Molecular Mechanisms of Calcium-sensing Receptor-mediated Calcium Signaling in the Modulation of Epithelial Ion Transport and Bicarbonate Secretion *

Background: Calcium-sensing receptor (CaSR) plays a critical role in the regulation of epithelial ion transport. Results: CaSR activators induce Ca 2 (cid:2) signaling and duodenal bicarbonate secretion (DBS). Conclusion: CaSR triggers Ca 2 (cid:2) -dependent DBS, likely through receptor-operated channels, intermediate conductance Ca 2 (cid:2) activated K (cid:2) channels, and the cystic fibrosis transmembrane conductance regulator. Significance: Dietary CaSR activators may modulate the physiological process of DBS that is critical for duodenal mucosal protection.

Cytosolic free Ca 2ϩ ([Ca 2ϩ ] cyt ) 4 plays an essential role in a variety of mammalian cells through the regulation of many biological functions, including neurotransmitter release, muscle contraction, gene regulation, cell proliferation, and apoptosis (1). Therefore, dysregulation of [Ca 2ϩ ] cyt homeostasis may result in pathological changes in many systems. Under physiological conditions, various mechanisms are controlling Ca 2ϩ homeostasis in the human body, one of which is the calciumsensing receptor (CaSR) (2). The CaSR is a plasma membrane protein initially cloned from bovine parathyroid cells. It is a member of the G protein-coupled receptor family and regulates the synthesis of parathyroid hormone in response to changes in serum Ca 2ϩ concentrations (3)(4)(5).
CaSR activation elicits complex intracellular signaling events through the modulation of a wide range of intracellular mediators, including G␣ q/11 proteins and phospholipase C (PLC). These, in turn, stimulate both inositol trisphosphate production and PKC activation, which increases [Ca 2ϩ ] cyt (4,5). Activation of the CaSR has been shown to increase [Ca 2ϩ ] cyt in different types of mammalian cells, especially in parathyroid cells, epithelial cells, osteocytes, cardiomyocytes, and smooth muscle cells (4,5). In addition, CaSR activation can stimulate G␣ i proteins and phosphodiesterase, leading to a decrease in cyclic AMP and cyclic GMP levels (4,5).
It has been demonstrated that the CaSR is expressed along the entire gastrointestinal tract and plays a critical role in normal gut physiology (6). Recent studies have been mainly performed on its functions in modulating gastrin and gastric acid secretion and intestinal fluid and electrolyte transports (6 -9). Extracellular Ca 2ϩ ([Ca 2ϩ ] o ) stimulates gastric acid and bicarbonate secretion in the guinea pig (10,11), suggesting that gastric surface epithelial cells are capable of sensing changes in Ca 2ϩ to modulate gastric secretion, likely through CaSR activation. Although it is well documented that the CaSR inhibits intestinal transepithelial Cl Ϫ secretion by blocking cyclic AMP signaling (7), little is known about the role of the CaSR in intestinal transepithelial HCO 3 Ϫ secretion, which is a critical factor in duodenal mucosal protection and mainly under the control of cyclic AMP and Ca 2ϩ signaling. Although the physiological roles and molecular mechanisms of cyclic AMP-induced HCO 3 Ϫ secretion are relatively well defined, those induced by Ca 2ϩ signaling remain poorly understood in most epithelia, especially in intestinal epithelia (12). Moreover, although it is known that Ca 2ϩ signaling is critical for duodenal bicarbonate secretion (DBS), the molecular mechanisms controlling [Ca 2ϩ ] cyt homeostasis in duodenal epithelial cells are poorly understood.
