Expression of the B splice variant of NBCe1 (SLC4A4) in the mouse kidney

Abstract

Sodium-coupled bicarbonate transporters are critical for renal electrolyte transport. The electrogenic, sodium-coupled bicarbonate cotransporter, isoform 1 (NBCe1), encoded by the SLC4A4 geneencoded by the SLC4A4 gene has five multiple splice variants; the A splice variant, NBCe1-A, is the primary basolateral bicarbonate transporter in the proximal convoluted tubule. This study’s purpose was to determine if there is expression of additional NBCe1 splice variants in the mouse kidney, their cellular distribution, and their regulation by metabolic acidosis. In wild-type mice, an antibody reactive only to NBCe1-A showed basolateral immunolabel only in cortical proximal tubule (PT) segments, whereas an antibody reactive to all NBCe1 splice variants (pan-NBCe1) showed basolateral immunolabel in PT segments in both the cortex and outer medulla. In mice with NBCe1-A deletion, the pan-NBCe1 antibody showed basolateral PT immunolabel in both the renal cortex and outer stripe of the outer medulla, and immunoblot analysis showed expression of a ~121-kDa protein. RT-PCR of mRNA from NBCe1-A knockout mice directed at splice variant-specific regions showed expression of only NBCe1-B mRNA. In wild-type kidney, RT-PCR confirmed expression of mRNA for the NBCe1-B splice variant and absence of mRNA for the C, D, and E splice variants. Finally, exogenous acid loading increased expression in the proximal straight tubule in the outer stripe of the outer medulla. These studies demonstrate that the NBCe1-B splice variant is present in the PT, and its expression increases in response to exogenous acid loading, suggesting it participates in the PT contribution to acid-base homeostasis.

INTRODUCTION

Acid-base homeostasis is necessary for almost all components of normal health. Multiple studies have shown that abnormal acid-base homeostasis, whether metabolic acidosis or metabolic alkalosis, is predictive of increased mortality (23, 41, 47). In addition, metabolic acidosis directly impacts almost every organ system, leading to osteopenia and osteoporosis, decreased insulin release and sensitivity, impaired sensitivity to thyroid hormone activity, decreased cardiovascular responsiveness to catecholamines and other vasopressor hormones, increased risk of cardiac arrhythmias, skeletal muscle atrophy and weakness, renal interstitial fibrosis, and increased progression of chronic kidney disease to end-stage renal disease (14, 16, 23–25, 58–60).

The SLC4A4 gene gives rise to proteins that mediate critical roles in several components of acid-base homeostasis. These roles include proximal tubule basolateral sodium-bicarbonate cotransport that is critical for proximal tubule filtered bicarbonate reabsorption (1, 12, 27, 48), regulation of proximal tubule ammonia metabolism (21), and regulation of renal organic anion excretion (42). In addition, the SLC4A4 gene gives rise to protein products that are expressed in multiple extrarenal sites, including the central nervous system, pancreas, gastrointestinal tract, and reproductive organs (27, 38, 45, 49).

Alternative splicing of the SLC4A4 gene gives rise to multiple splice variants, which have different sites of tissue expression, different regulatory mechanisms, and different physiologic roles (11, 45, 49). The B, C, and E splice variants use a different transcription initiation site than do the A and D splice variants. The C splice variant has a 97-bp deletion not found in the B variant, and the D and E splice variants have a 21-bp deletion not found in the A, B, and C splice variants (5, 9, 11, 38). Current evidence indicates that all SLC4A4 splice variants mediate electrogenic Na⁺-coupled bicarbonate transport, and because they are members of isoform 1 of this type of transporter, they are often termed NBCe1 proteins. The A splice variant, NBCe1-A, appears to be the primary basolateral NBCe1 isoform expressed in the renal proximal tubule. The B splice variant, NBCe1-B, is expressed primarily in the pancreas (3), and the C splice variant, NBCe1-C, is primarily expressed in the central nervous system (5). However, both NBCe1-B and NBCe1-C are expressed in a wide variety of additional tissues (28, 45, 49). The D and E splice variants, NBCe1-D and NBCe1-E, respectively, lack a 27-nucleotide cassette, were initially identified in mouse reproductive tract tissues, and may have a wide tissue distribution (38). Quantitative analysis suggests NBCe1-D and NBCe1-E expression is substantially lower than that of the other splice variants (38).

The purpose of the current study was to determine if the kidney expresses NBCe1 splice variants other than NBCe1-A, and, if so, to characterize their cellular expression pattern under basal conditions and in response to exogenous metabolic acidosis. We used an antibody specific to all NBCe1 splice variants (pan-NBCe1) and an antibody specific to only the NBCe1-A splice variant (anti-NBCe1-A). We combined this with the use of mice with global SLC4A4 deletion, thereby deleting all NBCe1 splice variants, and with a mouse with specific deletion of the NBCe1-A splice variant. We utilized a combination of immunohistochemistry, immunoblot analysis, and RT-PCR to determine the expression of alternative NBCe1 splice variant proteins in the mouse kidney, their regulation by metabolic acidosis, and, through the use of RT-PCR, the specific splice variant expressed. Our results show that NBCe1-B is expressed in the kidney and exhibits basolateral proximal tubule expression, and that its expression is increased specifically in the outer medulla in response to metabolic acidosis.

