NaHCO3 transporters are involved in maintenance of intracellular pH and transepithelial HCO3− movement in many rodent tissues. To establish the human relevance of the many investigations on rodents, this study aimed to map these transporters and a related polypeptide, NaBC1 [solute carrier 4 (SLC4)A11], to several human tissues by using PCR on reverse transcribed human mRNA and immunoperoxidase histochemistry. The mRNA encoding the electroneutral Na+:HCO3− cotransporter (NBCe1; SLC4A4), was expressed in renal cortex, renal medulla, stomach, duodenum, jejunum, ileum, colon, pancreas, choroid plexus, cerebellum, cerebrum, and hippocampus. NBCe2 (SLC4A5) and NBCn1 (SLC4A7) mRNAs were mainly found in kidney and brain tissues, as was mRNA encoding the Na+-dependent anion exchangers NCBE (SLC4A10) and NDCBE1 (SLC4A8). In addition to previous findings, NBCn1 protein was localized to human renal medullary thick ascending limbs and duodenal epithelial villus cells and NBCe2 protein to renal collecting ducts. Finally, the message encoding NaBC1 was found in kidney, stomach, duodenum, pancreas, and brain, and the corresponding protein in the anterior and posterior corneal epithelia, renal corpuscules, proximal tubules, collecting ducts, pancreatic ducts, and the choroid plexus epithelium. In conclusion, the selected human tissues display distinct expression patterns of HCO3− transporters, which closely resemble that of rodent tissues.
- solute carrier 4A
- acid/base transporters
- bicarbonate metabolism
- sodium transport
- chloride transport
the movement of HCO3− across the plasma membranes of living cells is crucial for maintaining a suitable internal milieu in and around the many different cells of higher organisms. Regulated HCO3− transport is necessary for such diverse functions as cellular pH regulation, cell division, migration, transepithelial HCO3−, Cl−, and/or Na+ movement (i.e., absorption and secretion), and acid-base balance of the whole organism. For these purposes, a number of epithelial and nonepithelial tissues express HCO3− transporters each with specialized tasks depending on the tissue localization and the transport mode of the given protein.
A substantial proportion of cellular HCO3− transport is mediated by proteins belonging to the solute carrier 4 (SLC4) family consisting of the Na+-independent Cl−/HCO3− exchangers (11, 12, 33), the electrogenic and electroneutral Na+:HCO3− cotransporters (NBCs) (5, 20, 21, 24), and the Na+-dependent Cl−/HCO3− exchangers, NDCBE1 (8), and NCBE (30). Most of these proteins have been extensively studied in tissues of both epithelial and nonepithelial origin. A great deal of attention has been given to studies of renal tubular HCO3− transport because of the central role of these structures in normal regulation of the acid-base balance and in some pathological states (for a review, see Ref. 23). The HCO3− transport of the epithelium in the stomach, duodenum, and pancreas has also been studied widely because the proper function of these tissues depends on high HCO3− output. Finally, the NaHCO3 transporters expressed in the brain and choroid plexus are believed to be involved in sustaining normal brain function through their effect on cellular and perhaps extracellular pH (4, 6). Thus the kidneys, the gastrointestinal tract, as well as the central nervous system have been prominent examples of tissues, where NaHCO3 transporters have been linked to important organ functions in rodents, and in many cases, the relevance for human biology remains obscure.
Another Na+-dependent transporter from the gene family is the NaBC1, a Na+:B(OH)4− transporter (16). Mutations in the human NaBC1 (formerly BTR1) were recently shown to be associated with recessive congenital hereditary endothelial dystrophy, and NaBC1 mRNA was detected in the human cornea (28). To date, NaBC1 protein has not previously been localized to any human tissues, and information on mammalian NDCBE1 protein expression is also lacking. Therefore, we investigated the expression patterns of the Na+-dependent transporters NBCe1 (SLC4A4), NBCe2 (SLC4A5), NBCn1 (SLC4A7), NCBE (SLC4A10), NDCBE1 (SLC4A8), and NaBC1 (SLC4A11) at the mRNA and, where possible, at the protein level in the human kidney, gastrointestinal tract, and brain as a first step toward identifying the similarities and differences in human and rodent HCO3− transport physiology.
