻ cotransport activity was of similar magnitude in mTALs from WT and NBCn1 KO mice that had not been acid loaded (Figure 2C). In contrast, NBCn1 was essential for the accelerated basolateral Na⁺/HCO₃⁻ cotransport activity in mTALs from WT mice during metabolic acidosis, because no such increase in amiloride-insensitive net acid extrusion capacity was observed in mTALs from NBCn1 KO mice (Figure 2C).

Systemic Acid Loading Upregulates NBCn1 Protein Expression Ten-Fold

In parallel with the increase in Na⁺/HCO₃⁻ cotransport activity, the expression of NBCn1 protein in the inner stripe of the renal outer medulla of WT mice increased almost ten-fold during oral acid loading, compared with mice receiving regular tap water (Figure 3A). NBCn1 protein expression levels were low in WT mice under control conditions, and indiscernible in NBCn1 KO mice both under control conditions and during metabolic acidosis (Figure 3A).

Figure 3.

NBCn1 protein expression is dramatically elevated during metabolic acidosis without change in NBCn1 mRNA levels. (A) NBCn1 protein expression in the inner stripe of renal outer medulla of NBCn1 KO and WT mice (n=6). (B) Expression of NBCn1 mRNA in enzymatically isolated TALs from NBCn1 KO and WT mice (n=6) under control conditions. (C) Expression of NBCn1 mRNA in enzymatically isolated TALs from WT mice (n=6) under control conditions, and after 4 days of oral acid loading. All data are presented relative to the average level in WT mice under control conditions. Data were compared by two-way ANOVA, followed by Sidak post-test or unpaired two-tailed t tests. ***P<0.001; NS, not significantly different versus WT under similar conditions. ^###P<0.001 versus WT under control conditions. MAc, metabolic acidosis.

Expression of NBCn1 mRNA detected in enzymatically isolated TALs from WT mice was completely absent in TALs from NBCn1 KO mice (Figure 3B). In striking contrast to the ten-fold upregulation of NBCn1 protein expression (Figure 3A), mRNA levels for NBCn1 changed only marginally during metabolic acidosis, and showed no significant differences between TALs isolated from WT mice under control conditions and after 4 days of oral acid loading (Figure 3C).

Several Other Na⁺HCO₃⁻ Cotransporters Are Expressed in TAL

As described above, the basal Na⁺/HCO₃⁻ cotransport in mice receiving standard laboratory chow and water did not depend on NBCn1 (Figure 2C). In congruence, in addition to NBCn1, we detect transcripts for a number of other known and putative Na⁺/HCO₃⁻ cotransporters from the Slc4 family in enzymatically isolated TALs (Figure 4). On the basis of these observations, NBCe1, NDCBE, AE4, and BTR1 are possible candidates for cellular HCO₃⁻ uptake in mTALs from unchallenged mice.

Figure 4.

Renal TALs express NBCn1, NBCe1, NDCBE, AE4, and BTR1 at mRNA level. The figure shows the results of end point RT-PCR on enzymatically isolated renal TALs used to screen for putative Na⁺/HCO₃⁻ cotransporters of the Slc4 family. In addition to the reactions on the basis of mRNA from TALs, for each primer set we performed a positive control reaction on a tissue known to express the given transporter, and a negative control reaction without template added. Reactions without addition of reverse transcriptase were performed in parallel with each reaction, and in each case produced no band (data not shown). The vertical white lines indicate delineations between separate agarose gels. Each reaction shown is representative of three experiments. Sequencing confirmed specificity of bands.