METHODS

Animals.

We used mice with a NBCe1-A deletion that we reported recently (33). This mouse has an 11-bp deletion in the 5′ portion of the coding sequence specific to NBCe1-A (and NBCe1-D) that is not present in the coding sequence of NBCe1-B, NBCe1-C, or NBCe1-E (33). All breeding involved use of heterozygous sires with heterozygous dams, and homozygous deletion mice were compared with age-matched wild-type (WT) mice derived from the same breeding regimen.

NBCe1-A knockout (KO) mice were genotyped using tail-clip samples as we described recently (33). Briefly, DNA was amplified using the primers: 5′-ACCTCAGCTCATCTGAGGG-3′ and 5′-CACGGGTCAGCCACTTAA-3′. The resulting product was digested with the restriction enzyme MseI, which recognizes a TTAA sequence present in the PCR product from WT mice but not from KO mice. Thus, a 250-bp band is present in the WT product, and a 520-bp band is present in KO product. In some studies, the NBCe1-A deletion was detected utilizing high-resolution melting (HRM) analysis, performed by GenoTyping Center of America (Ellsworth, ME). NBCe1-A was amplified using Roche HRM mix (forward CTTTCCTCAGGGTTTTCCAG, reverse CTCCCAAGATGAATCGGATG) and run on a Roche LightCycler 480 with conditions suggested for the Roche HRM master mix. Analysis was performed using the gene scanning module of the Roche LightCycler 480 Software, release 1.5.1 (Roche Diagnostics, Indianapolis, IN). Samples were differentiated based on melting profiles and difference plots that are affected by underlying sequence variation. Groupings were based upon comparison to known genotype controls.

We also used mice with complete NBCe1 KO involving all splice variants (18). All breeding involved use of heterozygous sires with heterozygous dams. Offspring were genotyped using tail-clip samples as described previously (18, 21, 42). WT offspring were compared with homozygous NBCe1 KO offspring.

The Institutional Animal Care and Use Committees of the University of Florida and the North Florida/South Georgia Veterans Health System approved all animal experiments. Trained personnel in the University of Florida College of Medicine Cancer and Genetics Transgenic Animal Core Facility oversaw all animal breeding.

Antibodies.

Pan-NBCe1 antibodies were obtained from Proteintech Group (11885–1-AP, Rosemont, IL). We have previously reported the specificity of this antibody in studies using NBCe1 KO mice (21). Antibodies specific to NBCe1-A (α-333) and antibodies directed against NBCe1-B (α-332) were described previously (50). We demonstrated the specificity of the NBCe1-A antibody recently in studies using NBCe1-A KO mice (33).

Tissue preparation for immunolocalization.

Kidney tissues were preserved for immunohistochemistry using standard approaches in our laboratory (20, 31, 32, 34, 35). Briefly, mice were anesthetized with inhalant isoflurane, and the kidneys were preserved by in vivo cardiac perfusion with PBS (pH 7.4) with addition of 6000 U/L sodium heparin and 120 mg/L lidocaine followed by periodate-lysine-2% paraformaldehyde, cut transversely into several 2- to 3-mm-thick slices, and immersed for 48 h at 4°C in the same fixative. Kidney samples from each animal were embedded in polyester wax made using polyethylene glycol 400 distearate (Polysciences, Warrington, PA) with 10% 1-hexadecanol, and 2-μm-thick sections were cut and mounted on gelatin-coated glass slides.

Immunohistochemistry.

Immunolocalization was accomplished using standard immunoperoxidase procedures. Briefly, sections were dewaxed in ethanol, rehydrated, heated in Trilogy (Cell Marque Corp., Rocklin, CA) to 88°C for 30 min and then to 96°C for 30 min, cooled for 30 min, and rinsed in PBS. Endogenous peroxidase activity was blocked by incubating the sections in 3% H₂O₂ in distilled water for 45 min. Sections were blocked for 15 min with either Serum-Free Protein Block (DakoCytomation) or 2.5% normal horse serum (Vector) containing 5% BSA, 0.05% saponin, and 0.2% gelatin, and then incubated at 4°C overnight with primary antibody diluted in either Dako antibody diluent or 0.1% BSA, 0.3% Triton X in PBS. Sections were washed in PBS and incubated for 30 min with polymer-linked, peroxidase-conjugated goat anti-rabbit IgG (MACH2, Biocare Medical, Pacheco, CA), washed again with PBS, and then exposed to diaminobenzidine for 5 min. Sections were washed in distilled water, dehydrated in graded ethanols and xylene, mounted, and observed by light microscopy. Comparisons of labeling were made only between sections of the same thickness from the same immunohistochemistry experiment using identical reagents. Sections were examined using a Leica DM2000 microscope equipped with DIC optics and a Leica DFC425 digital camera and Leica DFC Twain Software and LAS application suite (Leica Microsystems, Buffalo Grove, IL). Color correction was performed using Adobe Photoshop software (Adobe Systems, San Jose, CA).

Protein preparation.