In our previous studies, we proposed that Ca 2ϩ and cyclic AMP signaling may play different roles in the regulation of intestinal transepithelial HCO 3 Ϫ and Cl Ϫ secretion. We found that although cyclic AMP plays a major role in intestinal Cl Ϫ secretion, Ca 2ϩ signaling may be critical for transepithelial HCO 3 Ϫ secretion (13). However, activation of most well defined receptors expressed in intestinal epithelial cells usually increase both [Ca 2ϩ ] cyt and intracellular cyclic AMP levels, making it difficult to distinguish between [Ca 2ϩ ] cyt -and cyclic AMP-regulated epithelial ion transports. Because CaSR activation results in an increase in [Ca 2ϩ ] cyt but a decrease in intracellular cyclic AMP levels (3)(4)(5), we hypothesized that the CaSR is a suitable receptor system for further delineating the role of [Ca 2ϩ ] cytand cAMP-mediated intestinal epithelial ion transports in general and HCO 3 Ϫ secretion in particular. Therefore, in this study, we sought to investigate CaSR modulation of [Ca 2ϩ ] cyt -mediated DBS and the underlying mechanisms. We found that CaSR activation triggers Ca 2ϩ -dependent duodenal transepithelial HCO 3 Ϫ secretion, likely through the receptor-operated channels (ROCs), the intermediate-conductance Ca 2ϩ -activated K ϩ channels (IK Ca ), and the cystic fibrosis transmembrane conductance regulator (CFTR) channels. This study not only reveals that [Ca 2ϩ ] cyt signaling is critical to modulate DBS but also provides novel insights into the underlying molecular mechanisms of CaSR-induced Ca 2ϩ -dependent DBS.

EXPERIMENTAL PROCEDURES
Animal Study-The animal use protocol was approved by the University of California San Diego Committee on Investigations Involving Animal Subjects. All experiments were performed with adult Harlan C-57 black mice; homozygous CFTR knockout (CFTR Ϫ/Ϫ ) mice and their wild-type littermates (CFTR ϩ/ϩ ), which were established as described previously (13); and mice deficient in KCNQ1 (kcnq1 Ϫ/Ϫ ) and their wildtype littermates (kcnq1 ϩ/ϩ ), which were generated as described earlier (14).
Ussing Chamber Experiments in Vitro-The proximal duodenum removed from mice was immediately placed in ice-cold iso-osmolar mannitol with indomethacin (10 M) solution. The duodenal tissue was stripped of seromuscular layers and then mounted in the Ussing chambers (window area, 0.1 cm 2 ). Experiments were performed under continuous shortcircuited conditions (voltage current clamp, VCC 600, Physiologic Instruments, San Diego, CA), and luminal pH was maintained at 7.40 by the continuous infusion of 5 mM HCl under the automatic control of a pH-stat system (ETS 822, Radiometer America, Westlake, OH). Duodenal short circuit currents (I sc ) and HCO 3 Ϫ secretion were measured simultaneously as described previously (15). The rate of luminal bicarbonate secretion is expressed as micromolar per square centimeter per hour. The I sc was measured in microamperes and converted into Eq per square centimeters per hour. After basal parameters were measured for a 30-min period, CaSR activators were added to both the mucosal and serosal sides of the Ussing chamber because the CaSR has been identified on both the apical and basolateral membranes of epithelial cells (7,16). In some experiments, tissues were treated with inhibitors for 10 min after the baseline recording, followed by addition of CaSR activators. Electrophysiological parameters and bicarbonate secretion were recorded for a total of 90 min. During this experimental period, the vehicle did not significantly change I sc and HCO 3 Ϫ secretion, as shown in our previous control experiments (15). The mucosal solution used in Ussing chamber experiments contained the following: 140 mM Na ϩ , 5.4 mM K ϩ , 1. Epithelial Cell Culture-As described previously (17,18), SCBN, a duodenal epithelial cell line of canine origin (19), and Caco-2 and HEK-293 cells, human epithelial cell lines, were fed with fresh DMEM supplemented with 10% fetal bovine serum, L-glutamine, and streptomycin every 2-3 days. SW-480, a human colon cancer cell line, was fed with fresh L15 supplemented with 10% fetal bovine serum and streptomycin. After the cells had grown to confluence, they were replated onto 12-mm round coverslips (Warner Instruments Inc., Hamden, CT) and incubated for at least 24 h before use for [Ca 2ϩ ] cyt and pH i measurements.