Human DNAse-treated total RNA from renal cortex and medulla, stomach, duodenum, jejunum, ileum, colon, pancreas, choroid plexus, hippocampus, cerebrum, and cerebellum was purchased from Tebu-bio (Roskilde, Denmark). Rat RNA was isolated using RNeasy Mini Kits (Qiagen, Germantown, MD) or, for microisolated kidney tubules, isolated using Dynabeads mRNA Direct micro-kit (Dynal, Oslo, Norway) and immediately DNAse-treated (RQ DNaseI, Promega, Madison, WI). Renal structures were microisolated after digestion at 37°C in conical flasks containing 1 mg/ml collagenase A (Sigma, St. Louis, MO) and 2 mg/ml dispase (Invitrogen, Taastrup, Denmark) in oxygenated K+/gluconate solution (in mM: 10.0 Na+, 140.0 K+, 10.0 Ca2+, 1.0 Mg2+, 32.0 Cl−, 140.0 gluconate−, 10.0 HEPES, 10.0 sucrose, at pH 7.4) swirled at 80 rpm.
The RNA was reverse transcribed using 2 U/μl RT (Superscript II, Invitrogen). PCR (HotStarTaq Master Mix, Qiagen) with 10% cDNA and 1 pmol of each primer (human primers: Table 1, rat primers not shown) was performed for 30 cycles after 15 min at 95°C: 95°C, after which denaturation was performed for 30 s at 60°C, annealing for 30 s, and elongation at 72°C for 1 min. Negative PCR controls included omission of RT or omission of cDNA. PCR for β-actin was performed to validate each template. PCR products were separated by 2% agarose gel electrophoresis with ethidium bromide and photographed under ultraviolet illumination. Isolated renal structures were verified by RT-PCR for segment-specific markers [NBCe1 for proximal tubules, AQP1 for proximal tubules and thin descending limbs, ClC-2 for thin ascending limbs (pooled with thin descending limbs), distal tubules, NKCC2 for medullary thick ascending limbs (TALs) and cortical TALs, NCC for distal convoluted tubules, and AQP2 for collecting ducts]. All human and rat primer pairs were validated by nucleotide sequencing representative PCR products.
Human tissues were obtained post mortem from donated tissue according to the Danish guidelines for use of human material and immersion-fixed with 4% formaldehyde in isotonic PBS, pH 7.4. Rat tissues were from adult male Wistar rats (250–400 g, Taconic, Ry, Denmark). The animals were anesthetized by isoflurane inhalation before removing various organs and killing the animals. The tissues were dehydrated and embedded in paraffin wax, and 2-μm sections were cut using a rotary microtome (Leica, Heidelberg, Germany). The sections were dewaxed and rehydrated, and endogenous peroxidase was blocked by 0.5% H2O2 in absolute methanol. Epitopes were retrieved by boiling the sections in 10 mM Tris, pH 9, supplemented with 0.5 mM EGTA, and then quenched with 50 mM NH4Cl. Unspecific binding was blocked by 1% BSA in PBS with 0.05% saponin and 0.2% gelatin. The sections were incubated overnight at 4°C with the primary antibodies diluted in PBS with 0.1% BSA and 0.3% Triton X-100. Omission of primary antibody or, when possible, peptide preabsorption tests were run as negative controls.
The sections were incubated with horseradish peroxidase-conjugated goat anti-rabbit IgG (DAKO, Glostrup, Denmark) in PBS with BSA and Triton X-100. The staining was visualized by 0.05% 3,3′ diaminobenzidine tetrahydrochloride dissolved in PBS with 0.1% H2O2. Mayer's hematoxylin was used for counterstaining, and the sections were dehydrated in graded alcohol and xylene and mounted in hydrophobic Eukitt mounting medium (O. Kindler, Freiburg, Germany). Microscopy was performed on a Leica DMRE bright-field microscope equipped with a Leica DM300 digital camera.