Mice were anesthetized with inhalant isoflurane, and tissues were rinsed by in vivo cardiac perfusion with PBS (pH 7.4) with addition of sodium heparin and lidocaine, as described above. The right renal vasculature was clamped, the right kidney was rapidly removed, and the cortex, outer stripe of the outer medulla, and inner stripe of the outer medulla were isolated rapidly under a dissection microscope. All samples were snap frozen in liquid nitrogen and stored frozen at −80°C until used. Tissues were homogenized in T-PER Tissue Protein Extraction Reagent (Pierce Biotechnology, Rockford, IL) using microtube pestles (USA Scientific, Ocala, FL), and protein was extracted according to the manufacturer’s recommended procedures. An aliquot was obtained for protein determination using a bicinchoninic acid assay, and the remainder was stored frozen at −80°C until used.

Immunoblot analysis.

Five to ten micrograms of renal protein were electrophoresed on 10% PAGE ReadyGel (Bio-Rad Laboratories Inc., Hercules, CA). Gels were then transferred electrophoretically to nitrocellulose membranes, blocked with 5 g/dl nonfat dry milk in Blotto buffer (50 mM Tris, 150 mM NaCl, 5 mM Na₂EDTA, and 0.05% Tween 20, pH 7.6), and incubated at 4°C overnight with primary antibody diluted in nonfat dry milk. Loading and transfer equivalence were assessed with Ponceau S staining. After washing, membranes were exposed to secondary antibody, goat anti-rabbit IgG (Cell Signaling Technology, Beverly, MA), conjugated to horseradish peroxidase at a dilution of 1:5,000. Sites of antibody-antigen reaction were visualized by using enhanced chemiluminescence (SuperSignal West Pico Substrate, Pierce) and a Kodak Image Station 440CF digital imaging system. Band density was normalized such that mean density in the same region (cortex or outer stripe of the outer medulla) in control tissues was 100.0. The absence of saturation was confirmed by examining pixel intensity distribution in all immunoblots.

PCR methods.

Tissues from cortex and outer stripe of mouse kidney were dissected and frozen immediately in liquid nitrogen, then stored at −80°C. Frozen tissues were then placed into RNALater-ICE solution (Thermo Fisher Scientific, Waltham, MA). RNA was extracted from frozen tissue using RNeasy Mini Kit (Qiagen Inc., Germantown, MD) and reverse transcribed to cDNA using SuperScript III First-Strand Synthesis System for RT-PCR (Invitrogen, Carlsbad, CA). PCR amplification was performed using GoTaq Hot Start Polymerase (Promega, Madison, WI). When nested amplification was performed, the initial PCR amplification used LongAmp Hot Start Taq DNA Polymerase (New England Biolabs, Ipswich, MA). The PCR products were separated on agarose gels and bands were excised. DNA was purified from gels using QIAquick Gel Extraction Kit (Qiagen). PCR product sequencing was performed by GENEWIZ (South Plainfield, NJ). Sequence alignment was performed using ClustalX, v 2.1. PCR primers were designed using NIH Primer-Blast (40a). The GenBank sequences used for the different NBCe1 splice variants are shown in Table 1. Real-time RT-PCR was carried out with PowerUp SYBR Green (Fisher A25741) using Applied Biosystems 7500 Real Time PCR System. 18S RNA transcript expression was used as an internal control. Differences in gene expression were quantified using 2^−ΔΔCt analysis. Amplification of only a single product was confirmed by melting curve analysis.

Table 1.

NBCe1 splice variants

Splice Variant	GenBank GI Identifier	GenBank Accession Version
A	304359384	HQ018820.1
B	133922579	NM_018760.2
C	210147429	NM_001136260.1
D	310894150	HQ285250.1
E	308210767	NM_001197147.1

Acid loading.

An acid diet was prepared as we described previously (10, 30, 34, 35, 52, 53). Briefly, we added 0.4 M HCl to powdered standard rodent chow in a ratio of 1 ml/g chow. The control diet was identical, except we substituted deionized water for HCl.

Statistics.

Results are presented as means ± SE. P < 0.05 was taken as statistically significant; n refers to the number of animals studied.

RESULTS

Expression of NBCe1 and NBCe1-A in the normal mouse kidney.

Our first set of experiments tested the possibility that alternative NBCe1 splice variants were present in mouse kidney. Immunohistochemistry using a pan-NBCe1 antibody showed strong basolateral immunolabel in the proximal convoluted tubule, moderate intensity immunolabel in the proximal straight tubule in the medullary ray, and light immunolabel in the proximal straight tubule in the outer stripe of the outer medulla (Fig. 1). A slightly different pattern was obtained using an antibody specific to the A splice variant, NBCe1-A. Immunohistochemistry using the NBCe1-A-specific antibody showed strong basolateral immunolabel in the proximal convoluted tubule in the cortical labyrinth and moderate intensity immunolabel in the proximal straight tubule in the medullary ray, but, in contrast to the pan-NBCe1 antibody, no detectable immunolabel in the proximal straight tubule in the outer stripe of the outer medulla. There was no detectable immunolabel in any other renal cell population with either antibody. We reported previously the specificity of the pan-NBCe1 antibody (32), and there was no detectable immunolabel in NBCe1-A KO mice using the NBCe1-A-specific antibody (data not shown). These findings identify the normal expression pattern of NBCe1 proteins, inclusive of all splice variants, and of the NBCe1-A splice variant in the mouse kidney. The finding of immunolabel in the proximal straight tubule in the outer stripe of the outer medulla with the pan-NBCe1 antibody, but not with the NBCe1-A-specific antibody, suggests the presence of protein for NBCe1 splice variants other than NBCe1-A in the proximal tubule.

