Molecular Identification of the Apical Ca2+Channel in 1,25-Dihydroxyvitamin D3-responsive Epithelia*

In mammals, the extracellular calcium concentration is maintained within a narrow range despite large variations in daily dietary input and body demand. The small intestine and kidney constitute the influx pathways into the extracellular Ca2+ pool and, therefore, play a primary role in Ca2+ homeostasis. We identified an apical Ca2+influx channel, which is expressed in proximal small intestine, the distal part of the nephron and placenta. This novel epithelial Ca2+ channel (ECaC) of 730 amino acids contains six putative membrane-spanning domains with an additional hydrophobic stretch predicted to be the pore region. ECaC resembles the recently cloned capsaicin receptor and the transient receptor potential-related ion channels with respect to its predicted topology but shares less than 30% sequence homology with these channels. In kidney, ECaC is abundantly present in the apical membrane of Ca2+transporting cells and colocalizes with 1,25-dihydroxyvitamin D3-dependent calbindin-D28K. ECaC expression in Xenopus oocytes confers Ca2+influx with properties identical to those observed in distal renal cells. Thus, ECaC has the expected properties for being the gatekeeper of 1,25-dihydroxyvitamin D3-dependent active transepithelial Ca2+ transport.

Calcium is the most abundant cation in the human body, but less than 1% is present in ionic form in the extracellular compartment (1). The extracellular Ca 2ϩ concentration is precisely controlled by parathyroid hormone (PTH) 1 and 1,25-dihydroxyvitamin D 3 (1,25-(OH) 2 D 3 ). Daily dietary intake is less than 1000 mg of which only 30% is absorbed in the intestinal tract. This percentage is significantly enhanced during growth, pregnancy, and lactation by increased levels of circulating 1,25-(OH) 2 D 3 . Although there is a continuous turnover of bone mass, there is no net gain or loss of Ca 2ϩ from bone in a young and healthy individual. This implicates that healthy adults excrete maximally 300 mg Ca 2ϩ in the urine to balance intestinal Ca 2ϩ uptake and that the remaining filtered load of Ca 2ϩ has to be reabsorbed by the kidney. Recently, the mechanism by which extracellular Ca 2ϩ is sensed by the parathyroid gland was elucidated by cloning of the Ca 2ϩ -sensing receptor (2), and mutations in this receptor gene explained familial hypocalciuric hypercalcemia (3). The importance of 1,25-(OH) 2 D 3 in Ca 2ϩ homeostasis of the body is reflected by mutations in the genes coding for 1␣-hydroxylase (4), a renal enzyme controlling its synthesis, and the 1,25-(OH) 2 D 3 -receptor (5). Transepithelial Ca 2ϩ transport is a three-step process consisting of passive entry across the apical membrane, cytosolic diffusion facilitated by 1,25-(OH) 2 D 3 -dependent calcium-binding proteins (calbindins), and active extrusion across the opposing basolateral membrane mediated by a high affinity Ca 2ϩ -ATPase and Na ϩ -Ca 2ϩ exchanger (6). Until now, the molecular mechanism responsible for Ca 2ϩ entry into small intestinal and renal cells, which serve as the influx pathways into the extracellular Ca 2ϩ pool, is still elusive (6).

EXPERIMENTAL PROCEDURES
Primary Cultures of Kidney Cells-Rabbit connecting tubule (CNT) and cortical collecting duct (CCD) cells were immunodissected from New Zealand White rabbits (ϳ0.5 kg) with monoclonal antibody R2G9, set in primary culture on permeable filter supports (0.33 cm 2 , Costar), and grown to confluence for 5 days, as described previously (7).
Expression Cloning and DNA Analysis-Poly(A) ϩ RNA, which induced 45 Ca 2ϩ uptake in Xenopus laevis oocytes, was isolated from primary cultures of rabbit CNT and CCD cells and used to construct a directional cDNA library using a SuperScript TM cDNA synthesis system (Life Technologies, Inc.). cDNA was ligated into the pSPORT1 vector, and ElectroMax DH10B cells were transformed using a Bio-Rad Gene Pulser. cRNA synthesized in vitro from pools of ϳ30.000 independent bacterial clones from this cDNA library was injected in oocytes. A pool expressing highest 45 Ca 2ϩ uptake rates was sequentially subdivided and analyzed until a single clone (ECaC) was identified that was double-stranded sequenced using an automatic sequencer (ABI Prism 310 Genetic Analyzer). The mean hydrophobicity index was computed according to the algorithm of Kyte and Doolittle (8) with a window of 9 residues. Homology searches were performed against the nonredundant GenBank TM data base.
Immunohistochemistry-Rabbit kidney slices were fixed in 1% (v/v) * This work was supported in part by a grant from the Dutch Organization of Scientific Research (NWO-ALW 805-09.042). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
periodate-lysine-paraformaldehyde fixative for 2 h, washed with 20% (w/v) sucrose in phosphate-buffered saline, and subsequently frozen in liquid N 2 . Sections (7 m) were blocked with 5% (w/v) blocking reagent (NEN Life Science Products) in phosphate-buffered saline for 15 min. Sections were washed three times with Tris-buffered saline (TBS; 150 mM NaCl, 100 mM Tris-HCl, pH 7.5) and incubated with affinity-purified guinea pig antiserum raised against the ECaC C-tail (amino acids 580 -706) and rabbit anti-calbindin-D 28K antiserum (9) for 16 h at 4°C. After thorough washing with TBS, the sections were incubated with the corresponding fluorescein isothiocyanate-or tetramethylrhodamine isothiocyanate-conjugated anti-immunoglobulin G for 60 min. Subsequently, sections were washed with TBS, distilled water, and methanol and finally mounted in Mowiol (Hoechst). All controls, including sections treated with preimmune serum or with conjugated antibodies only, were devoid of any staining.