The protein content of rat kidney or brain homogenates was determined using a bicinchoninic acid protein assay reagent kit (Pierce, Rockford, IL) after clearing the suspension of nuclei and unbroken cells by 4,000 g centrifugation. Protein samples were adjusted to 1.5% (wt/vol) sodium dodecyl sulfate, 40.0 mM 1,4-dithiothreitol, 6% (vol/vol) glycerol, 10 mM Tris(hydroxymethyl) aminomethane, pH 6.8, and added bromophenol blue. The samples were heated to 65°C for 15 min and stored at −20°C until use. Protein samples of 2–10 μg were separated by 9% polyacrylamide gel electrophoresis. Proteins were then electrotransferred onto nitrocellulose membranes, which were blocked by incubation in 5% nonfat dry milk in a Tween-containing PBS (PBS-T, PBS with 0.1% vol/vol Tween 20). The membranes were incubated with primary antibody overnight at 5°C in PBS-T. After washing, we incubated the membranes with horseradish peroxidase-conjugated goat anti-rabbit IgG secondary antibody (DAKO) for 2 h in PBS-T. Excess antibody was then removed by extensive washing, and bound antibody was detected by ECL chemiluminiscence kit (Amersham, Little Chalfont, Buckinghamshire, UK).
NaBC1 cloning and transfection.
Full-length NaBC1 was obtained by 3′ RACE (rapid amplification of cDNA ends), 5′ RACE and according to manufacturer's protocols (BD SMART RACE cDNA amplification kit; BD Biosciences Clontech, Palo Alto, CA, USA). Full-length NaBC1 constructs were obtained by PCR using 5′ and 3′ UTR primers, and after separation, products were extracted from agarose gels (Gel extraction kit, Qiagen). After ligation into a topoisomerase I-activated pcDNA™3.1/V5-His-TOPO vector (Invitrogen), the resulting plasmids were used for transformation of competent “One Shot Top10” E. coli (Invitrogen). Colonies were grown with ampicillin selection and NaBC1-positive colonies were multiplied in mini- and midi-preps (Qiagen). The resulting plasmids were purified using Qiaprep (Qiagen), and the inserts were verified by sequencing. Errors due to the Taq polymerase-based RACE reactions were corrected using site-directed mutagenesis with PFU Turbo polymerase (Invitrogen).
Human embryonic kidney (HEK)293 cells (European Collection of Cell Cultures ECACC, Porton Down, Salisbury, Wiltshire, UK) were grown on coverslips in DMEM supplemented with 10% (vol/vol) FCS at 37°C in 5% CO2. After 3 or 4 passages, 50% confluent cells were transiently transfected with NaBC1 plasmids by lipofection technique (Effectene Transfection Reagent, Qiagen). After 72 h, cells were fixed and stained as described below (immunocytochemistry).
The cells were fixed for 10 min in 2% paraformaldehyde 72 h after transfection, and rinsed twice in PBS (280 mM Na+, 80 mM HPO42−, 20 mM H2PO4−, 100 mM Cl−, pH 7.5). After permeabilization and blocking in PBS with 0.1% Triton and 1% BSA for 30 min (PBS-TA), cells were incubated overnight at 4°C with primary antibodies diluted in PBS-TA and washed and incubated for 1 h at 21°C with Alexa488-conjugated goat anti-rabbit secondary antibodies and TO-PRO3 nuclear counterstain (both Molecular Probes, Eugene, OR). After three final washes with PBS, the coverslips were mounted with glycergel (DAKO). Cells were inspected on an inverted Leica DMRS confocal microscope using an HCX PlApo 64 × (1.32 NA) objective. The immunofluorescence images were merged with differential interference contrast images.