Fig. 1.

Immunohistochemistry using pan-NBCe1 and NBCe1-A-specific antibody. Top: localization of the NBCe1-A immunolabel in the mouse kidney. Results using antibody specific to NBCe1-A (left), and results using an antibody that recognizes an epitope present in all NBCe1 splice variants, which we term pan-NBCe1 (right). Results are from serial sections. Bottom: high magnification images obtained from the areas outlined by rectangles in the top row images. Strong basolateral immunolabel is present in the proximal convoluted tubule in the cortical labyrinth with both NBCe1-A and pan-NBCe1 antibodies (A and D, respectively). There is modest basolateral immunolabel with both antibodies in the proximal straight tubule in the cortical medullary ray (B and E, respectively). No detectable immunolabel is present using the NBCe1-A antibody in the proximal straight tubule in the outer stripe of the outer medulla (C). In contrast, basolateral immunolabel using the pan-NBCe1 antibody is clearly present in the proximal straight tubule in the outer stripe of the outer medulla (F). This difference between pan-NBCe1 and NBCe1-A-specific antibodies suggests the presence of alternative NBCe1 splice variants in the mouse kidney. Results are representative findings in 4 kidneys in each group. NBCe1-A, A splice variant of NBCe1 protein; pan-NBCe1, antibody reactive to all NBCe1 splice variants.

Expression of NBCe1 splice variants in NBCe1-A KO kidney.

We then used mice with NBCe1-A deletion (33) to facilitate identification of alternative NBCe1 splice variant protein expression in the mouse kidney. These mice do not have early mortality and grow to adulthood. They exhibit spontaneous metabolic acidosis, are normokalemic, and have an intact, even exaggerated, ability to acidify their urine (33). First, we performed immunohistochemistry studies using an antibody that recognizes all NBCe1 splice variants. These experiments identified basolateral NBCe1 immunolabel in the proximal tubule of mice that do not express NBCe1-A (Fig. 2). This expression was present in both the proximal convoluted tubule, in the proximal straight tubule in the medullary ray, and in the proximal straight tubule in the outer stripe of the outer medulla. To confirm further the presence of alternative NBCe1 splice variants in the NBCe1-A deletion kidney, we performed immunoblot analysis of kidney tissue from mice with NBCe1-A deletion. This demonstrated expression of a protein with molecular mass of ~121 kDa (Fig. 2), consistent with the expected molecular mass of NBCe1 splice variants. Thus, there appears to be expression of NBCe1 splice variants other than NBCe1-A in the NBCe1-A KO kidney. These splice variants are expressed in the renal proximal tubule and exhibit an axial distribution that differs from that observed for NBCe1-A.

Fig. 2.

NBCe1 protein expression in the NBCe1-A knockout (KO) kidney. Immunoblot analysis of proteins from cortex and outer stripe of outer medulla (OSOM) from NBCe1-A KO kidney (A). Expression of protein of ~121 kDa is shown. Low-power (B) and high-power (C–E) micrographs of NBCe1 immunolabel in kidney of mice with genetic deletion of the NBCe1-A splice variant. Weak basolateral immunolabel is present in the proximal convoluted tubule and in the proximal straight tubule in both the cortical medullary ray and in the OSOM. Results are representative of findings in kidneys from four mice. We reported previously that there is no detectable immunolabel in the kidney of mice with deletion of all NBCe1 splice variants using this antibody (21). These results indicate the presence of an NBCe1 splice variant protein other than NBCe1-A in the NBCe1-A KO mouse kidney. Results are representative of findings in kidneys from 4 mice. NBCe1-A, A splice variant of NBCe1 protein.

RT-PCR evidence.

To determine which NBCe1 splice variants are expressed, we used RT-PCR to identify the specific NBCe1 transcript(s) present in the mouse kidney. The mouse has a total of five different NBCe1 splice variants (45). The B, C, and E splice variants differ from the A and D splice variants because they use an alternative initiation codon, which results in a different 5′ end of the mRNA coding sequence (Fig. 3). To determine whether transcripts for the B, C, or E splice variants were present, we designed two sets of RT-PCR reactions, both of which used a forward primer in the coding region that is specific to the B, C, and E splice variants and is not present in the A and D splice variants (primer sets 1 and 2, Table 2). Both RT-PCR reactions generated products of the expected size (Fig. 4). Sequencing these products demonstrated that the amplified mRNA sequence was identical to that expected for the 5′ mRNA coding sequence for the NBCe1-B/C/E transcript. Control samples amplified without the reverse transcriptase step showed no amplification (data not shown).

Fig. 3.

Overview of transcript differences for the five NBCe1 splice variants. NBCe1-A and NBCe1-D have a different transcription initiation site than do NBCe1-B, NBCe1-C, and NBCe1-E, resulting in different 5′ coding sequences. Both NBCe1-D and NBCe1-E have a 27-bp deletion. NBCe1-C has a 97-bp deletion. This results in a frameshift change in downstream amino acid residues and a different location of the stop codon. Figure also shows the approximate site of the regions amplified by the different RT-PCR primer sets used. Diagram is not drawn to scale. UTR, untranslated region.

Table 2.

