Na+-dependent Cl−/HCO exchange activity helps maintain intracellular pH (pHi) homeostasis in many invertebrate and vertebrate cell types. Our laboratory cloned and characterized a Na+-dependent Cl−/HCO exchanger (NDAE1) fromDrosophila melanogaster (Romero MF, Henry D, Nelson S, Harte PJ, and Sciortino CM. J Biol Chem 275: 24552–24559, 2000). In the present study we used immunohistochemical and Western blot techniques to characterize the developmental expression, subcellular localization, and tissue distribution of NDAE1 protein in D. melanogaster. We have shown that a polyclonal antibody raised against the NH2terminus of NDAE1 (αCWR57) recognizes NDAE1 electrophysiologically characterized in Xenopus oocytes. Moreover, our results begin to delineate the NDAE1 topology, i.e., both the NH2and COOH termini are intracellular. NDAE1 is expressed throughoutDrosophila development in the central and peripheral nervous systems, sensilla, and the alimentary tract (Malpighian tubules, gut, and salivary glands). Coimmunolabeling of larval tissues with NDAE1 antibody and a monoclonal antibody to the Na+-K+-ATPase α-subunit revealed that the majority of NDAE1 is located at the basolateral membranes of Malpighian tubule cells. These results suggest that NDAE1 may be a key pHi regulatory protein and may contribute to basolateral ion transport in epithelia and nervous system of Drosophila.
- fusion proteins
sodium-dependentCl−/HCO exchange regulates intracellular pH (pHi) in a variety of invertebrate and vertebrate cell types. First characterized in the squid giant axon (50), snail neurons (59), and barnacle muscle fibers (8), this acid-extruding mechanism was shortly thereafter functionally characterized in vertebrate cells:Necturus proximal tubule (18) and mammalian fibroblasts (26). Subsequently, Na+-dependent Cl−/HCO exchange activity has been characterized in many invertebrate and vertebrate cell types including rat mesangial cells (9, 16), the basolateral membranes of mammalian proximal tubules (4, 35), Chinese hamster ovary cells (11), vascular smooth muscle cells (22), chicken myocytes (31), lens (15), intestinal cells (40), nervous system (51, 55), lymphocytes (58), and mammary tumors (28). In all these cell types, Na+-dependent Cl−/HCO exchange helps to maintain pHi homeostasis.
We recently reported the cloning and characterization of a Na+-driven anion exchanger, NDAE1, from Drosophila melanogaster (47). NDAE1 is the first recombinant protein functionally capable of Na+-dependent Cl−/HCO exchange and functions as a Na+-driven anion exchanger in the nominal absence of HCO (i.e., solutions bubbled with 100% O2) (53). NDAE1 is a member of the bicarbonate transporter superfamily (BTS) (44), which includes the anion exchangers (AEs) (3, 25, 32), electrogenic/electroneutral Na+-bicarbonate cotransporters (NBCs) (1, 10, 41a, 42, 46), and a recently described mammalian Na+- and Cl−-dependent HCO transporter (NCBE; GenBank accession no. AB033759) (63). These proteins share 30–40% amino acid homology and predicted gross topology based on sequence alignments and hydropathy analysis, respectively [reviewed by Romero and Boron (44)]. N-linked glycosylation mutagenesis studies on the transmembrane domains of AE1 supports a 12-transmembrane-spanning model for AE (for review see Ref. 61). Additionally, subcellular immunolocalization studies with polyclonal (29) and monoclonal (62) antibodies indicate that the COOH terminus of AE1 is intracellular. Similarly, the AE1 and AE2 COOH termini interact with the cytosolic enzyme carbonic anhydrase II (60). However, structural analysis of NDAE1, the NBCs, and NCBE remains unexplored.
Drosophila has been widely used as a model for studies of development, molecular, and population genetics. However, much less is known about Drosophila physiology. Insect Malpighian tubules (MTs) have been used as a model system of fluid secretion by epithelia (6, 12, 33, 34). The MT is a brush-border epithelium made up of principal cells and secondary stellate cells (57), which together regulate the osmolarity of hemolymph (36, 37, 64). The physiology of insect MT transport has been studied in a variety of species, and a general transport model has been proposed to which Drosophila (D. melanogaster and D. hydei) generally adhere (12,14, 30). The MT and the gut of insects have many functional similarities to the mammalian kidney.