Measurement of [Ca 2ϩ ] cyt in Epithelial Cells by Digital Ca 2ϩ
Imaging-[Ca 2ϩ ] cyt levels in epithelial cells were measured by digital Ca 2ϩ imaging as described previously (20). Cells grown on coverslips were loaded with 5 M Fura-2/AM in physiological salt solution, described below, at room temperature (ϳ22°C) for 50 min and then washed for 30 min. Thereafter, the coverslips with epithelial cells were mounted in a perfusion chamber on a Nikon microscope stage (Nikon Corp., Tokyo, Japan). The ratio of Fura-2/AM fluorescence with excitation at 340 or 380 nm (F 340/380 ) was followed over time and captured using an intensified charge-coupled device camera (ICCD200) and a MetaFluor imaging system (Universal Imaging Corp., Downingtown, PA). The physiological salt solution used in digital Ca 2ϩ measurement contained the following: 140 mM Na ϩ , 5 mM K ϩ , 2 mM Ca 2ϩ , 147 mM Cl 2Ϫ , 10 mM Hepes, and 10 mM glucose (pH 7.4). For the Ca 2ϩ -free solution, Ca 2ϩ was omitted, but 0.5 mM EGTA was added. The osmolality for all solutions was ϳ300 mosmol/kg of H 2 O.

Measurement of HCO 3
Ϫ Fluxes in SCBN Cells -pH i measurements in SCBN cells were applied as described previously (21). Briefly, cells grown on coverslips were incubated with 2 M 2Ј,7Ј-bis(2-carboxyethyl)-5-(and -6)carboxyfluorescein-AM in physiological salt solution, described above, for 30 min at room temperature and then washed for 30 min. The ratio of 2Ј,7Јbis(2-carboxyethyl)-5-(and -6)carboxyfluorescein fluorescence with excitation at 495 or 440 nm (F 495/440 ) was captured using an intensified charge-coupled device camera and a MetaFluor imaging system. The NaCl/HCO 3 Ϫ solutions contained the following: 120 mM NaCl, 25 mM NaHCO 3 , 2.5 mM K 2 HPO 4 , 1 mM MgSO 4 , 1 mM CaCl 2 , and 10 mM glucose equilibrated with 5% CO 2 /95% O 2 (pH 7.4). In Na ϩ -free (Na ϩ -free/HCO 3 Ϫ ) solutions, Na ϩ was replaced with N-methyl-D-glucamine. In HCO 3 Ϫ -free solutions, NaHCO 3 was replaced with NaCl (in NaCl/Hepes solution) or with N-methyl-D-glucamine (in Na ϩfree/Hepes solution). In experiments in which cells were acidified, 30 mM NH 4 C1 replaced an equal amount of N-methyl-Dglucamine. The ratio of the 2Ј,7Ј-bis(2-carboxyethyl)-5-(and -6)carboxyfluorescein fluorescence was calibrated in terms of pH i by incubating the cells in a high K ϩ solution (KCl replaced NaCl) and then permeabilizing the cells with 10 M nigericin. Then the pH of the bathing solution was stepped between pH 6.3 and 7.4. The F 495/440 was linear over this pH range. Cells were first perfused with either the NaCl/HCO 3 Ϫ or NaCl/Hepes solution in the chamber for 15 min to allow the pH i to stabilize. Then the cells were switched to Na ϩ -free/HCO 3 Ϫ or Na ϩ -free/ Hepes for 5 min to remove Na ϩ from the cells. The cells were then treated with the NH 4 -containing solution for 5 min, and when the NH 4 -containing solution was removed, cells were acidified to pH i 6.0 -6.5. Rates of pH i recovery after treatment with drugs were calculated by linear regression analysis between pH 6.0 and 6.5.
Western Blot Analysis-The specific anti-CaSR antibody used in this study is an affinity-purified monoclonal antibody raised against a synthetic peptide corresponding to the extracellular domain (residues 214 -235) of the human CaSR (Labome, Princeton, NJ). Its cross-reactivity with rodents, specificity, and applications have been described previously (7,22). A Western blot analysis of mouse duodenal mucosae and intestinal epithelial cells was applied as described previously (15). PVDF membranes (Millipore Corp., Billerica, MA) with resolved proteins were incubated with the anti-CaSR antibody or GAPDH (1:5000, Ambion, Austin, TX). After washing with PBS plus 1% Tween (PBST), the rabbit anti-mouse secondary antibody was applied to the membranes, which were treated with a chemiluminescent solution (Fivephoton Biochemicals, San Diego, CA) and then exposed to x-ray film. Densitometric analysis of the blots was performed with the use of an Alpha-Imager digital imaging system (Alpha Innotech, San Leandro, CA).