Primers used for mRNA amplification

Primer Set	Forward Primer	Reverse Primer	Amplification Purpose	Location of Forward Primer in NBCe1-B	Location of Reverse Primer in NBCe1-B
1	CCCAGGAGGATGGAGGATGA	GGGCTGCCATTATCAGGGTT	Detection of NBCe1-B, NBCe1-C, or NBCe1-E transcripts	58–77	784–803
2	AGAATGCGGATGAGTCCAGC	CATGGCTGGGCTGCCATTAT	Detection of NBCe1-B, NBCe1-C, or NBCe1-E transcripts	278–297	791–810
3	AGAAGGAGACGCCACAAGAG	CACAGGTGTGCAACAACCAC	Nested amplification to differentiate B and C transcripts; will not amplify NBCe1-A transcript	187–206	4652–4671
4	TGTATATGGGGGTGGCCTCA	GAGGCACGACTTTCACTGGA	Internal primer set for nested amplification to differentiate B and C transcripts	2795–2814	3355–3374
5	TGAGCTTCTGGCTGTGGATG	ATTCCTGTGGGTCATGGCTG	Identification of possible NBCe1-D or NBCe1-E transcript	405–424	803–822

NBCe1-A, A splice variant of NBCe1 protein; NBCe1-B, B splice variant of NBCe1 protein; NBCe1-C, C splice variant of NBCe1 protein; NBCe1-D, D splice variant of NBCe1 protein; NBCe1-E, E splice variant of NBCe1 protein.

Fig. 4.

RT-PCR showing presence of “non-NBCe1-A” transcript. mRNA from wild-type (WT) and NBCe1-A knockout (KO) kidney was amplified using RT-PCR with two different sets of primers. Both Primer Set 1 and Primer Set 2 have the forward primer in the 5′ region specific for the B and C splice variants of NBCe1 (NBCe1-B and NBCe1-C; primer details are in Table 2). Reverse primer for both is in the NBCe1 transcript region conserved in all five splice variants. Top: results of PCR amplification. Gene products were obtained of the appropriate bp size with each set of primers. Primary bands were excised from the gel and sequenced. Bottom: sequence alignment that confirms amplification of mRNA transcripts identical to that present in non-NBCe1-A splice variants, i.e., NBCe1-B or NBCe1-C, in both WT and NBCe1-A KO kidney. Region amplified is also identical to that present in NBCe1-E, which is not shown for simplification. Regions common to all NBCe1 splice variants may not be shown. Results are representative of findings in four WT and four NBCe1-A KO kidneys. NBCe1-A, A splice variant of NBCe1 protein; NBCe1-E, E splice variant of NBCe1 protein.

The B and C splice variants differ because of a 97-bp deletion, beginning at bp-3169 of the NBCe1-B splice variant and extending to bp-3266. This 97-bp deletion causes a frameshift alteration in the downstream amino acid sequence, yielding different carboxy tails of the B and C splice variants. To differentiate between NBCe1-B and NBCe1-C, we performed RT-PCR using primer set 3, which had a forward primer unique to the NBCe1-B/C transcript and not present in the A splice variant, and a reverse primer downstream of the deletion site. A product of the expected size, 4.4 kb, was obtained that overlaps the portion of the coding sequence with the splice differences. We then amplified this PCR product using primer set 4, which flanks the potential deletion site. A ~530-bp product, consistent with the predicted size of the B splice variant, was obtained (Fig. 5). Sequencing this product confirmed amplification of a region specific to the B splice variant and not the C splice variant. Control samples amplified without the reverse transcriptase step showed no amplification (data not shown). These findings are consistent with NBCe1-B transcript expression in the NBCe1-A KO kidney.

Fig. 5.

Nested RT-PCR for differentiation of NBCe1-B and NBCe1-C mRNA products. mRNA from wild-type (WT) and NBCe1-A knockout (KO) kidneys underwent nested RT-PCR amplification. mRNA underwent RT-PCR amplification using primer set 3. The amplified product was excised, purified, and amplified by PCR using primer set 4. Top: results of the second PCR amplification. A product of ~500 bp was obtained from kidneys of both WT and NBCe1-A KO mice. Bottom: product was isolated and sequenced, and sequence alignment with consensus NBCe1-B and NBCe1-C sequence is shown. Area highlighted shows that sequences from both WT and the NBCe1-A KO mouse kidneys had 100% homology with the NBCe1-B mRNA sequence. A product with the 97-bp deletion found in the NBCe1-C mRNA sequence was not present. Thus, the NBCe1-C splice variant does not appear to be expressed in either WT or NBCe1-A KO kidney. Results are representative of findings in four WT and four NBCe1-A KO kidneys. NBCe1-A, A splice variant of NBCe1 protein; NBCe1-B, B splice variant of NBCe1 protein; NBCe1-C, C splice variant of NBCe1 protein.

To determine whether the NBCe1-B transcript is present in the WT kidney, we also performed this nested RT-PCR evaluation using mRNA from WT kidney. Identical results as observed in the NBCe1-A KO kidney were obtained (Fig. 5). Control samples amplified without the reverse transcriptase step showed no amplification (data not shown). These findings indicate that the NBCe1-B mRNA transcript is expressed in both WT and NBCe1-A KO kidney.