The rates of ionic and fluid transport across MTs are among the highest known (33) and are driven primarily across principal cells by an apical bafilomycin A1-sensitive, V-type H+-ATPase (21, 24). Proton secretion, in concert with K+/H+ exchange, drives apical K+ secretion (24), whereas Cl−secretion seems to occur mainly via Cl− channels in stellate cells (39). Less is known about the movement of ions across the basolateral membrane of principal cells. The ouabain-sensitive Na+-K+-ATPase has been both immunolocalized and functionally characterized at the basolateral membrane of MTs (21, 30). Inhibitor studies using bumetanide suggest that a Na+-K+-2Cl− cotransporter may mediate basolateral ion flux in some species (20, 38), although not in D. melanogaster (14). A model proposed by Linton and O'Donnell (30) couples outward movement of K+ via potassium channels with inwardly directed (dihydroindenyl)oxy alkanoic acid-inhibitable KCl transport and an unknown Na+/solute transporter (30). In concert with the Na+-K+-ATPase, these transporters drive basolateral flux of K+ and Na+.
Control of systemic and intracellular pH is important for nutrient absorption and insect viability (19) as well as being a potential signaling mechanism for gametogenesis and infiltration of malaria parasites into insect vector species (i.e., mosquitoes) (56). However, study of pHiregulation in invertebrates has been mainly with squid giant axon (49), snail neurons (59), and barnacle muscle fiber (8), all of which contain Na+-dependent Cl−/HCO exchange activity. pHi studies on the gut and MTs have focused on the apical H+ pump, with experiments performed in the nominal absence of HCO (5).
To initiate the understanding of the role of NDAE1 in D. melanogaster physiology, we developed an NDAE1 antibody to characterize the structural properties and subcellular localization of the endogenous protein. Using this antibody and a monoclonal antibody to the Na+-K+-ATPase α5-subunit (27), we have characterized the tissue and membrane distribution of NDAE1 protein in developing D. melanogaster. Structurally, we have demonstrated that both the NH2 and COOH termini of NDAE1 are located intracellularly as predicted by hydropathy analysis.
Portions of this work have been presented in preliminary form (52).
Polyclonal Antibody Production and Affinity Purification
Recombinant glutathione S-transferase (GST)-NDAE1 fusion protein was used as an immunogen for the production of rabbit polyclonal antibodies by following standard production protocols (Cocalico Biologicals, Reamstown, PA). PCR primers designed to add 5′EcoRI (5′-GGGGAATTCATGG CCGAAAAGAATGAG-3′) and 3′XhoI (5′-CCCCTCGAGCTCCACATCCTCCTCGAA-3′) restriction enzyme sites onto the DNA sequence encoding to the first 100 amino acids of NDAE1 were used to amplify a 300-bp NH-terminal fragment (NTERM100) from NDAE1-pTLN cDNA. The amplified fragment was subcloned into the pGEX-4T.1 (GST) expression vector (Amersham Pharmacia, Piscataway, NJ). Sequence was verified by automated sequencing (Cleveland Genomics, Cleveland, OH). DH5α bacteria were transformed with the NTERM100-pGEX-4T.1 construct. Bacteria were grown to an optical density (at a wavelength of 600 nm) of 0.4 when GST-fusion protein production was induced for 3 h by the addition of 0.4 mM isopropyl β-d-thiogalactopyranoside. The GST-fusion protein was isolated using glutathione agarose beads (Sigma, St. Louis, MO). Briefly, bacteria were lysed by snap freeze/thaw, vortexing, and sonication in PBST (PBS plus 0.1% Triton X-100) and 10 μl/ml of bacterial protease cocktail (P-8465; Sigma), followed by overnight incubation with glutathione agarose beads at 4°C. The fusion protein was eluted by the addition of 40 mM reduced glutathione for 1 h, which was subsequently removed by centrifugal filtration using an Amicon filter system (Millipore, Bedford, MA). Fusion protein was stored at 4°C in 10 mM Tris, pH 8.0, and 0.02% NaN3until use.
To affinity purify antibodies specific to NDAE1 epitopes, GST-NTERM100 was run on a single-lane 10% SDS gel and Western blotted as described (see Western Blot Analysis). A band of nitrocellulose membrane containing the fusion protein was cut into small pieces and incubated in PBS plus 5% BSA for 2 h at room temperature (RT). To bind the NDAE1 antibody, we incubated the membrane pieces in 5 ml of serum containing the target antibody overnight at 4°C. Subsequently, the membrane pieces were washed with PBS, and the antibody was eluted with 0.2 M glycine (pH 2.2) for 2 min. The pH was then quickly titrated to 7.4 with 1 M Tris base. Affinity-purified antibodies were stored in 50% glycerol with 0.1% BSA at −20°C until use.