Na⁺/H⁺ Exchange Activity Is Upregulated in NBCn1 KO Mice and during Systemic Acid Loading

The absence of NBCn1-mediated HCO₃⁻ uptake in mTALs of NBCn1 KO mice could lead to compensatory upregulation of other acid-base transporters.¹⁸ We therefore explored the capacity for basolateral Na⁺/H⁺ exchange, which represents a parallel pathway for cellular net acid extrusion (Figure 1), in mTALs from NBCn1 KO and WT mice under control conditions and during metabolic acidosis (Figure 5). Figure 5A shows an original trace of the experimental protocol, which was performed in its entirety without CO₂/HCO₃⁻. Application of Na⁺-free solution—first luminally and then basolaterally—caused intracellular acidification (Figure 5A). Amiloride was added to the luminal solution to block contribution from Na⁺/H⁺ exchange across the apical membrane. From the new acidic steady-state pH_i, we measured Na⁺/H⁺ exchange activity on the basis of the pH_i recovery rate upon readdition of basolateral Na⁺ (Figure 5, A and B). Returning standard CO₂/HCO₃⁻-free solution also to the tubule lumen brought back pH_i to the initial steady-state level (Figure 5A).

Figure 5.

Na⁺/H⁺ exchange is upregulated by oral acid loading and KO of NBCn1. (A) Original trace illustrating the protocol for determining CO₂/HCO₃⁻-independent acid-base transport activities in mTALs. All of the applied solutions were without CO₂ and HCO₃⁻. The dashed rectangle indicates the initial recovery phase, which is expanded in (B) and used for evaluating Na⁺/H⁺ exchange activity in (C). (B) Average traces (n=5–9) of Na⁺-dependent pH_i recovery from intracellular acidosis corresponding to the dashed rectangle in (A). The illustrated experiments were performed in absence of CO₂/HCO₃⁻ on mTAL tubules from mice under control conditions, or after 4 days of oral acid loading. (C) Na⁺/H⁺ exchange activity in isolated mTALs from NBCn1 KO and WT mice (n=5–9) under control conditions, and after 4 days of oral acid loading. For mTALs from WT mice, average pH_i in the analyzed recovery phase was 6.79±0.06 (control) and 6.84±0.03 (MAc 4 days). Corresponding values for mTALs from NBCn1 KO mice were not significantly different (P=0.55; two-way ANOVA), at 6.78±0.04 (control) and 6.92±0.04 (MAc 4 days). Data were compared by two-way ANOVA, followed by Sidak post-test or unpaired two-tailed t test. *P<0.05; NS, not significantly different versus WT exposed to the same duration of oral acid loading. ^#P<0.05 versus WT under control conditions. BL, basolateral; Lu, luminal; MAc, metabolic acidosis.

Figure 5C shows the rate of Na⁺-dependent net acid extrusion in absence of CO₂/HCO₃⁻, taking into account the intracellular buffering capacity. KO of NBCn1 resulted in two-fold upregulation of Na⁺/H⁺ exchange activity under control conditions (Figure 5C). Although oral acid loading for 4 days doubled the basolateral Na⁺/H⁺ exchange activity in mTALs from WT mice, it had no effect in mTALs from NBCn1 KO mice (Figure 5C). Thus, we observed no difference in the capacity for basolateral Na⁺/H⁺ exchange activity between mTALs from acid-loaded WT and NBCn1 KO mice (Figure 5C).

Restoration of Systemic Acid-Base Balance Is Delayed in Acid-Loaded NBCn1 KO Mice

Giving WT mice free access to drinking water containing 196 mM NH₄Cl caused their arterial blood to acidify by 0.2 pH units within 12 hours, and then return gradually toward the normal resting level over the next few days, despite continued supply of NH₄Cl (Figure 6A). In NBCn1 KO mice, we observed similar degrees of acute acidification, but the recovery of arterial blood pH was slower when compared with WT mice (Figure 6A). Arterial standard [HCO₃⁻] followed the same pattern observed for arterial pH: after the initial decline, the recovery was delayed in NBCn1 KO compared with WT mice (Figure 6B).

Figure 6.