A previous report indicated the presence of cytoplasmic NBCe1-B immunolabel in the proximal tubule (50). Using the same antibody, we observed faint cytoplasmic, faint nuclear, and moderate subapical immunolabel in the proximal tubule. To determine if this immunolabel was specific, we examined the kidneys of mice with deletion of all NBCe1 splice variants (18, 21). Identical proximal tubule immunolabel was present in WT and NBCe1 KO mice (Fig. 6). Accordingly, we cannot confirm that the mouse kidney immunolabel found with this antibody is specific to NBCe1-B.

Fig. 6.

Immunolabel with antibody directed against NBCe1-B. Left: immunolabel in wild-type (WT) kidney using an antibody directed against amino-terminus of NBCe1-B that differs from the amino-terminus of NBCe1-A. Faint cytoplasmic and nuclear immunolabel and stronger label in the apical region is present. Right: immunolabel in the NBCe1 knockout (KO) mouse kidney. An identical immunolabel pattern is present. Accordingly, we cannot confirm this antibody's specificity in the mouse kidney. Results are representative of findings in three WT and three NBCe1-A KO kidneys. NBCe1-A, A splice variant of NBCe1 protein; NBCe1-B, B splice variant of NBCe1 protein.

Determination of NBCe1-D and NBCe1-E mRNA expression.

Two other, apparently minor, splice variants of the NBCe1 gene are the NBCe1-D and NBCe1-E splice variants. The NBCe1-D splice variant is generated from the same transcription initiation site as the A splice variant and is identical to the A splice variant in the region in which the genome-editing approach used to generate the NBCe1-A deletion created an early stop codon. Thus, the NBCe1-D splice variant protein product is also deleted in mice with NBCe1-A deletion and cannot explain the “non-NBCe1-A” protein expression findings in the NBCe1-A KO mice.

The E splice variant differs from the B and C splice variants because of a 27-bp deletion starting at nucleotide 776. Primer set 5 was used to amplify this region from NBCe1-A KO kidney. A product of ~320 bp was obtained, consistent with that predicted for the B splice variant (318 bp) and slightly larger than the predicted for either the D or E splice variant, 281 bp (Fig. 7). Sequencing this PCR product showed a sequence identical to that predicted for the B splice variant; a product with the 27-bp deletion present in the NBCe1-D and NBCe1-E splice variants was not present. Identical results were obtained using mRNA from WT kidney. Control samples amplified without the reverse transcriptase step showed no amplification (data not shown). These results thereby show no detectable expression of NBCe1-E mRNA transcript in either WT or NBCe1-A KO kidney.

Fig. 7.

RT-PCR to test for NBCe1-D and NBCe1-E splice variants. mRNA from NBCe1-A knockout (KO) mouse kidney was amplified with RT-PCR with primer set 5, which amplifies the region in which a 27-bp deletion is present in the NBCe1-D and NBCe1-E sequences. Left: products of ~400 bp were obtained. Right: Sequencing showed no evidence of the 27-bp deletion present in NBCe1-D and NBCe1-E transcripts. Results are representative of findings in four WT and four NBCe1-A KO kidneys. NBCe1-A, A splice variant of NBCe1 protein; NBCe1-B, B splice variant of NBCe1 protein; NBCe1-C, C splice variant of NBCe1 protein; NBCe1-D, D splice variant of NBCe1 protein; NBCe1-E, E splice variant of NBCe1 protein.

Because there was no evidence in the WT kidney of expression of a transcript with the 27-bp deletion present in the D splice variant, these results also indicate NBCe1-D mRNA is not present in the normal mouse kidney.

Effect of acid loading on NBCe1-B protein expression.

Metabolic acidosis increases proximal tubule bicarbonate reabsorption and stimulates proximal tubule ammonia metabolism (12, 15), proximal tubule functions critical to acid-base homeostasis. To determine whether NBCe1-B might contribute to these responses, we determined the effect of metabolic acidosis on NBCe1-B expression. Mice were fed acid-loading or control diet for seven days, and then kidney tissues were obtained. Acid loading decreased serum bicarbonate significantly in both genotypes (data not shown). In NBCe1-A KO kidneys, immunoblot analysis using the pan-NBCe1 antibody showed increased protein expression in the outer stripe of the outer medulla but not the cortex (Fig. 8). Immunohistochemistry demonstrated increased basolateral immunolabel in the proximal straight tubule in the outer stripe of the outer medulla. Acid loading also increased NBCe1-B mRNA expression significantly (Fig. 8).

Fig. 8.

Effect of experimental metabolic acidosis on NBCe1-B expression in the NBCe1-A knockout (KO) mouse kidney. A: immunoblot analysis with quantification. In the outer stripe of the outer medulla (OSOM), metabolic acidosis increased expression of NBCe1 total protein. Because only the B splice variant, NBCe1-B, is present in the NBCe1-A KO house kidney, these findings indicate that metabolic acidosis increases NBCe1-B protein expression. B: immunohistochemical findings. Exogenous acid loading increased basolateral NBCe1-B expression in the proximal straight tubule in the OSOM. Results are representative of findings from four mice on normal diet and six on acid diet. C: acid loading increased NBCe1-B mRNA expression. Expression was quantified using SYBR Green real-time RT-PCR using primer set 1 and was normalized to 18S RNA expression using 2^−ΔCt analysis. Results were then normalized such that mean normal diet expression was 100. n = 3 normal diet kidneys and 6 acid-loaded diet kidneys. NBCe1-A, A splice variant of NBCe1 protein.