Tissue Membrane Preparations
Oregon-R strain D. melanogaster membrane preparations were prepared using a modified method of Wilcox (66). Briefly, ∼150 mg of tissue were homogenized with a Polytron (Fisher, Pittsburgh, PA) in 5 ml of ice-cold modified balanced salt solution (BSS) containing (in mM) 10 Tris base, 55 NaCl, 40 KCl, 7 MgCl2, 5 CaCl2, 20 glucose, 50 sucrose, 0.2 4-(2-aminoethyl)benzenesulfonyl fluoride, 1 benzamidine, and 1 EDTA as well as (in μg/ml) 5 aprotinin, 10 pepstatin, 10 leupeptin, and 150 phenylmethylsulfonyl fluoride. Homogenates were then centrifuged at 1,000 g for 5 min at 4°C. The supernatant was transferred to a new tube and centrifuged at 1,500 g for 5 min at 4°C. The supernatant was then transferred to an ultracentrifuge tube and spun at 100,000 g for 90 min at 4°C. The membrane-enriched pellet was resuspended in 500 μl of BSS, and the protein concentration was quantified by Bradford colorimetric assay (Bio-Rad, Hercules, CA).
Immunoprecipitation of NDAE1
COS-7 cells grown on 60-mm dishes were incubated for 10 min on ice in 1 ml of RIPA buffer containing (in mM) 50 NaCl and 20 Tris, pH 8.0, plus 0.1% SDS, 0.5% deoxycholate, 1% Triton X-100, and 10 μl/ml of mammalian protease cocktail (Sigma). Cells were scraped, and the lysate was centrifuged at 13,000 rpm to pellet-insoluble material. The supernatant was then incubated with primary antibody at 4°C overnight. Next, 50 μl of a 50% slurry of protein A-Sepharose beads were added, and the mixture was incubated at 4°C for 2 h. The beads were washed with RIPA buffer, and protein-antibody complexes were released from the beads and denatured by the addition of Laemmli sample buffer (2% SDS, 10% glycerol, 60 mM Tris, pH 6.8, and 0.02% bromphenol blue) and boiling for 5 min.
Tagging NDAE1 With Enhanced Green Fluorescent Protein or Hemagglutinin A and Transient Transfection of COS-7 Cells
A COOH terminal-tagged enhanced green fluorescent protein (EGFP) version of NDAE1 (NDAE1-EGFP) was created using the QuickChange site-directed mutagenesis kit (Stratagene, La Jolla, CA) and appropriate primers to remove the stop codon of the NDAE1 cDNA and introduce an ApaI restriction site 3′ of the original stop codon. The mutated NDAE1 cDNA was subsequently removed from the pTLN vector via KpnI and ApaI restriction sites and directionally subcloned into pEGFPN-1 (Clontech, Palo Alto, CA). The sequence was verified by automated DNA sequencing (Cleveland Genomics).
To create a version of NDAE1 with a COOH-terminal hemagglutinin A (HA) tag (NDAE1-HA), site-directed mutagenesis primers were designed to remove the stop codon of NDAE1 and create a 3′ EcoRI restriction site in NDAE1-pTLN. NDAE1 was nondirectionally subcloned via EcoRI restriction sites (second site in the 5′ multiple cloning sites of pTLN) into the pMH mammalian expression vector (Boehringer Mannheim, Indianapolis, IN), and the sequence was verified by automated DNA sequencing (Cleveland Genomics).
COS-7 cells plated on glass coverslips were grown to 50% confluence and transiently transfected using the Lipofectamine Plus kit (GIBCO-BRL, Gaithersburg, MD) according to the manufacturer's protocol and were used 48 h posttransfection.
Western Blot Analysis
Proteins were separated by SDS polyacrylamide gel electrophoresis (SDS-PAGE) using the indicated percentage of acrylamide and subsequently transferred to a nitrocellulose membrane. Membranes were blocked for 1–2 h with 10% skim milk in Tris-buffered saline containing (in mM) 20 Tris, 137 NaCl, and 0.1% Tween 20 titrated to pH 7.5 (TBST). Blots were then incubated overnight at 4°C in TBST plus 10% skim milk with the appropriate dilution of primary antibody. Next, blots were washed with TBST and incubated for 2 h at RT in TBST plus 10% skim milk with a 1:5,000 dilution of horseradish peroxidase-conjugated secondary antibody (Jackson ImmunoResearch, West Grove, PA). Blots were washed with TBST, and signals were detected by enhanced chemiluminescence (ECL; Amersham Pharmacia).
Cell and Tissue Preparation/Immunohistochemistry
COS-7 cell preparation.
COS-7 cells attached to glass coverslips were fixed during a 30-min incubation at RT in PBS containing 4% paraformaldehyde (PFA). Cells were then incubated for 10 min in 100 mM glycine in PBS, washed with PBS, and permeabilized by 10 min of incubation in PBST.