NBCn1 KO delays recovery of arterial blood gas parameters during metabolic acidosis induced by oral NH₄Cl loading. Arterial blood was sampled from the carotid artery of mice ventilated mechanically to normocapnia. (A) Arterial blood pH (solid lines) during oral loading with NH₄Cl. The corresponding Pco₂ (dashed lines) was maintained at approximately 40 mm Hg through mechanical ventilation (n=9–14). (B) Arterial blood standard [HCO₃⁻] measured under the same conditions shown in (A) (n=9–14). Data were compared by two-way ANOVA, followed by Sidak post-test. Values for standard [HCO₃⁻] were log-transformed before statistical comparisons were performed, to improve normal distribution. *P<0.05; **P<0.01 versus WT at same time point.

Apart from the differences in pH and standard [HCO₃⁻], the arterial blood composition was largely similar in NBCn1 KO and WT mice (Table 1).

Table 1.

Biochemical values of arterial blood collected from WT and NBCn1 KO mice (n=9–11) under control acid-base conditions

Biochemical values	WT	NBCn1 KO
Hematocrit, %	42.4±0.9	42.3±1.0 NS
Po2, mm Hg	100±2	101±2 NS
[Na+], mM	149.6±0.5	150.6±0.3 NS
[K+], mM	3.64±0.10	3.63±0.13 NS
[Ca2+], mM	1.05±0.02	1.03±0.03 NS
[Cl−], mM	111.9±0.7	112.1±0.5 NS

Data were compared by unpaired two-tailed t tests. Figure 6 shows corresponding values for arterial Pco₂, standard [HCO₃⁻], and pH. NS, not significantly different versus WT.

Reduced Capacity for Urinary NH₄⁺ Excretion in Acid-Loaded NBCn1 KO Mice

To evaluate if the delayed recovery of arterial blood pH in NBCn1 KO mice challenged with a metabolic acid load relates to reduced capacity for renal net acid excretion, we next measured urinary pH, [NH₄⁺], titratable acids, and the corresponding diuresis to estimate excretion rates (Figure 7).

Figure 7.

NBCn1 KO limits the capacity for urinary NH₄⁺ excretion without affecting urine pH or titratable acid. (A, C, and E) Urinary pH (A), [NH₄⁺] and corresponding NH₄⁺ excretion rates (C), and titratable acid (E), before and after WT and NBCn1 KO mice were acid-loaded through their drinking water for 1 day (n=6–12). Urinary excretion rates were estimated by multiplying the measured urine concentrations with the average urine output in WT and NBCn1 KO mice kept in metabolic cages. (B, D, and F) Urinary pH (B), [NH₄⁺] (D), and titratable acid (F), with corresponding excretion rates for WT and NBCn1 KO mice (n=5–6) acid-loaded for 1 day through their drinking water, and given an additional acute H₂O or NH₄Cl challenge (20 µl/g) at the indicated concentration by intragastric gavage. The error bars indicate SEM for urinary pH and concentrations of NH₄⁺ and titratable acid, and these data were compared by two-way ANOVA, followed by Sidak post-tests. ***P<0.001; NS, not significantly different versus WT under same conditions. MAc, metabolic acidosis.

We first collected spot urine at baseline and after oral acid loading for 1 day. As illustrated in Figure 7, A, C, and E, we identified no significant differences in urinary acid-base parameters between WT and NBCn1 KO mice, using this approach. Assuming that pH and [HCO₃⁻] of the extracellular fluid volume follows arterial blood, then the additional acidity in NBCn1 KO mice after 24 hours (Figure 6A) can be explained by a reduced urinary excretion rate corresponding to 10–20 mM acid equivalents during the first day of oral acid loading (see Methods section for calculation details). A difference of this magnitude—corresponding to 5%–10% of the [NH₄⁺] in urine from acid-loaded WT mice—would be extremely difficult to document in particular when considering that a transient difference in NH₄⁺ excretion between WT and NBCn1 KO mice may not be fully temporally synchronized within the investigated groups of mice.