We next examined the possibility that acid loading induced de novo NBCe1-C, NBCe1-D, or NBCe1-E expression. To do so, we repeated the mRNA analysis detailed above for identification of these transcripts in kidneys from acid-loaded WT and NBCe1-A KO mice (n = 3 in each group). We found no identifiable expression of any of these transcripts (data not shown), indicating that the increased protein expression seen in acid-loaded NBCe1-A KO mice cannot be ascribed to expression of the C, D, or E splice variants. Thus, in NBCe1-A KO mice, experimental metabolic acidosis increases NBCe1-B expression in the proximal straight tubule in the outer stripe of the outer medulla.

DISCUSSION

The current studies identify several new findings regarding splice variants of the SLC4A4 gene that encodes NBCe1 proteins in the mouse kidney. First, there is expression of an NBCe1 splice variant other than NBCe1-A in both the normal and the NBCe1-A KO mouse kidney, and this protein has a basolateral proximal tubule expression pattern. This splice variant appears to be the primary splice variant in the proximal straight tubule in the outer stripe of the outer medulla. Second, mRNA sequencing confirms expression of a “non-NBCe1-A” transcript and identifies this transcript as NBCe1-B. Finally, metabolic acidosis increases NBCe1-B protein expression in the proximal straight tubule in the outer stripe of the outer medulla. These results add significantly to our understanding of proximal tubule bicarbonate transporter expression and suggest NBCe1-B has an important role in renal acid-base homeostasis.

The first major finding of this study is that an NBCe1 splice variant other than the A splice variant is present in the kidney. Immunohistochemistry of WT kidney using an NBCe1 antibody directed against a region of the protein present in all splice variants shows immunolabel in the proximal straight tubule in the outer medulla, a region where NBCe1-A expression is not detectable. It is theoretically possible that differences in protein-protein interactions, or other factors that alter protein tertiary structure in fixed tissues, could cause these differences in the proximal straight tubule in the outer medulla. However, these theoretical possibilities cannot explain the finding that a pan-NBCe1 antibody identifies basolateral immunolabel expression in mice with specific genetic deletion of the NBCe1-A splice variant.

We conclude that NBCe1-B is present in the kidney based upon several lines of evidence. First, there is basolateral immunolabel with an antibody specific to all NBCe1 splice variants in portions of the proximal tubule where an antibody specific to NBCe1-A is negative. Second, there is basolateral immunolabel present in this same region in mice genetically modified to prevent expression of NBCe1-A, and thus the immunolabel does not represent detection of the NBCe1-A splice variant. Third, immunoblot analysis shows expression of a protein with the appropriate molecular weight for NBCe1-B in the kidney of mice that do not express NBCe1-A. Finally, extensive RT-PCR analysis, including amplified product sequencing, shows expression of NBCe1-B mRNA but not mRNA for either NBCe1-C, NBCe1-D, or NBCe1-E in both the WT mouse kidney and in the NBCe1-A KO mouse kidney.

The presence of NBCe1-B in the proximal straight tubule can explain discrepancies between reports of NBCe1-A expression and transport activity. Several previous studies examining mouse, rat, rabbit, and human kidney using NBCe1-A-specific antibodies have shown basolateral immunolabel in cortical proximal tubule segments but not in proximal straight tubule in the outer medulla (17, 33, 39, 51, 61). In contrast, functional NBCe1 activity has routinely been demonstrated in the proximal straight tubule from the outer medulla (4, 19, 26, 40). Thus, the finding of NBCe1-B expression in the proximal tubule, including the proximal straight tubule in the outer stripe of the outer medulla, can explain these differences between NBCe1-A expression and NBCe1 activity. Moreover, the functional activity studies provide evidence that basolateral NBCe1-B in the proximal straight tubule in the outer stripe of the outer medulla is active and contributes to basolateral bicarbonate exit. These findings differ from a report using immunogold electron microscopy that did not find detectable NBCe1 protein in the S3 segment of the rat proximal tubule (39). However, immunogold electron microscopy is less sensitive than is immunohistochemistry for detecting protein expression, which may explain the different findings.

The SLC4A4 gene has five splice variants. An alternative transcription initiation site gives rise to the NBCe1-A and NBCe1-D transcripts and results in a different 5′ region than in NBCe1-B, NBCe1-C, and NBCe1-E. The current studies, by showing presence of a transcript with the 5′ portion present in the B/C/E transcript, thus provide evidence of expression of a non-NBCe1-A transcript. The NBCe1-C transcript differs from NBCe1-B because of alternative splicing that excludes exon 24 (45). The current study sequenced mouse renal mRNA in this region and found no evidence of NBCe1-C expression. Finally, NBCe1-D and NBCe1-E transcripts differ from NBCe1-A and NBCe1-B, respectively, because of a cryptic splice site in exon 6 that deletes 27 bp from the transcribed mRNA (38). Direct sequencing showed no evidence of either the NBCe1-D or NBCe1-E transcript in either normal or NBCe1-A deletion kidneys. Thus, of the five SLC4A4 splice variants, only two, NBCe1-A and NBCe1-B, are expressed in the mouse kidney. The presence of NBCe1-B mRNA in the mouse kidney is consistent with Northern blot findings on human tissues showing expression of a transcript with the 5′ mRNA region that is present in NBCe1-B and NBCe1-C (3). It is also consistent with another study examining rabbit kidney, which identified NBCe1 transcript expression, without differentiating between different splice variants, in the proximal straight tubule in the outer medulla (4), a region where previous studies had not found NBCe1-A protein expression (17, 39, 51, 61). In total, these findings support expression of a non-NBCe1-A transcript, specifically NBCe1-B, in the kidney.