Oregon-R strain D. melanogaster embryos were collected, and the chorion membranes (the outermost protective membrane) were removed by 5 min of incubation in 2.6% sodium hypochlorite (50% bleach) and then rinsed with a solution containing 0.9% NaCl and 0.03% Triton X-100 (41). Embryos were placed into a 25-ml glass scintillation vial and suspended in 10 ml of heptane. Next, 10 ml of fixing solution was added containing (in mM) 100 PIPES, 2 EGTA, and 1 MgSO4 and 4% PFA titrated to pH 7.0. Embryos were agitated at RT for 30 min. The aqueous layer was then aspirated, and 100 mM glycine in PBS was added for 60 min to clear remaining fixative. The aqueous layer was then removed, and 10 ml of methanol were added. The embryos were vigorously shaken for 1 min, and the vitelline membranes at the heptane/methanol interface and the heptane were removed. Embryos were washed with methanol and stored at −20°C until rehydrated in PBS for use.
D. melanogaster third-instar larvae (43) were dissected in PBS using Teflon-coated no. 5 forceps and a fine dissecting probe. Tissues were washed in PBS, fixed for 20 min in 2% PFA in PBST, and treated with 100 mM glycine in PBS for 45 min.
Oocytes used for electrophysiological studies (see Oocyte Preparation and Electrophysiology) were immediately transferred to PBS containing 4% PFA for 1 h at RT. Subsequently, oocytes were incubated in 100 mM glycine in PBS for 1 h and washed in PBS. Oocytes were then frozen in OCT compound (EM Sciences, Ft. Washington, PA) using liquid nitrogen and sliced to 10 μm with a cryotome. Slices were mounted on gelatin-coated microscope slides and stored at −20°C until use.
Nuclei staining with 4′,6-diamidino-2-phenylindole dihydrochloride.
In some experiments 4′,6-diamidino-2-phenylindole dihydrochloride (DAPI) stain was used to demarcate cell nuclei. After staining was completed with the appropriate primary and secondary antibodies, samples were incubated in PBS with a 1:1,000 dilution of DAPI for 5 min at RT, followed by washes with PBS.
Cells and tissues were incubated in PBST containing 1% BSA and 10% donkey and/or 10% goat serum for 30–60 min at RT. Primary antibodies at the indicated dilution were then added for 2.5 h. Samples were then washed extensively in PBST and treated for 2 h with the indicated secondary antibodies (Jackson ImmunoResearch) in PBST containing 1% BSA, 10% donkey, and/or 10% goat serum, followed by washes with PBST. Embryos were cleared by incubation in PBS plus 50% glycerol for 30 min. Samples were mounted on microscope slides using DAKO fluorescence mounting medium (DAKO, Carpinteria, CA). Images were acquired with an AxioCam digital camera and AxioVision software (Carl Zeiss, Oberkochen, Germany). Confocal images were acquired using a Zeiss LSM 410 confocal microscope. Samples treated with secondary antibody alone or with affinity-purified αCWR57 (αCWR57ap) incubated for 1 h at RT with GST-NTERM fusion peptide were used as negative controls.
Oocyte Preparation and Electrophysiology
Xenopus laevis oocytes were prepared for cRNA injection as previously described (48). cRNA (50 nl of 0.7 μg/μl) or water was injected into oocytes that were incubated at 18°C in OR3 medium (47). Oocytes were studied 3–11 days postinjection, and experimental procedures were repeated in two batches of oocytes from different frogs to take into account possible biological variation. Ion-selective microelectrode fabrication and experimentation were performed as previously described (54).
Characterization of NH2-Terminal NDAE1 Antibody and Fusion Protein
We constructed an NH2-terminal NDAE1-GST fusion protein (GST-NTERM100) and used this recombinant protein for competition assays to determine the epitope specificity of the NH2-terminal NDAE1 antibody. Figure1 shows a Western blot of the GST-NTERM100 peptide probed with the affinity-purified NH2-terminal antibody αCWR57ap (blot 1). The molecular mass of GST is 26 kDa, and the predicted molecular mass of GST-NTERM100 is 37 kDa. The αCWR57ap recognizes a band in the GST-NTERM100-loaded lane with an apparent molecular mass of ∼40 kDa as well as several other bands (35, 32, and 28 kDa), which likely represent peptide degradation products. The bands in the GST-NTERM100 lane can be competed by 50 μg peptide (Fig. 1, blot 3), indicating that αCWR57ap recognizes the NH2-terminal epitope conjugated to GST, and not GST alone. Reprobing these blots with anti-GST antibody (αGST) indicates that the bands recognized by αCWR57ap represent the GST-fusion protein (Fig. 1, blots 2and 4). The presence of a 28-kDa degradation band indicates that the epitope recognized by αCWR57ap is likely in the first 20–30 amino acids.