Immunohistochemistry-Immunohistochemistry was carried out as described previously (7). Briefly, the slides with duodenal tissues from C-57 mice or with intestinal epithelial cells were incubated with an anti-CaSR monoclonal antibody (1:100 dilution, Labome). The primary antibodies were detected with biotinylated goat anti-mouse IgG (Vector Laboratories, Burlingame, CA) secondary antibodies. Immunoreactivity was detected using a horseradish peroxidase (3Ј-,3Ј-diaminobenzidine) kit (BioGenex, San Francisco, CA) followed by counterstaining with hematoxylin, dehydration, and mounting. Slides were then examined with a Nikon Eclipse 800 Research microscope. To demonstrate the CaSR specificity of the antibody labeling, a control experiment was performed in which the primary antibody was omitted. All incubations were performed at room temperature.
Statistical Analysis-Results are expressed as mean Ϯ S.E. Differences between means were considered to be statistically significant at p Ͻ 0.05 using Student's t test or one-way analysis of variance followed by Newman-Keuls post hoc test, as appropriate.

Protein Expression of the CaSR in Mouse Duodenum Mucosal
Tissues-To examine CaSR expression in mouse duodenum mucosa, both Western blot and immunohistochemistry analyses were performed. As shown in our Western blot analysis ( Fig.  1), the antibody identified a significant band at ϳ120 -130 kDa in lysates of mouse duodenum mucosal tissues, indicating protein expression of the CaSR (8,23). Fig. 2A shows typical villous crypt structures of mouse duodenum mucosa with H&E staining. Fig. 2B shows representative images of CaSR immunohistochemistry in duodenum mucosa. Intense CaSR immunoreactivity (brown) was noted on both apical and basolateral membranes of the villous and crypt epithelial cells (Fig. 2B, right panel). However, no specific signal for the CaSR was observed when CaSR primary antibody was omitted (Fig. 2C). Therefore, cellular distribution and location of the CaSR in mouse duodenum mucosa was detected by immunohistochemistry.
Protein Expression of the CaSR in Intestinal Epithelial Cells-To examine CaSR expression in intestinal epithelial cells, both Western blot and immunohistochemistry analyses were performed on SCBN, SW-480, and Caco-2 cells, two human intestinal epithelial cell lines commonly used in the literature for physiological and pathological studies of intestinal ion transports. As shown by the Western blot analysis in Fig. 1, the antibody identified a strong band at ϳ120 -130 kDa in both SW480 and Caco-2 cells. However, the antibody identified one band of ϳ120 -130 kDa and another band of ϳ140 -150 kDa in the SCBN cell line, which is similar to previous reports (8,23), indicating CaSR protein expression in duodenal epithelial cells. However, our Western blot results show that the expression of CaSR protein is severalfold higher in human intestinal epithelial cells than in mouse duodenum mucosal tissues (Fig. 1), suggesting a higher CaSR expression in pure epithelial cells than in mucosal tissues that contain various cell types. To rule out the possible nonspecific staining of the CaSR in the tissues and cell lines, we also used parental HEK-293 cells as negative controls. Indeed, the antibody did not detect any CaSR expression in these cells (Fig. 1). Fig. 3 shows representative images of CaSR immunocytochemistry in these epithelial cells. Intense CaSR immunoreactivity was noted in intestinal epithelial SCBN, SW-480, and Caco-2 cells (Fig. 3, A, C, and E) but not in epithelial HEK-293 cells (Fig. 3G). No specific signal for the CaSR was observed when the CaSR primary antibody was omitted (Fig. 3, B, D, F, and H). Therefore, by immunocytochemistry, the CaSR was verified in intestinal epithelial cells, which is consistent with its presence in the epithelial cells of rat duodenum mucosa (24).