Several previous studies have examined expression of protein from NBCe1 splice variants in the kidney. One report suggested that the B splice variant was present in the rat renal proximal tubule but identified a different subcellular pattern, namely cytoplasmic in addition to basolateral expression (13). However, in our studies using the same antibody, we observed similar labeling in kidneys from WT mice and in those from mice with deletion of all NBCe1 splice variants, indicating that the observed immunolabel was not due to NBCe1-B. Another report using a different anti-NBCe1-B antibody identified apical expression in the proximal tubule (17). It is interesting to note that we also observed subapical proximal tubule immunolabel with an antibody directed against NBCe1-B, but this immunolabel was also present in the kidney of mice with deletion of all NBCe1 splice variants and thus should be considered nonspecific. We are unable to comment further on the reason for the difference between the findings in Ref. 17 and the current study. A third report using an antibody directed against the carboxy-tail that is specific to NBCe1-C identified weak, at least compared with the central nervous system, expression in the rat kidney of a protein with an appropriate molecular weight for NBCe1-C (9). However, mRNA expression was not examined and no immunohistochemistry was shown. The current studies, in contrast, show NBCe1-C mRNA is not present in the mouse kidney. We are unable to comment further on the reason for the different findings from the current study.

Identification that the proximal tubule has basolateral NBCe1-B suggests that a variety of novel mechanisms may regulate basolateral bicarbonate exit and thereby alter transepithelial bicarbonate reabsorption. The inositol (1,4,5)-trisphosphate receptor binding protein released with inositol (1,4,5)-trisphosphate (IRBIT) binds to the amino-terminus NBCe1-B domain that is not present in NBCe1-A and activates transport (55). IRBIT also reduces the affinity for intracellular Mg²⁺-dependent inhibition of NBCe1-B (62), and it may modulate effects of cAMP on NBCe1-B activity (22). Other studies show that IRBIT forms signaling complexes with phosphatidylinositol phosphate kinases that regulate NBCe1-B through phosphatidylinositol 4,5-bisphosphate (8). Intracellular Cl⁻ inhibits NBCe1-B activity through involvement of a low-affinity GXXXP site, and IRBIT activation unmasks a second Cl⁻ inhibitory site (54). Protein phosphatase 1, which is involved in multiple signal transduction pathways, binds NBCe1-B, altering both activity and plasma membrane expression level (36). The IRBIT homolog S-adenosylhomocysteine hydrolase-like 2 binds to NBCe1-B via specific serine residues in amino-terminus and activates NBCe1-B activity (44). Understanding the role of these and other currently unidentified regulators of NBCe1-B activity will be important for future studies examining the regulation of proximal tubule transport.

A third major finding of the current studies is that exogenous acid loading increases NBCe1-B expression in the NBCe1-A KO kidney. This finding is consistent with evidence that metabolic acidosis increases NBCe1-B expression in mouse ameloblast-like LS8 cells (43, 56). NBCe1-B is also regulated by the second messenger molecule, IRBIT, whereas NBCe1-A is not (37, 55, 62, 63). Thus, the presence of and the increased expression of NBCe1-B in response to metabolic acidosis may contribute to the increase in proximal tubule basolateral Na⁺-coupled bicarbonate transport activity during metabolic acidosis (6, 46, 57), which occurs despite no change in total NBCe1 expression (7, 29).

In summary, the current studies provide important new information regarding renal acid-base homeostasis. Immunohistochemistry, immunoblot analysis, and RT-PCR analysis show expression of SLC4A4 splice variant NBCe1-B in the mouse proximal tubule. Metabolic acidosis increases expression, suggesting that NBCe1-B may contribute to the increased filtered bicarbonate reabsorption that occurs in response to metabolic acidosis. Thus, these studies demonstrate the presence of NBCe1-B, a basolateral integral membrane protein, in the proximal tubule, where it likely contributes significantly to renal acid-base and electrolyte homeostasis.

GRANTS

These studies were supported by funds from the the National Institutes of Health (Grant R01-DK045788 to I. D. Weiner and Grant R01-DK107798 to I. D. Weiner and J. W. Verlander). A. N. Harris was supported by funds from NIH Grant 5T32DK104721.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

J.W.V. and I.D.W. conceived and designed research. L.F., H.-W.L., C.C., A.N.H., and J.W.V. performed experiments; H.-W.L., A.N.H., J.W.V., and I.D.W. analyzed data; H.-W.L., A.N.H., M.F.R., J.W.V., and I.D.W. interpreted results of experiments; L.F., H.-W.L., J.W.V., and I.D.W. prepared figures; I.D.W. drafted the manuscript; L.F., H.-W.L., A.N.H., J.W.V., and I.D.W. edited and revised manuscript; L.F., H.-W.L., C.C., A.N.H., M.F.R., J.W.V., and I.D.W. approved final version of manuscript.