Detection of Full-Length NDAE1
We next tested the ability of αCWR57ap to detect full-length endogenous NDAE1 protein from tissues (Drosophila) as well as in cells heterologously expressing NDAE1 (oocytes and COS-7 cells). Figure 2shows a representative (n = 5) Western blot of membrane preparations of whole D. melanogaster embryos and adults. A single band with an apparent molecular mass of ∼105 kDa is present in these lanes and in the lane with protein from COS-7 cells transiently transfected with NDAE1 cDNA. Within the first 100 amino acids of NDAE1, a stretch of sequence is present, ETARWIKFEED, that is highly conserved among the AEs, NBCs, and NDAE1. However, no signal is detected in either COS-7 cells alone or COS-7 cells transfected with the rat kidney electrogenic Na+-HCO cotransporter (rkNBC) (45). This finding indicates that αCWR57ap recognizes a NDAE1-specific epitope. The signal caused by αCWR57ap was eliminated by preincubation for 1 h with 25 μg of GST-NTERM100 peptide (not shown).
Detection, Localization, and Physiology of Recombinant NDAE1 Expressed in Xenopus Oocytes
Initial characterization of NDAE1 as expressed inXenopus oocytes revealed that it functions as a Na+-driven anion exchanger, with the capacity to mediate Na+-driven Cl−/HCO exchange (47). Figure 3 Ashows an experiment in which pHi of a water-injected control and a NDAE1 cRNA-injected oocyte were simultaneously monitored. The resting pHi of the NDAE1 cRNA-injected oocyte was ∼0.3 pH units more alkaline than that of the control (0.27 ± 0.03, n = 9). Addition of CO2/HCO resulted in an ∼0.2 pH unit acidification in the control oocyte due to the diffusion, hydration, and subsequent dissociation of CO2 to H+ and HCO . In contrast, the NDAE1 oocyte only transiently acidified and then alkalinized by ∼0.05 pH units. This finding indicates that HCO influx is greater than H+ evolution due to CO2 diffusion across the plasma membrane. Bath Cl− replacement with gluconate (0 Cl−) had no effect on the control oocytes pHibut resulted in an ∼0.1 pH unit alkalinization in the NDAE1 oocyte, illustrating the exchange of Cl− for HCO as mediated by NDAE1.
We cryosectioned the oocytes tested in Fig. 3 A and used indirect immunofluorescent staining to determine the location of NDAE1 protein (n = 3). As shown in Fig. 3 B, staining with αCWR57ap was observed in the NDAE1 oocyte sections as a thin band that coincides with the margin of the plasma membrane. The intracellular staining in Fig. 3 B represents NDAE1 protein not processed to the plasma membrane. No signal was observed in NDAE1 oocyte sections stained with secondary antibodies alone or in water-injected control oocyte sections probed with αCWR57ap. Together with previous physiological data showing that the membrane-impermeable stilbene DIDS inhibits NDAE1 function (47), these histological data indicate that NDAE1 can traffic to the plasma membrane.
Plasma Membrane Sidedness of the NH2 and COOH Termini of NDAE1
To determine the location of the COOH terminus of NDAE1 with respect to the plasma membrane, we constructed an EGFP-tagged version of NDAE1 (NDAE1-EGFP). Figure4 A shows that when immunoprecipitated with αCWR57ap, both αCWR57ap and a monoclonal anti-GFP antibody (αGFP) recognize a band with an apparent molecular mass of ∼140 kDa (NDAE1 = 105 kDa, EGFP = 35 kDa) from COS-7 cells transfected with NDAE1-EGFP. Figure 4 Bshows confocal micrograph optical sections of COS-7 cells transiently transfected with EGFP (top) or NDAE1-EGFP (bottom). Overlapping αCWR57ap staining and the EGFP signal are apparent at the perimeter of the cells transfected with NDAE1-GFP, although some perinuclear signal is observed (a common artifact of transient overexpression). In contrast, cells transfected with EGFP alone show diffuse EGFP signal and no αCWR57ap staining. This finding indicates that NDAE1-EGFP protein localizes at or near the plasma membrane. Preliminary data indicate that NDAE1-EGFP mediates Na+-driven anion exchange in the same manner as NDAE1 in transfected COS-7 cells (M. B. Ganz, C. M. Sciortino, and M. F. Romero, unpublished observations). Thus some of the NDAE1-EGFP protein is at the plasma membrane.
To localize the NH2 terminus of NDAE1 relative to the plasma membrane, COS-7 cells cotransfected with NDAE1 and EGFP were fixed either with or without Triton X-100 permeabilization and subjected to indirect immunofluorescent staining with αCWR57ap primary antibody. As shown in Fig. 5, although a punctate EGFP signal is apparent in both permeabilized (top) and nonpermeabilized (bottom) cells, only permeabilized cells show αCWR57ap staining. These data indicate that the NH2 terminus of NDAE1 is intracellular, as predicted by hydropathy analysis and structural analogy to AE1.