All antibodies were raised in rabbits, and the following were previously validated for use with rodent and human tissues: SLC4A4-NBCe1 (15), SLC4A5-NBCe2 [73 c-terminal amino acids of human NBCe2, (4)], SLC4A7-ctNBCn1 (29), -ntNBCn1 (7), and -ctNBC3 (22), as well as SLC4A10-NCBE (18, 19). The novel anti-NDCBE1 (SLC4A8) and anti-NaBC1 (SLC4A11) antibodies were shown to bind the respective immunizing peptides, bind to proteins of expected molecular size, and for NaBC1 to label cells transiently transfected with full-length rat NaBC1 (⇓Fig. 2). Table 2 compares the amino acid composition of the immunizing peptides to the corresponding human sequences.
Detection of the SLC4A-derived mRNA in human tissues.
Total RNA from a panel of human tissues was reverse transcribed, and the resulting cDNA was subjected to PCR analysis for transcripts arising from the SLC4 gene family to determine the tissue expression profile of the Na+-dependent members. As shown in Fig. 1A, mRNA signal for NBCe1 (SLC4A4) was strong in kidneys, duodenum, ileum, pancreas, and in all four tissues from the CNS, although it was weak in the remaining gastrointestinal tissues. Message for NBCe2 (SLC4A5) was detected mainly in the kidneys and CNS (Fig. 1B), whereas the small intestine and pancreas displayed little signal. Only the stomach, colon, and pancreas were negative for NBCe2 mRNA. Fig. 1C shows that expression of NBCn1 (SLC4A7) mRNA was detected in all tested tissues, with little signal in the stomach, colon, and pancreas. NDCBE1 (SLC4A8) mRNA was consistently observed in the renal medulla and all four tissues of the CNS, whereas weaker signal was detected in renal cortex and throughout the gastrointestinal tract (Fig. 1D). The mRNA encoding NCBE (SLC4A10) displayed a more restricted expression pattern, as illustrated in Fig. 1E. Robust signal was observed in the choroid plexus, hippocampus, and cerebral cortex, whereas weaker signal could be observed in the renal cortex, stomach, duodenum, ileum, and cerebellum. NaBC1 (SLC4A11) mRNA was found in kidneys, stomach, duodenum, pancreas, choroid plexus, hippocampus, and cerebral cortex (Fig. 1F). The signal was least convincing in stomach, small intestine, pancreas, and cerebellum.
The specificity of each PCR product was verified by sequence analysis revealing 100% identity between the renal products and the corresponding human cDNA sequence according to GenBank. Both human NBCe2 products matched variants “a” and “c” of human NBCe2, the shorter product being a truncated version of NBCe2 “a” or “c”. The NDCBE1 band corresponds to both variant 1 and 2 of human NDCBE1. The two NCBE bands in rat brain both represented NCBE forms.
Validation of anti-NaBC1 and anti-NDCBE1 antibodies.