To better evaluate the maximal capacity for urinary acid excretion and optimize temporal synchronization, we next acutely challenged the acid-loaded mice with additional NH₄Cl by intragastric gavage. The urinary NH₄⁺ excretion increased much less in NBCn1 KO mice compared with WT mice when challenged with this acute acid load (Figure 7D), whereas urinary pH and titratable acids were similarly affected in NBCn1 KO and WT mice (Figure 7, B and F). The lower capacity for urinary NH₄⁺ excretion at a similar degree of urinary acidification supports that NBCn1 KO mice are restricted in their ability to create the corticomedullary NH₄⁺ gradient that is required for efficient diffusion trapping in the collecting ducts (Figure 1).

The Corticomedullary NH₄⁺ Gradient Is Absent in Acid-Loaded NBCn1 KO Mice

We next evaluated renal tissue concentrations of NH₄⁺ to more directly ascertain whether the lower urinary NH₄⁺ excretion can be explained by reduced accumulation of NH₄⁺/NH₃ in the renal medulla, and thus a smaller gradient for NH₄⁺ secretion in the collecting ducts (Figure 1).

We measured the tissue content of NH₄⁺ in renal cortex and outer and inner medulla from WT and NBCn1 KO mice under control conditions (Figure 8A), and in mice that had been acid-loaded through their drinking water for 1 day (Figure 8B). Consistent with observations from others,⁷ a robust corticomedullary NH₄⁺ gradient developed in kidneys from WT mice during the 1-day systemic acid load; however, no such gradient developed in kidneys from NBCn1 KO mice under the same conditions (Figure 8, A and B).

Figure 8.

NBCn1 KO disrupts medullary accumulation of NH₄⁺ during systemic acid loading, and inhibits NH₄⁺ reabsorption across mTALs. (A and B) NH₄⁺ content (relative to total protein content) in the renal cortex and outer and inner medulla of NBCn1 KO and WT mice under control conditions (A) (n=4–5), and after mice were acid-loaded through their drinking water for 1 day (B) (n=6–7). Data were compared by two-way ANOVA, followed by Sidak post-tests. (C) Rate of net NH₄⁺ reabsorption across isolated mTALs from WT and NBCn1 KO mice (n=5) that had been acid-loaded through their drinking water for 1 day. Data were compared by unpaired two-tailed t test. *P<0.05; NS, not significantly different versus WT under similar conditions. ^###P<0.001 versus WT inner medulla. MAc, metabolic acidosis.

Attenuated NH₄⁺ Reabsorption across Renal mTALs from NBCn1 KO Mice

The above findings support that NBCn1 is involved in the countercurrent multiplication mechanism that accumulates NH₄⁺/NH₃ in the medullary interstitium (Figure 1). To directly evaluate the contribution of NBCn1 to vectorial transport of NH₄⁺ across mTALs, we next assessed the net NH₄⁺ reabsorption in isolated mTALs, where perfusion flow rates can be controlled and [NH₄⁺] measured.⁷ As illustrated in Figure 8C, we found that NH₄⁺ reabsorption was approximately 60% lower in mTAL from NBCn1 KO compared with WT mice, when both groups of mice had been acid-loaded for 1 day.

Taken together, we demonstrate impeded capacity for excretion of systemic acid loads in NBCn1 KO mice (Figure 7D), which is consistent with absence of upregulated basolateral Na⁺/HCO₃⁻ cotransport (Figure 2, B and C) and inhibited NH₄⁺ reabsorption in mTALs (Figure 8C), lack of medullary NH₄⁺ accumulation (Figure 8B), and prolonged acidification of arterial blood (Figure 6) during systemic acid loading. Likewise, the unaffected acid-base status of arterial blood in unchallenged NBCn1 KO mice (Figure 6) is consistent with their maintained Na⁺/HCO₃⁻ cotransport activity (Figure 2, B and C), and low demand for medullary NH₄⁺ accumulation (Figure 8A) and urinary NH₄⁺ excretion (Figure 7C), under these conditions.