The location of the COOH terminus of NDAE1 was determined using a COOH-terminal, HA-tagged NDAE1 construct (NDAE1-HA). As shown in Fig.6 A, a band with an apparent molecular mass of 105 kDa from COS-7 cells transfected with NDAE1-HA (NDAE1 = 105 kDa, HA = 0.8 kDa) is recognized on a Western blot by anti-HA monoclonal antibody (αHA). Immunoprecipitation using αHA antibody followed by Western blot probed with αCWR57ap revealed that this band was NDAE1-HA. COS-7 cells transiently cotransfected with NDAE1-HA and EGFP (for visualization and control purposes) were fixed either with or without Triton X-100 permeabilization and probed with αHA. As shown in Fig. 6 B, only permeabilized cells had a positive signal with αHA antibody. These results indicate that the NDAE1 COOH terminus is also located on the cytosolic face of the plasma membrane.
Localization of Endogenous NDAE1 in Drosophila melanogaster
αCWR57ap and a monoclonal antibody to the α-subunit of the Na+-K+-ATPase (27) were used to determine the tissue distribution of NDAE1 and the Na+-K+-ATPase, respectively (n= 5), in Drosophila. Figure7 A shows that NDAE1 (left) but not Na+-K+-ATPase (middle) protein is expressed in blastoderm embryos (stage 4 of embryogenesis). By stage 9 (Fig. 7 B), both proteins are expressed in the developing alimentary tract and nervous system, including the mesectoderm (a band of neuronal precursor cells). Additionally, NDAE1 and the Na+-K+-ATPase are expressed in the regions corresponding to the developing gut (Fig.7 B). By stage 15 (Fig. 7 C), the midgut has completely fused and neuronal differentiation has progressed such that neuronal elongation can be observed. Sensory end organs, including Bolwig's organ and the dorsal organ, which develop in the optic and antennal system, respectively, become distinct (17). Both NDAE1 and Na+-K+-ATPase are expressed in the midgut, MTs, and anal pads as well as the central nervous system (CNS), peripheral nervous system (PNS), and developing sensilla (i.e., Bolwig's organ and the dorsal organ). By the end of embryogenesis, stage 17 (Fig. 7 D), the major segments of the alimentary tract have fully developed (fore-, mid-, and hindgut, MTs, and salivary glands). NDAE1 is expressed in all segments of the alimentary tract. Although the Na+-K+-ATPase protein is expressed in the fore- and midgut as well as in the MTs, this protein is not detectable in the hindgut. Expression of both proteins persists in the CNS and PNS as well as in the more mature sensilla in the head region. NDAE1 protein expression is coincident with mRNA expression patterns observed previously by in situ hybridization studies of D. melanogaster embryos (47). Minimal background staining is observed with secondary antibody alone (Fig. 7 E) or with αCWR57ap preincubated for 1 h with 25 μg of GST-NTERM100 (Fig. 7 F).
NDAE1 protein was next localized in various tissues of third-instar larvae. Preliminary experiments with larvae and adults revealed a high level of NDAE1 protein expression in four major tissues: MTs, gut, salivary gland, and brain (not shown). We dissected these tissues to characterize the tissue distribution of NDAE1. Like many insects, D. melanogaster has anterior (AMT) and posterior (PMT) MTs, each with three major segments: ureter/proximal, medial, and distal. Figure 8 Ashows the immunolocalization of NDAE1 and the Na+-K+-ATPase in the AMT. Both proteins are abundantly expressed in the ureter and proximal portions of the MT. However, expression is attenuated in the medial tubule, and neither protein is expressed in the distal (blind ending) AMT. In contrast, Fig. 8 B shows that both NDAE1 and Na+-K+-ATPase are expressed along the entire length of the PMT. Figure 8 C shows that αCWR57ap signal is competed away by preincubation of this primary antibody for 1 h at RT with GST-NTERM100 peptide.
Confocal microscopy was used to more precisely visualize the expression patterns of NDAE1 in MTs, gut, and salivary gland. Figure9 A shows a confocal micrograph of the proximal segments of AMT (left) and PMT (right). Both NDAE1 and the Na+-K+-ATPase are more highly expressed in the PMT. Additionally, NDAE1 expression is more discrete at the outer perimeter of the AMT principal cells. Because the Na+-K+-ATPase localizes at the basolateral membrane of MTs in D. melanogaster, these results indicate that the vast majority of NDAE1 protein is also located basolaterally, although NDAE1 may be located apically. Figure 9 B shows the staining pattern in the mid/hindgut at the point of insertion of the ureter of the AMT. The levels of NDAE1 and Na+-K+-ATPase staining are similar in the midgut and the AMTs. Figure 9 C shows that NDAE1 is also expressed in the salivary glands, apparently also basal in location. The Na+-K+-ATPase protein is much less abundant in the salivary gland than in MTs or gut. The data in Fig. 9 show that NDAE1 expression persists throughout gut development and may play an important role in the function and/or pH regulation of all parts of the alimentary tract in the mature fly.