The recognition of the NaBC1 peptide used for rabbit immunization by the corresponding antiserum was secured by subjecting the peptide to SDS-PAGE and subsequent immunoblotting with anti-NaBC1 antiserum (Fig. 2A). HEK293 cells were transiently transfected with rat NaBC1 and immunocytochemically stained with the affinity-purified antibody to demonstrate its recognition of the full-length NaBC1 protein (Fig. 2B). Binding of affinity-purified anti-NaBC1 to NaBC1 protein in transfected cells is indicated by the green fluorescence signal of the secondary antibody. A red counterstain was used to visualize the nuclei, and the fluorescence images were merged with the corresponding differential interference contrast images. Only around 20% of the cells were successfully transfected. Omission of primary antibody from the labeling procedure prevented labeling of transfected cells. The NaBC1 protein seemed largely retained intracellularly in transfected cells. Normal targeting of membrane proteins cannot be expected from transient overexpression systems. Therefore, immunoblot analysis was done on renal homogenates enriched in plasma membranes (17,000 g pellets) or vesicles (200,000 g pellets), as illustrated in Fig. 2C. The NaBC1 immunoreactivity was more predominant in the plasma membrane-enriched samples, although bands were also detectable in the vesicle fraction of medullary homogenates. The renal immunoreactivity was not detectable when the anti-NaBC1 was preabsorbed by the immunizing peptide (Fig. 2D). Similar results were obtained with affinity-purified antibodies against the same epitope from a second rabbit (not shown). The NDCBE1 peptide used for rabbit immunization was recognized by the anti-NDCBE1 serum by SDS-PAGE and subsequent immunoblotting, as shown in Fig. 2E. The affinity-purified anti-NDCBE1 antibody recognized polypeptides of the expected 130 kDa by immunoblot analysis of 17,000 g pellets from rat cerebral and cerebellar homogenates (Fig. 2F). The right panel shows the successful preabsorption of the anti-NDCBE1 by the immunizing peptide. Similar results were obtained with affinity-purified antibodies against another NDCBE1 epitope (not shown).
Immunolocalization of NBCe1, NBCe2, NBCn1, and NaBC1 to human renal structures.
An immunohistochemical approach was applied to verify the above molecular biological findings and to determine the tissue, cellular, and subcellular distribution of the transporters in human kidneys, duodenum, jejunum, pancreas, and choroid plexus. NBCe1 has previously been localized to the basolateral plasma membrane domain of renal proximal tubules and pancreatic ducts, which could be confirmed in the current study (not shown). Fig. 3A illustrates that anti-NBCe2 labeling was observed corresponding to the apical plasma membrane domain of a subset of collecting duct cells in the renal medulla. Identical labeling was produced with two anti-NBCe2 antibodies directed against other epitopes (not shown). The staining was prevented by preabsorption with the immunizing peptide (Fig. 3A, inset). A similar pattern was shown in the renal cortex, and immuofluorescence double labeling with AQP2 in both human and rat kidney revealed that the NBCe2 antibodies labeled the intercalated cells (not shown).
Three anti-NBCn1 antibodies were employed for mapping NBCn1/NBC3 to human renal structures. The antibody against a human and rat NH2 terminal, ntNCBn1, localized NBCn1 to medullary TALs, as shown in Fig. 3B. Figure 3B, inset illustrates the lack of immunostaining upon preabsorption of antibody with the immunizing peptide. The antiserum directed to the human COOH terminal, ctNBC3 revealed similar labeling to the anti NH2 terminal antibody (Fig. 3C), but an antiserum against the COOH terminal of rat NBCn1, ctNBCn1, failed to produce convincing staining (cytosol of intercalated cells, not shown). The sensitivity and specificity of the novel anti-NaBC1 antibody were studied, as it has not previously been used for publication (Fig. 2, A–D). Anti-NaBC1 immunoreactivity was observed in the podocytes of renal corpuscules and the brush border of proximal tubules, as illustrated in Fig. 3D. In the cortex and outer medulla, anti-NaBC1 staining was observed corresponding to apical membrane domains of intercalated cells (not shown) and to the basolateral membranes in inner medullary collecting ducts (Fig. 3E). A similar labeling pattern was observed in rat kidney (Fig. 6, B–D). Peptide preabsorption of the antibody prevented labeling of all structures (Fig. 3, B, D, and E, insets). Fig. 3F summarizes the renal expression of Na+-dependent transporters of the SLC4A gene family.
Immunolocalization of NBCn1 and NBCe1 in the human duodenum.
Application of all of the three anti-NBCn1 antibodies resulted in staining of the basolateral plasma membrane domains of duodenal epithelial cells of both villi and crypts (Fig. 4A: ctNBC3, Fig. 4B: ntNCBn1, and Fig. 4C: ctNBCn1, respectively). Figure 4, B–E, insets show that staining was prevented by binding the antibodies to the corresponding immunizing peptides before the application (peptide was not available for the anti-human COOH-terminal antibody). Anti-NBCe1 antibodies against a common epitope of the renal and duodenal/pancreatic NBCe1 labeled the basolateral membrane domain of duodenal epithelial villus cells, as shown in Fig. 4D. This staining was also prevented by peptide preabsorption (inset).