Role of the CaSR in Regulating Duodenal HCO 3 Ϫ Secretion and I sc -The CaSR has been functionally demonstrated along the entire gastrointestinal epithelium, where it plays an important role in the regulation of gastric acid and intestinal Cl Ϫ secretion. Therefore, in our initial studies, Ussing chamber experiments were conducted to test whether the CaSR is involved in duodenal mucosal ion transports, especially DBS. Because it is now evident that CFTR channels are essential for transepithelial HCO 3 Ϫ and Cl Ϫ secretion in most gastrointestinal epithelia (25,26), both CFTR knockout and wild-type mice were used to test whether CaSR activation can modulate duodenal I sc and HCO 3 Ϫ secretion. After basal I sc and HCO 3 Ϫ secretion were recorded for 30 min, two commonly used CaSR activators, spermine (1 mM) and Gd 3ϩ (0.5 mM), were added to both sides of the tissues because the CaSR is not restricted to one side of epithelial cells (7,16). As shown in Fig. 4, A and B, in both CFTR knockout and wild-type mice, spermine and Gd 3ϩ did not significantly affect duodenal basal I sc (p Ͼ 0.05, n ϭ 6). The net peak HCO 3 Ϫ secretion, calculated as the difference between the baseline and the peak value at 10 min, was used to describe the CaSR-activated HCO 3 Ϫ secretion. As shown in Fig. 4, C and D, both spermine and Gd 3ϩ markedly stimulated DBS in wildtype mice (p Ͻ 0.01, n ϭ 6), which was inhibited significantly by U73122 (10 M), a selective PLC inhibitor (p Ͻ 0.01, n ϭ 6). However spermine and Gd 3ϩ did not stimulate DBS in CFTR knockout mice (not significant, n ϭ 6). Therefore, these data   (30,31), markedly inhibited spermine-induced net peak DBS (p Ͻ 0.01, n ϭ 6). 2-APB (100 M), a commonly used blocker of ROC (32), also markedly inhibited spermine-induced net peak DBS (p Ͻ 0.01, n ϭ 6) (Fig. 5A). Moreover, when spermine-induced net peak DBS were compared between KCNQ1 knockout and wild-type mice, no significant differences were found between these two types of mice (NS, n ϭ 6) (Fig. 5B). Again, spermine (1 mM) did not significantly affect the basal duodenal I sc of both KCNQ1 knockout and wild-type mice (NS, n ϭ 6) (Fig. 5C). Therefore, our data indicate that Ca 2ϩ signaling, ROC, and IK Ca , but not cAMP signaling and KCNQ1, are involved in CaSR-mediated DBS. CaSR Activation-induced HCO 3 Ϫ Fluxes across SCBN Cells-Because expression and function of CFTR channels has been well established in SCBN cells (18), they are commonly used for studies of small intestinal epithelial ion transports (18,33,34). We first tested the role of CFTR in HCO 3 Ϫ fluxes in SCBN cells. To this end, cells were treated with NH 4 Cl in Na ϩ -free/HCO 3 Ϫ solution that caused the pH i first to increase (because of the entry of the weak base NH 3 ) and then to decrease when the NH 4 was washed from the bath. The cells remained acidic in the Na ϩ -free/HCO 3 Ϫ solution, in which the pH i was kept relatively stable but recovered when cells were returned to NaCl/HCO 3 Ϫ solution (Fig. 6A), likely because of the operation of Na ϩ /H ϩ exchange and other Na ϩ -and HCO 3 Ϫ -dependent pH i regulators. To test for the ability of HCO 3 Ϫ to permeate though CFTR, genistein (50 M), a commonly used CFTR activator (35), was added to cells that were acidified in Na ϩ -free/HCO 3 Ϫ solution.
We observed that pH i quickly recovered, and further recovery occurred after adding back NaCl/HCO 3 Ϫ solution (Fig. 6B). To examine whether genistein indeed activates HCO 3 Ϫ fluxes through the CFTR, cells were pretreated with CFTR inh -173 (10 M), a commonly used CFTR blocker (35), which reversed genistein-induced pH i recovery in Na ϩ -free/HCO 3 Ϫ solution (Fig. 6D). To test whether the genistein-activated, Na ϩ -independent recovery of pH i was HCO 3 Ϫ -dependent, these experiments were also repeated in Na ϩ -free and HCO 3 Ϫ -free Hepesbuffered solutions in which the acidified cells responded to genistein with only a slight effect on pH i , but a sustained pH i recovery occurred when Na ϩ was present in the Hepes solution (data not shown). Together, these findings are consistent with genistein-regulated HCO 3 Ϫ fluxes through the CFTR in the presence of extracellular HCO 3 Ϫ . To test for the role of the CaSR in modulating HCO 3 Ϫ fluxes through the CFTR, similar experiments were performed with spermine (1 mM), which was added to cells acidified in Na ϩ -   DECEMBER 12, 2014 • VOLUME 289 • NUMBER 50 JOURNAL OF BIOLOGICAL CHEMISTRY 34647 free/HCO 3 Ϫ solution. As shown in Fig. 6C, spermine, like genistein, induced pH ii recovery, which was reversed by 2-APB (100 M). Similarly, Gd 3ϩ (0.5 mM) induced pH i recovery in Na ϩfree/HCO 3 Ϫ solution (Fig. 6D). However, spermine and Gd 3ϩ did not induce a significant pH i recovery in Na ϩ -free, Hepesbuffered solutions (data not shown). These data from single cell studies are in agreement with those from a duodenal tissue study, indicating that ROC and CFTR channels are involved in CaSR-mediated transepithelial HCO 3 Ϫ secretion.