Discussion

Renal elimination of nonvolatile acid or base compensates systemic acid-base disturbances during physiologic and pathophysiologic challenges. In this study, we impose a continuous metabolic acid load on mice, and demonstrate a sequence of molecular, cellular, and integrative adaptations that rely on the electroneutral Na⁺/HCO₃⁻ cotransporter NBCn1. We show that during metabolic acidosis: (1) NBCn1 protein expression is ten-fold upregulated in the inner stripe of the renal outer medulla; (2) NBCn1 is responsible for the four-fold increase in basolateral HCO₃⁻ uptake in mTALs; (3) NBCn1 facilitates NH₄⁺ reabsorption across mTALs; (4) NBCn1 is necessary for establishing the corticomedullary NH₄⁺ gradient; (5) NBCn1 increases the capacity for urinary NH₄⁺ excretion; and (6) NBCn1 is required for the early rapid recovery of arterial blood pH and standard [HCO₃⁻].

The renal corticomedullary NH₄⁺ gradient (Figure 8B) provides a readily available source of acid equivalents that can quickly transfer to, and be trapped in, the luminal fluid of collecting ducts when H⁺ secretion increases and the preurine maximally acidifies. As illustrated in Figure 1, we propose that NBCn1 facilitates transepithelial NH₄⁺ reabsorption and contributes to the corticomedullary NH₄⁺ gradient by neutralizing intracellular H⁺ liberated during transport of NH₄⁺/NH₃ across mTALs. In support of this model, we measure lower NH₄⁺ reabsorption across mTALs (Figure 8C), attenuated renal medullary NH₄⁺ accumulation (Figure 8B), and reduced capacity for urinary NH₄⁺ excretion (Figure 7D) during early metabolic acidosis in NBCn1 KO compared with WT mice, despite maintained ability for urinary acidification (Figure 7B). During prolonged systemic acidosis, other mechanisms (e.g., additional increases in proximal tubule ammoniagenesis) are likely able to compensate for the lack of NBCn1-facilitated renal net acid excretion. Indeed, after 4 days of systemic acid loading, NBCn1 KO mice have reestablished arterial blood pH and standard [HCO₃⁻] to the same extent as WT mice (Figure 6).

Pharmacologic tools targeting Na⁺/HCO₃⁻ cotransporters are sparse, and not sufficiently selective to provide evidence for molecular contribution of individual transporters.^15–17 DIDS inhibits most Na⁺/HCO₃⁻ cotransporters of the Slc4 family, but the DIDS sensitivity of NBCn1 is controversial and most likely cell-type-dependent.^15,18,28 NBCn1 is usually reported DIDS-insensitive in epithelial cells. Our conclusion that NBCn1 is responsible for the basolateral HCO₃⁻ uptake in renal mTALs only during systemic acid loading (Figure 2, B and C) therefore contrasts with some previous investigations from unchallenged rats, where contribution of NBCn1 was suggested on the basis of DIDS insensitivity of the HCO₃⁻ uptake mechanism.^13,14 However, other previous reports provide support for DIDS-sensitive Na⁺/HCO₃⁻ cotransport in mTALs.⁸

Compensatory upregulation of other Na⁺/HCO₃⁻ cotransporters could, in principle, explain why we see no measurable difference in basolateral Na⁺/HCO₃⁻ cotransport activity between mTALs from NBCn1 KO and WT mice under control conditions (Figure 2, B and C). Notably, the basolateral Na⁺/H⁺ exchange activity in NBCn1 KO mice under control conditions is increased compared with WT mice (Figure 5, B and C), suggesting that NBCn1 plays some role in mTALs in unchallenged mice. The very low NBCn1 protein expression level in the inner stripe of renal outer medulla from WT mice under control conditions, however, supports that NBCn1 becomes functionally important only during metabolic acidosis when the protein expression level increases dramatically (Figure 3A). The identification of transcripts (Figure 4) for other known and putative Na⁺/HCO₃⁻ cotransporters of the Slc4 family (e.g., NBCe1, NDCBE, AE4, and BTR1) in TALs is consistent with their possible involvement, but functional investigations addressing their individual contributions are required to fully settle the molecular mechanism of basolateral HCO₃⁻ uptake in mTALs under control conditions.