Immunolocalization of NaBC1 to the human pancreatic ducts.
Figure 4E illustrates that anti-NaBC1 labeling was found in the ductal structures, the intercalated ducts, of the human pancreas. The labeling was sensitive to peptide preabsorption (Fig. 4, B–E, insets). In the pancreas, anti-NBCe1 immunoreactivity was restricted to intercalated duct, as has previously been reported (not shown).
Immunolocalization of NDCBE1 and NaBC1 to the human brain structures.
NaBC1 labeling was restricted to the apical membrane domain and the brush border of choroid plexus epithelial cells, as well as to the capillary endothelia (Fig. 5A). The inset shows that the labeling was prevented by preabsorbing the antibody by the corresponding immunizing peptide. Figure 5B illustrates the pyramidal cell localization of anti-NDCBE1 immunoreactivity in human hippocampus.
Immunolocalization of NaBC1 to the human eye.
Figure 5C shows NaBC1 immunoreactivity corresponding to the posterior corneal epithelium (also called corneal endothelia), and in Fig. 5D, also to the anterior corneal epithelium. The anterior epithelial staining is more restricted to cell boundaries than the posterior epithelial staining, which is most predominant intracellularly. The anti-NaBC1 labeling was prevented by peptide preabsorption (Fig. 5, A, C, and D, insets). Similar staining was found in rat cornea (Fig. 6F).
RT-PCR analysis on renal structures and immunolocalization of NaBC1 in rat tissues.
Figure 6A exemplifies the detection of NaBC1 mRNA in isolated renal structures by RT-PCR. NaBC1 was found in glomeruli, proximal tubules, thin limbs, and collecting ducts in repeated isolates. This renal distribution was supported by the NaBC1 immunoreactivity in rat glomeruli, proximal tubular brush border (Fig. 6B), thin descending limbs (Fig. 6C), and inner medullary collecting ducts (Fig. 6D). Immunoreactivity was also observed corresponding to a nerve adjacent to an afferent artery and peripheral nerves (Fig. 6, B and E, respectively). NaBC1 mRNA was also demonstrated in rat cornea (Fig. 6A), and the corresponding immunolabeling was found in rat cornea, as shown in Fig. 6F.
RT-PCR analysis on rat tissues and immunolocalization of NDCBE1 in rat brain.
Figure 7A exemplifies the detection NDCBE1 mRNA in various rat tissues by RT-PCR. NDCBE1 was found in kidney inner medulla, the submandibular gland, eye, cerebrum, cerebellum, as well as in testis. Figure 7B shows the detection of NDCBE1 immunoreactivity in Purkinje cells and dendrites in the molecular layer of rat cerebellum. ML is the molecular layer, and GrL is the granular layer. Figure 7B, inset shows that the labeling was prevented by preabsorbing the antibody by the corresponding immunizing peptide.
Very often, rodents are used in acid-base physiological investigations as a convenient model for human biology and pathophysiology. Although the expression of some acid-base transporters has already been studied in humans (1, 3, 14), it is unknown to what extent the human expression pattern of many of these transporters resembles that of rodents.