CaSR Activation Induces Ca 2ϩ Signaling in Epithelial Cells-
It is well documented that CaSR activation inhibits the intracellular cyclic AMP pathway in intestinal epithelial cells. However, little is known about Ca 2ϩ signaling downstream of CaSR activation in these cells. In addition, although it is known that Ca 2ϩ signaling is a critical regulator for DBS, so far Ca 2ϩ signaling in duodenal epithelial cells is poorly understood. We therefore monitored [Ca 2ϩ ] cyt changes in epithelial cells stimulated with different CaSR activators.
Following a short exposure to Ca 2ϩ -free solutions for 3 min, cells were superfused with different concentrations of [Ca 2ϩ ] o (1.0 -4.0 mM), which are close to the EC 50 (ϳ2.0 mM) for CaSR activation (23). Although [Ca 2ϩ ] o at 1.0 mM did not affect basal [Ca 2ϩ ] cyt , significant increases were seen when [Ca 2ϩ ] o increased to 4.0 mM (Fig. 7, A and B). Although [Ca 2ϩ ] o is an endogenous CaSR activator, it may enter healthy cells through the store-operated Ca 2ϩ entry pathway (32,36) or may even directly leak into unhealthy cells. We also used the CaSR activator spermine and found that it dose-dependently increased [Ca 2ϩ ] cyt in SCBN cells (Fig. 7, C and D). Moreover, both [Ca 2ϩ ] o -and spermine-induced [Ca 2ϩ ] cyt signaling was inhibited markedly by calhex 231 (3 M), a selective CaSR antagonist (Fig. 7, E and F). These results provide direct evidence for the CaSR-mediated increase in [Ca 2ϩ ] cyt in duodenal epithelial cells.
The CaSR is a member of the G protein-coupled receptor family, and its activation mobilizes different Ca 2ϩ sources in different cell types (4,5). Therefore, we sought to elucidate the mechanisms of [Ca 2ϩ ] cyt mobilization by CaSR activation in SCBN cells. To test whether ROCs are involved in CaSR activation, cells were superfused with spermine (3 mM) in the presence or the absence of 2 mM [Ca 2ϩ ] o . As shown in Fig. 8A (Fig. 7, C and  D). However, treatment with 2-APB (100 M) (Fig. 8, C and D) or SKF96365 (10 M) (Fig. 7, E and F), two commonly used ROC inhibitors (37), significantly inhibited spermine-induced  4 Cl in the solution caused the pH i first to increase and then to decrease when the NH 4 Cl was washed out. The cells remained acidic, and the pH i was relatively stable in Na ϩ -free/HCO 3 Ϫ solution, but the pH i began to recover when the cells were returned to NaCl/HCO 3 Ϫ solution (Na ϩ ). B, genistein-induced HCO 3 Ϫ fluxes through CFTR channels. The time course of pH i changes in SCBN cells was similar to the control in A. However, after the NH 4 Cl was washed out, genistein (Gen, 50 M) was added to the cells acidified in Na ϩ -free/HCO 3 Ϫ solution, and the pH i began to recover, but further recovery occurred after adding back NaCl/HCO solution (Na ϩ ). C, spermine-induced HCO 3 Ϫ fluxes through CFTR channels. The time course of pH i changes was similar to B, but spermine (Sper, 1 mM) was added to the cells acidified in Na ϩ -free/ [Ca 2ϩ ] 0 entry, indicating that the CaSR-mediated Ca 2ϩ entry pathway in SCBN cells involves the ROC.