RESULTS AND DISCUSSION
Here, we report the expression cloning, tissue distribution, immunolocalization, and functional characterization of the apical Ca 2ϩ influx channel, which is expressed solely in proximal small intestine, the distal part of the nephron, and placenta. In analogy to the recently cloned amiloride-sensitive and aldosterone-dependent epithelial Na ϩ channel (ENaC) (10), present in the apical membrane of sodium-transporting epithelia, this novel epithelial Ca 2ϩ channel was named ECaC. By screening for maximal 45 Ca 2ϩ influx activity in oocytes a single 2.8kilobase pair cDNA was isolated from a directional cDNA library prepared from poly(A) ϩ RNA of rabbit distal tubular cells. The ECaC cDNA contains an open reading frame of 2190 nucleotides that encodes a protein of 730 amino acids with a predicted relative molecular mass of 83 kDa (M r 83,000) (Fig.  1A). Hydropathy analysis suggests that ECaC contains three structural domains: a large hydrophilic amino-terminal domain of 327 amino acids containing three ankyrin binding repeats and several potential protein kinase C phosphorylation sites, suggesting an intracellular location; a six transmembrane-spanning domain with two potential N-linked glycosylation sites and an additional hydrophobic stretch between transmembrane segments 5 and 6 indicative of an ion pore region; and a hydrophilic 151-amino acid carboxyl terminus containing potential protein kinase A and C phosphorylation sites (Fig. 1B).
A protein data base search revealed only a significant homology of less than 30% between ECaC and the recently cloned capsaicin receptor (VR1) (11), the transient receptor potential (TRP)-related ion channels (12) and olfactory channels (13). The capsaicin receptor is a nonselective cation channel and functions as a transducer of painful thermal stimuli (11). Members of the TRP family have been proposed to mediate the entry of extracellular Ca 2ϩ into cells in response to depletion of intracellular Ca 2ϩ stores (12). These proteins resemble ECaC with respect to their predicted topological organization and the presence of multiple NH 2 -terminal ankyrin repeats (14). There was also striking amino acid sequence similarity between ECaC, VR1, and TRP-related proteins within and adjacent to the sixth transmembrane segment, including the predicted area that may contribute to the ion permeation path (Fig. 1C) (15). Outside these regions, however, ECaC shares only 20% sequence similarity with VR1 and TRP family members, suggesting a distant evolutionary relationship among these channels.
High stringency Northern blot analysis of ECaC transcripts revealed prominent bands of ϳ3 kb in small intestine, kidney, and placenta (Fig. 2). We found that in the intestine ECaC mRNA expression was highest in duodenum, decreased in jejunum, and absent in ileum and colon. ECaC mRNA expression in kidney and placenta was comparable with jejunum. In addition, ECaC transcripts in lung, skeletal muscle, stomach, heart, liver, spleen, and brain were undetectable. Most important is that expression of ECaC coincides with that of calbindin-D 9K in intestine and placenta and calbindin-D 28K in kidney (16,17).
Immunofluorescence staining revealed that in kidney ECaC is abundantly present along the apical membrane in the majority of cells lining the distal part of the nephron including distal convoluted tubule, connecting tubule and cortical collecting duct, where it colocalizes with calbindin-D 28K (Fig. 3). This part of the nephron is the site of PTH-and 1,25-(OH) 2 D 3regulated transcellular Ca 2ϩ reabsorption (18).
By functional expression of ECaC in Xenopus oocytes we observed that the 45 Ca 2ϩ uptake was linear for at least 2 h (Fig.  4) and increased with increasing extracellular Ca 2ϩ concentra- Cations were added at a final concentration of 0.5 mM, pH was lowered from 7.4 to 5.9, 50 mM NaCl was replaced by 50 mM KCl to depolarize the membrane, methoxyverapamil (m-verapamil) and felodipine were added at 10 M. In transcellular Ca 2ϩ transport measurements across renal cells these additions were applied to the apical compartment only. C and D, concentration dependence of 45 Ca 2ϩ uptake in oocytes expressing ECaC (C) and transcellular 45 Ca 2ϩ transport in rabbit kidney cells (D). Measurements were performed at the indicated Ca 2ϩ concentrations in extracellular and apical medium of oocytes and confluent monolayers, respectively. Water-injected oocytes exhibited a Ca 2ϩ uptake of less than 6% of the ECaC-injected oocytes. Values are means Ϯ S.E. of five experiments. Statistical significance (*, p Ͻ 0.05) was determined by analysis of variance.