NBCe1 (SLC4A4) and NBCe2 (SLC4A5) are both capable of transporting 2–3 HCO3− for each Na+ (9, 10, 24, 27). The direction of transport differs among the tissues, so that, for example, NBCe1 in renal proximal tubules extrudes these ions 3:1, while the pancreatic duct variant takes the ions into the cell 2:1. The tissue distribution of NBCe1 in humans is quite well described. The renal NBCe1 variant is situated in the basolateral plasma membrane domain of proximal tubules (32), and the pancreatic NBCe1 form is also basolateral in the human pancreatic ducts (14, 26). The pancreatic pNBC was reported to be expressed at lower levels in other organs, whereas kNBC is expressed uniquely in kidney (1). We can confirm the above localization in human kidney and pancreas, and in addition, we found NBCe1 immunoreactivity corresponding to the basolateral plasma membrane of duodenal enterocytes. Thus, the human NBCe1 displays an identical expression pattern to rodent NBCe1 and is very likely involved in transepithelial HCO3− movement at these sites.
Human NBCe2/NBC4 mRNA has been detected in liver, testis, spleen, and heart (21), as well as in brain, pancreas, muscle, and kidney (25). Xu and coworkers (31) reported NBC4 mRNA expression in TALs of the human kidney. The exact protein localization is fairly undecided even in rodents. Two reports have mapped NBCe2 protein to rat urothelia, liver (2), and choroid plexus (4). Here, we report a single novel tubular expression site for NBCe2 protein in human kidney, as immunostaining outside the kidney was not preabsorbed by the immunizing peptide (apical in choroid plexus epithelium, pancreatic ducts, and duodenal villus cells, not shown). In the collecting duct, NBCe2 immunoreactivity was found in luminal plasma membranes. This was also observed in rat kidneys. Thus, NBCe2 seems predominantly to be a luminal transport protein in contrast to the other electrogenic transporters, basolateral NBCe1. Thus, although the two transporters display very similar transport properties, these proteins must serve quite distinct roles in transepithelial bicarbonate transport.
For the electroneutral Na+:HCO3− cotransporter, NBCn1/NBC3 (SLC4A7), a low cellular Na+ concentration is usually sufficient to drive inward directed ionic transport (3, 5). NBCn1 is most likely a cellular base loader or “acid-extruder” like the NHEs. Actually, NBC3 was first localized in human collecting ducts by application of the ctNBC3 antibody (22). Later, Vorum and coworkers (29) localized NBCn1 in rat mTAL using the anti-ctNBCn1 antibody (29). In the choroid plexus epithelium, both humans and rats express NBC3/NBCn1 in the basolateral plasma membrane (18, 19). In the present study of the human kidney, we found basolateral labeling of medullary TALs with both the ntNBCn1 and ctNBC3 antibodies for the first time, whereas the ctNBCn1 did not label any membranes of renal structures. In contrast, all three antibodies clearly labeled the basolateral plasma membrane domain of human duodenal enterocytes. We believe our current unmasking procedure has exposed the epitope more efficiently than previously and allows the NBC recognition of this protein in human TAL. The lack of ctNBCn1 binding in TAL may indicate that the protein abundance is not sufficient to allow detection by this particular antibody. The antibodies are unlikely to display equal affinity toward the target proteins. The apparent discrepancy in immunoreactivity using the three antibodies could also be caused by differential binding to accessory proteins, organ-specific variable splicing, or demasking/unfolding of NBCn1 in the tissues, although this has never been proven to be the case. Hence, although some progress has been made recently in both man and rodents, the complete tissue distribution of NBCn1 remains to be uncovered. The A-intercalated cells may represent a site of diverging NBCn1 expression between humans and rodents.
The Na+-dependent Cl−/HCO3− exchangers are, like NBCn1, transporting in an electroneutral manner (8, 30). The equivalent of one Na+ and two HCO3− ions are believed to be taken up by the cells in exchange for each intracellular Cl−. The mRNA encoding NCBE (SLC4A10) was initially found in rat brain, pituitary, testis, kidney, and ileum (30). Later, NCBE was found to be expressed in high abundance in rodent choroid plexus, in which the protein was situated in the basolateral plasma membrane (18). NCBE has also previously been localized to human choroid plexus (19), and in the current study, the protein was not detected elsewhere, although mRNA was also readily found in, for example, the hippocampus. Thus, it is not possible at present to expand the cellular localization of NCBE in human tissues, and therefore, it is impossible to identify any differences in expression profile between humans and rodents.