The functional activity of the CaSR was also characterized in human intestinal epithelial cells. As shown in Fig. 9, [Ca 2ϩ ] o dose-dependently increased [Ca 2ϩ ] cyt in SW-480 and Caco-2 intestinal epithelial cells with CaSR expression (Fig. 9, A, B, and  D) but not in HEK-293 cells without CaSR expression (Fig. 9, C  and D), confirming functional expression of the CaSR in human intestinal epithelial cells. Moreover, in SW-480 cells, spermine did not alter [Ca 2ϩ ] cyt in [Ca 2ϩ ] 0 -free solutions but induced a marked [Ca 2ϩ ] cyt rise in [Ca 2ϩ ] 0 -containing solutions (Fig. 9, E and F), which was inhibited significantly by 2-APB (100 M) (Fig. 9, G and H), further indicating an important role of the ROC in CaSR-mediated Ca 2ϩ entry in human intestinal epithelial cells.

DISCUSSION
In this study, we demonstrated a novel role for the CaSR in controlling [Ca 2ϩ ] cyt signaling in duodenal epithelial cells to regulate Ca 2ϩ -dependent DBS and advance our understanding of the molecular mechanisms underlying CaSR-mediated [Ca 2ϩ ] cyt rise in these cells and Ca 2ϩ -dependent transepithelial HCO 3 Ϫ secretion. The CaSR is a member of the pheromone class of G-proteincoupled receptors that is expressed in a variety of tissues throughout the body and has been identified to mediate a wide array of physiological effects (3)(4)(5). In the parathyroid gland, it is responsible for regulating body calcium homeostasis by modulating the levels of parathyroid hormone and calcium in the circulation (2,38). Following the cloning of the CaSR from bovine parathyroid cells in 1993 (2), studies were conducted to determine the expression of the receptor. The CaSR has been shown to be expressed along the entire gastrointestinal tract, where it has many physiological roles, such as modulation of gastrin and gastric acid secretion, intestinal fluid, and ion transports by sensing the concentrations of electrolytes, amino acids, and polyamines (2,9,39). Although the CaSR has been cloned for two decades, only one previous study implicated its role in pancreatic HCO 3 Ϫ secretion (40), and another study suggested its possible involvement in L-glutamate-mediated DBS (41). So far, CaSR-me-   Ϫ secretion and the underlying molecular mechanisms are largely unknown.
The DBS is critical to defend the vulnerable duodenal epithelium against various aggressive factors (42,43). The importance of DBS in protecting duodenal mucosa has been confirmed in patients with duodenal ulcer whose acid-stimulated DBS is only 41% of that in healthy subjects (44). The DBS is impaired in the duodenal tissues from patients with cystic fibrosis, suggesting a pivotal role of the CFTR in mediating the DBS (45). Because the CaSR has been demonstrated to regulate gastric secretion and intestinal Cl Ϫ secretion, it is reasonable to infer that it may also modulate intestinal HCO 3 Ϫ secretion. We applied both CaSR agonists and antagonists in two models of duodenal mucosal tissues and intestinal epithelial cells and confirmed that CaSR activation indeed stimulates duodenal transepithelial HCO 3 Ϫ secretion, which is consistent with a previous observation that perfusion of Ca 2ϩ and spermine increased DBS in anesthetized rats (41). However, that study did not further test whether Ca 2ϩ and spermine stimulate the DBS through CaSR activation in the duodenum. Therefore, our study provides novel insights into the CaSR-mediated DBS.
Following our observation that CaSR activation induces DBS, we aimed to elucidate the underlying mechanisms, as established previously for pancreatic HCO 3 Ϫ secretion (40). We demonstrate that CaSR activation raises [Ca 2ϩ ] cyt in SCBN, SW-480, and Caco-2 cells, likely by evoking [Ca 2ϩ ] 0 entry through the ROC. The SCBN cell model was used in this study because this cell line is the only well characterized nontransformed duodenal epithelial cell line (17,18); because it expresses functional CFTR channels and has been used widely in the study of Ca 2ϩ -dependent Cl Ϫ secretion (18,33) Ϫ fluxes through the CFTR in duodenal epithelial cells. Because the physiological roles and molecular mechanisms of [Ca 2ϩ ] cyt -induced HCO 3 Ϫ secretion remain poorly understood in most epithelia (12), this study focuses on CaSR-mediated Ca 2ϩ signaling in intestinal epithelial cells. CaSR activity is demonstrated in canine duodenal epithelial SCBN cells and human intestinal epithelial SW-480 and Caco-2 cells with CaSR expression but not in human epithelial HEK-293 cells without CaSR expression. These data strongly support an important role of the CaSR in regulating the [Ca 2ϩ ] cyt -dependent function in both human and animal duodenal epithelial cells.