Na+-dependent Cl−/HCO exchange is an important mechanism by which many invertebrate and vertebrate cells maintain pHi homeostasis. To better understand how and where NDAE1 may contribute to cellular pHi regulation as well as Na+, HCO , and Cl− transport across secretory epithelia, we developed a polyclonal antibody to the NDAE1 NH2 terminus. We have characterized NDAE1 protein distribution in embryos and larvae of D. melanogaster.
Antibody Specificity and Subcellular Localization of NDAE1
The αCWR57ap polyclonal antibody was raised against the first 100 amino acids of NDAE1, which contains a stretch of amino acids that is highly conserved among members of the BTS (ETARWIKFEED). However, αCWR57ap did not cross-react with rkNBC (rkNBC-EGFP) expressed in COS-7 cells either on a Western blot (Fig. 2) or by immunofluorescence (not shown), indicating that this antibody does not recognize this epitope. Based on the Drosophila genome sequence (2), only one other sequence is related to the BTS (CG8177). The size of the predicted open reading frame is 1,239 amino acids, with a predicted molecular mass of 140 kDa. CG8177 has 31.6% overall homology to NDAE1, but the putative NH2 termini are only 26% homologous [sharing identity only in the EWKETARWIK(F/Y)EEDVE sequence]. Because αCWR57ap does not cross-react with this common sequence of rkNBC and does not recognize a 140-kDa protein in Drosophila, our data indicate that αCWR57ap should not recognize the putative CG8177 protein. Thus far, of known BTS proteins, αCWR57ap specifically detects NDAE1.
Although the predicted molecular mass of NDAE1 is 114 kDa, αCWR57ap recognizes a band with an apparent molecular mass of ∼105 kDa from D. melanogaster tissue membrane preparations and a slightly larger band in COS-7 cells transiently transfected with NDAE1 cDNA. The band from native tissue is much more discrete than that of the COS-7 transfections, perhaps indicating a difference in posttranslational modification of the protein. The observation of a 28-kDa degradation band in Fig. 1 indicates that the epitope recognized by αCWR57ap is likely located within the first 20–30 amino acids of the NH2 terminus of NDAE1. Accordingly, it is unlikely that this ∼105-kDa band on Western blots is due to an NH2-terminal truncation. Our cell transfection studies with HA- and EGFP-tagged versions of NDAE1 imply that the COOH terminus remains intact when expressed in COS-7 cells. Initial characterization of an antibody raised against the COOH terminus of NDAE1 also recognizes a band on Western blots with an apparent molecular mass of ∼105 kDa from D. melanogaster tissue, indicating that the COOH terminus remains intact in vivo (unpublished observations) and that the NDAE1 protein runs at a lower molecular mass on an SDS-PAGE gel than predicted.
The primary sequence of NDAE1 contains one putative protein kinase A and seven putative protein kinase C regulatory sites on the NH2- and COOH-terminal tails. cAMP (14), cGMP (13), and intracellular Ca2+ as well as nitric oxide have been shown to regulate fluid secretion in the D. melanogaster MTs. In mammalian cells, Na+-dependent Cl−/HCO exchange activity is also regulated by hormones and growth factors such as angiotensin II (16). Permeabilization studies of NDAE1 expressed in COS-7 cells revealed that both termini face the cytosolic side of the plasma membrane. Thus the termini prospectively can be affected by these cell regulatory mechanisms. Initial experiments have indicated that NDAE1 function can be modulated by an angiotensin II-mediated signaling mechanism (M. B. Ganz, C. M. Sciortino, and M. F. Romero, unpublished observations).
Developmental Expression of NDAE1 Protein in the Alimentary Tract of D. melanogaster
The alimentary tract of insects is responsible for the absorption of nutrients and maintenance of solute concentration and water balance of the hemolymph. Luminal gut pH in many insect species, including those in the order Diptera, i.e., Drosophila and mosquitoes, is quite alkaline (pH 8–11) (19). This alkaline pH facilitates proper nutrient absorption and may be a key signal for the movement of malarial parasites (Plasmodium) from lumen to cell in mosquitoes (56). An apical V-type H+-ATPase, in part, facilitates this alkaline pH across the gut epithelia (19). Yet, the epithelia maintains pHi homeostasis, indicating that another acid-base transporter at the basolateral membrane facilitates this process.