The upregulation of NBCn1 protein expression (Figure 3A) and function (Figure 2C) in TALs during systemic acid loading is not accompanied by a proportional increase in NBCn1 mRNA expression (Figure 3C). Our findings therefore support that upregulation of NBCn1 in response to metabolic acidosis mainly involves enhanced translation and/or increased protein stability. Disproportional changes in mRNA and protein expression are common for NBCn1, and also observed, for instance, during breast carcinogenesis.²⁹

In addition to Na⁺/HCO₃⁻ cotransport mediated via NBCn1, net acid extrusion across the basolateral membrane of epithelial cells in mTALs can occur via the Na⁺/H⁺ exchanger NHE4 (Slc9a4), which has previously been found to contribute to renal NH₄⁺ excretion.⁷ This Na⁺/H⁺ exchanger, which is essentially amiloride-insensitive, does not appear to play a major role in our setting, because amiloride completely abolished the CO₂/HCO₃⁻-independent recovery from intracellular acidification (compare Figures 2C and 5C). Instead, the amiloride-sensitive NHE1 (Slc9a1) is a likely candidate for basolateral Na⁺/H⁺ exchange activity in mTALs.^7,30 Although mTALs from NBCn1 KO mice show elevated basolateral Na⁺/H⁺ exchange activity compared with WT mice under control conditions, they show no further elevation during systemic acid loading (Figure 5C). These findings may suggest an overlap between the molecular mechanisms underlying accelerated Na⁺/H⁺ exchange activity in WT mice during metabolic acidosis and in mice with disrupted NBCn1 expression. The enhanced Na⁺/H⁺ exchange activity in mTALs from unchallenged NBCn1 KO compared with WT mice (Figure 5, B and C) does not apparently result in altered systemic acid-base balance (Figure 6), renal tissue [NH₄⁺] (Figure 8A), changes in urinary acid excretion (Figure 7), or obvious electrolyte imbalances (Table 1). Likewise, although Na⁺/H⁺ exchange and Na⁺/HCO₃⁻ cotransport are functionally equivalent when the CO₂/HCO₃⁻ buffer is in equilibrium, the substantial basolateral Na⁺/H⁺ exchange activity in mTALs (Figure 5C) does not sufficiently compensate for the loss of NBCn1, with respect to renal NH₄⁺ handling and excretion (Figures 7D and 8, B and C), or the recovery of arterial blood pH and [HCO₃⁻] during metabolic acidosis (Figure 6). The distinct contributions from NBCn1 and basolateral Na⁺/H⁺ exchangers (in our setting, most likely NHE1) is not yet understood, but may be related to their substantial differences in subcellular expression patterns, interacting partners, activation pathways, and sensitivities to intra- and extracellular pH in many cell types.^31–33

The pathways allowing entry of NH₄⁺ and/or NH₃ from the medullary interstitium into the collecting duct lumen are still controversial. Na⁺/K⁺-ATPase activity can mediate basolateral uptake of NH₄⁺ in collecting ducts, and transmembrane diffusion of NH₃ relies on gas permeability of the lipid membrane and facilitated diffusion through gas channels made up of RhCG or RhBG proteins (Figure 1).^6,34,35 The NH₃ permeability of the collecting duct epithelium, together with the regulated acidity of the preurine, controls the degree of diffusion trapping and the concentration of NH₄⁺ in the final urine.

In conclusion, we show—as illustrated in Figure 1—that NBCn1 mediates the elevated Na⁺/HCO₃⁻ cotransport that accelerates basolateral net acid extrusion and transepithelial NH₄⁺ reabsorption in mTALs during systemic acid loading, leads to medullary NH₄⁺ accumulation, increases the capacity for urinary NH₄⁺ excretion, and is required for rapid renal compensation of metabolic acidosis.

Disclosures

E. Boedtkjer is inventor on a patent (PA 2016 70746) describing inhibitors of NBCn1 for anticancer therapy. All remaining authors have nothing to disclose.

Funding

This work was supported by Independent Research Fund Denmark grants 4004-00137B and 10-094816 (to E. Boedtkjer).