The anti-NDCBE1 antibody was validated by peptide recognition, by binding a single brain plasma membrane protein of the expected molecular size, and by peptide preabsorption, as the antibody has not previously been used for publication. Human NDCBE1 (SLC4A8) mRNA was found in several parts of the human brain, testis, kidney, and ovary (8). In the current report, the tissue expression was expanded to include intestinal tissues, although the expression level may be lower than in brain and kidney. The NDCBE1 protein has never before been localized to any specific structure in either rodent or human tissues. Presently, anti-NDCBE1 immunoreactivity was only convincingly associated with pyramidal cells of the human hippocampus. Similar labeling—as well as staining of the functionally related Purkinje cells—was also observed in rat brain. Staining was seen in endothelia in kidneys and the choroid plexus. Again, a similar staining pattern was found in rat, but the expression at these specific sites was not corroborated by RT-PCR. It will be interesting to learn whether a similar expression profile is observed in rodents and what functional implication this may have.
The only functional report to date on NaBC1 (SLC4A11) has classified the protein as a sodium borate cotransporter (16). The mRNA encoding this atypical SLC4A-derived protein is found in kidney, salivary glands, testis, thyroid glands, and trachea (17). Interestingly, NaBC1 is also expressed in the eye, and mutations in the human NaBC1 gene cause recessive congenital hereditary endothelial dystrophy (28). To ascertain the quality of the antibody, anti-NaBC1 was subjected to peptide blotting, immunocytochemical analysis on NaBC1-transfected culture cells, and immunoblotting. Anti-NaBC1 bound both the immunizing peptide, as well as a single plasma membrane polypeptide of the expected 100 kDa from rat kidneys (and other tissues), only when it was not preabsorbed by the immunizing peptide. In the present study, the expected corneal endothelial NaBC1 immunoreactivity was demonstrated for the first time. NaBC1 labeling was, however, not restricted to this tissue, as similar staining was evident also in the corneal epithelium, in vascular endothelia, as well as secretory and in some absorptive epithelia. The broad apparent renal NaBC1 protein expression calls for special caution. Thus, it was crucial for our interpretation that the human renal NaBC1 distribution matched that of the rat kidney, and the latter was confirmed by RT-PCR analysis on validated microisolated renal structures.
Most of the NaBC1-expressing epithelia are known to display vectorial NaHCO3 transport at high rates, but their role in borate transport remains elusive. It will be interesting to learn whether the functional characterization of NaBC1 can be verified and perhaps expanded by other investigators, taking the low borate concentration in body fluids into consideration.
In conclusion, the human distribution of most Na+-dependent transporters of the SLC4 gene family was very similar to that of rodent tissues. Thus, it seems that past investigations of rodent HCO3− transporters may be highly relevant for human physiology. The endothelial and epithelial expression profiles of NDCBE1 and NaBC1 call for further studies of the significance of the transport proteins at these sites.
The Water and Salt Research Center at the University of Aarhus is established and supported by the Danish National Research Foundation (Danmarks Grundforskningsfond). The work was also supported by the Novo Nordisk Foundation, the Danish Medical Research Council, and by the University of Aarhus Research Foundation.
We wish to thank Inger Merete S. Paulsen, Mette F. Vistisen, Helle Hoyer, and Christian V. Westberg for expert technical assistance. The anti-NBCe2 antibody was kindly provided by W.F. Boron (Department of Cellular and Molecular Physiology, Yale University School of Medicine, New Haven, CT). We are grateful to Ira Kurtz (Division of Nephrology, School of Medicine, University of California at Los Angeles, CA) for supplying the NBC3c1 antibody (ctNBC3). Human hippocampus was a gift from Carsten Bjarkam, Institute of Anatomy, University of Aarhus, Aarhus, Denmark.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2007 the American Physiological Society