On the basis of determinations of ion content across the basolateral membrane (5, 64, 65), as well as the pH gradient and membrane potential, a basolaterally located NDAE1 likely functions as an acid extruder in MTs, gut, and salivary glands. The current model of basolateral ion transport includes a Na+-K+-ATPase, KCl cotransporter, and an unidentified Na+/solute transport mechanism (30). Basolateral localization of NDAE1 in MTs intimate the role of NDAE1 basolateral solute transport. This hypothesis is further supported by the colocalization of NDAE1 and the Na+-K+- ATPase. Dow and coworkers (14) showed that PMTs are capable of fluid secretion along their entire length, whereas AMT only secrete fluid at their proximal end. Our findings that both NDAE1 and the Na+-K+- ATPase are homogenously expressed along the length of the PMTs but only proximally in AMTs indicate that protein expression of these proteins parallels the secretory function of MT epithelia.
We also observed that NDAE1 is abundant in the cuboidal cells of the salivary gland, whereas the Na+-K+-ATPase is much less abundant (Fig. 9). This low level of Na+-K+-ATPase expression correlates with the low mRNA signal detected by Northern blot analysis, presented in the original characterization of the α5 antibody in D. melanogaster (27). NDAE1 is apparently located on the basolateral membrane (Fig. 9) because fluorescence signal was observed on the outer edge, but not the interior (not shown), of the salivary gland cells. Thus the role of NDAE1 may be to mediate basal HCO transport, ultimately leading to the secretion of an alkaline fluid across the luminal surface of the gland.
Future studies should explore the in vivo role of NDAE1 inDrosophila as well as NDAE1 expression inDiptera. Perhaps NDAE1, functioning as an acid-extruding Na+-dependent Cl−/HCO exchanger, enables the sustained transepithelial alkaline pH observed in the Diptera gut. Controlling NDAE1 function and/or expression may therefore allow control of the alkaline pH signal forPlasmodium gametogenesis in the mosquito gut (7,56) as well as the progression of the parasites to the salivary glands.
NDAE1 Expression in D. melanogaster Nervous System
Work in our laboratory has shown that NDAE1, when expressed inXenopus oocytes, functions as a Na+-driven Cl−/HCO exchanger (47) and a Na+-driven Cl−/base exchanger (53). Na+-dependent Cl−/HCO exchange has been demonstrated as a key pHi regulator of invertebrate neurons (49,50, 59). NDAE1 is expressed in both the CNS and PNS as well as in the developing sensory organs (dorsal organ and Bolwig's organ). Maintenance of pHi is critical for proper neuronal function (23). Understanding the role of NDAE1 in pHiregulation of the Drosophila nervous system may provide genetic models for nervous system physiology and pathophysiology (i.e., hypoxia) as well as a potential means for controlling insect viability. Future studies using cell type-specific markers are necessary to more precisely identify cell types expressing NDAE1 in both the alimentary tract and nervous systems.
This study demonstrates that the Na+-driven Cl−/HCO exchanger NDAE1 is expressed in both secreting epithelia of D. melanogaster as well as in the developing nervous system (CNS, PNS, and sensory organs). Moreover, our data structurally assign the NH2- and COOH termini of NDAE1 to the cytosol, allowing the termini to be potential targets for cell regulation known to mediate ion and water flux. Future studies are necessary to determine the physiological role of NDAE1 in vivo.
We thank Montelle C. Sanders for technical assistance and Maryanne Pendergast for assistance in obtaining confocal images.
NOTE ADDED IN PROOF
Since this manuscript was accepted, other investigators (Grichtchenko II, Choi I, Zhong X, Bray-Ward P, Russell JM, and Boron WF. Cloning, characterization, and chromosomal mapping of a human electroneutral Na+-driven Cl-HCO3 exchanger.J Biol Chem 276: 8358–8363, 2001) have reported another cDNA encoding a mammalian Na+-driven Cl−/HCO exchanger (NDCBE; GenBank accession no. AF069512).
This work was supported by an American Heart Association grant, a Howard Hughes Medical Institute Grant to Case Western Reserve University (M. F. Romero), and National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) Grant DK-56218 (M. F. Romero). C. M. Sciortino was supported by an NIDDK Predoctoral Fellowship (DK-07678). L. D. Shrode was supported by a National Heart, Lung, and Blood Institute Postdoctoral Fellowship (HL-07415).
Address for reprint requests and other correspondence: M. F. Romero, Dept. of Physiology & Biophysics, Case Western Reserve Univ. School of Medicine, 2119 Abington Rd., Cleveland, OH 44106-4970 (E-mail:).
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 © 2001 the American Physiological Society