FIG. 2.
ECaC expression in small intestine, kidney, and placenta. High-stringency Northern blot analysis of RNA from a range of tissues probed with 32 P-labeled ECaC cDNA. Lanes were loaded with 2.5 g of poly(A) ϩ RNA from rabbit duodenum, jejunum, ileum, kidney, distal colon, lung, skeletal muscle, stomach, heart, spleen, liver, brain, and human placenta, respectively. Molecular mass standards (in kilobases) are indicated on the left. tions between 0.01 and 2.0 mM with an apparent affinity for Ca 2ϩ of ϳ0.2 mM (Fig. 5C). This is well within the range of physiologically relevant extracellular calcium concentrations. As shown in Fig. 5A, trivalent and divalent cations inhibited 45 Ca 2ϩ influx in the following rank order of potency: La 3ϩ Ͼ Cd 2ϩ Ͼ Mn 2ϩ , while Ba 2ϩ , Mg 2ϩ , and Sr 2ϩ had no effect. It is striking that Ba 2ϩ and Sr 2ϩ , which are highly permeant in voltage-gated Ca 2ϩ channels (19), do not interfere with ECaC. In addition, the L-type Ca 2ϩ channel antagonists and depolarization with 50 mM KCl were without effect. VR1, to which ECaC has the highest homology, shows a relative low permeability to monovalent ions such as Na ϩ (11). In a double labeling experiment, ECaC-injected oocytes did not exhibit a significant Na ϩ influx (0.88 Ϯ 0.05 and 0.95 Ϯ 0.05 nmol⅐h Ϫ1 ⅐ oocyte Ϫ1 for ECaC and water-injected oocytes, respectively; n ϭ 29 oocytes; p Ͼ 0.2), while displaying a markedly increased Ca 2ϩ influx. In humans metabolic acidosis induces hypercalciuria (18,20), and here we demonstrate that acidification of the extracellular medium to pH 5.9 significantly inhibits 45 Ca 2ϩ influx (Fig. 5, A and B). If extrapolatable to the in vivo situation this effect could well be the molecular explanation of acidosisinduced calciuresis. Taken together, these characteristics indicate that ECaC is distinct from previously described Ca 2ϩ channels. Furthermore, the above described pharmacological and functional properties of ECaC are identical to those of Ca 2ϩ transport across the monolayers (Fig. 5, C and D), providing evidence that the protein is a major constituent of the transcellular Ca 2ϩ transport system in renal cells. Together with the previous finding that the Ca 2ϩ influx rate at the apical membrane of renal distal cells is tightly coupled to transepithelial Ca 2ϩ flux over a wide range of transport rates (21), this suggests that the apical Ca 2ϩ influx is the rate-limiting step in transcellular Ca 2ϩ transport. Moreover, this implicates that hormonal regulation of a single influx pathway, i.e. ECaC, may control the rate of transcellular Ca 2ϩ transport.
In humans, approximately 4% of the population suffers from idiopathic hypercalciuria with the characteristics of autosomal dominant transmission (22). In some of the affected individuals, hypercalciuria is secondary to hyperabsorption of Ca 2ϩ (22). Gain of function mutations in ECaC or its dysregulation may well be the cause of absorptive hypercalciuria. The present elucidation of ECaC allows to study these possibilities with molecular genetic approaches.