Although the CFTR has been thought to be principally activated by cyclic AMP, Ca 2ϩ signaling can activate the CFTR or potentiate cyclic AMP-mediated CFTR activation (12). The [Ca 2ϩ ] cyt elevation can stimulate mitochondrial ATP production, which is necessary for the process of epithelial HCO 3 Ϫ secretion (46). During the activation of CFTR, PKA uses ATP to phosphorylate and activate the R domain of CFTR (47). Therefore, the rise in [Ca 2ϩ ] cyt can activate apical CFTR channels. It is also known that a rise in [Ca 2ϩ ] cyt modulates the activities of Cl Ϫ /HCO 3 Ϫ exchangers, Na ϩ /H ϩ exchangers, and Na ϩ -HCO 3 Ϫ cotransport in epithelial cells (20, 48 -51). The Ca 2ϩ -activated chloride channel has been suggested to contribute to HCO 3 Ϫ secretion in some epithelia (12). We reported previously that [Ca 2ϩ ] cyt activates basolateral IK Ca in murine duodenal epithelium to provide a driving force for HCO 3 Ϫ secretion (15). This study combining selective pharmacological inhibitors and knockout mice is in good agreement with other reports on the CaSR-Ca 2ϩ -IK Ca pathway in the vascular system (52), further supporting our pervious notion that IK Ca plays an essential role in Ca 2ϩ -mediated DBS (15). All of these actions of [Ca 2ϩ ] cyt in epithelial cells may contribute to the molecular mechanisms underlying Ca 2ϩ -mediated transepithelial HCO 3 Ϫ secretion.  However, our data demonstrate that CaSR-[Ca 2ϩ ] cyt -IK Ca -CFTR is a major pathway involved in CaSR-mediated DBS observed in this study (Fig. 10).
It is generally assumed that the regulatory mechanisms involved in intestinal epithelial HCO 3 Ϫ and Cl Ϫ secretion are similar, but this notion has never been fully studied and confirmed. Epithelial HCO 3 Ϫ and Cl Ϫ secretion is mainly under the control of cyclic AMP and Ca 2ϩ signaling, which may interact and cross-talk to regulate epithelial ion transports (25,42,53,54). Previous studies demonstrated that most well known secretagogues, such as forskolin, ACh, 5-HT, and PGE 2 , usually stimulate intestinal HCO 3 Ϫ and Cl Ϫ secretion in parallel (20,42,55,56). It is not known, however, whether epithelial HCO 3 Ϫ and Cl Ϫ secretion has to occur in parallel and whether they are regulated by the same or different signaling/mechanisms. Notably, estrogen inhibits forskolin-and carbachol-induced rat colonic Cl Ϫ secretion (57). However, we revealed that estrogen stimulates DBS in humans and mice likely through Ca 2ϩ signaling without altering basal duodenal I sc , an index primarily of epithelial Cl Ϫ secretion (13,58,59). Therefore, estrogen may play different roles in regulating intestinal HCO 3 Ϫ and Cl Ϫ secretion. These findings suggest that epithelial HCO 3 Ϫ and Cl Ϫ secretion may not be always triggered in parallel and they may not be regulated by the same signaling/mechanism. We therefore propose that different regulatory mechanisms may exist for intestinal HCO 3 Ϫ and Cl Ϫ secretion. Ca 2ϩ signaling may play a key role in HCO 3 Ϫ secretion, but cyclic AMP may play a major role in Cl Ϫ secretion. Indeed, in this study, CaSR activation, resulting in an increase in [Ca 2ϩ ] cyt but a decrease in intracellular cyclic AMP (4 -6, 9), leads to a specific DBS without simultaneously altering duodenal I sc . Moreover, IK Ca rather than cyclic AMP-activated K ϩ channels (KCNQ1) are found to be involved in CaSR-mediated DBS, indicating that a sole Ca 2ϩ signaling in the absence of cyclic AMP can trigger the DBS. Therefore, this study not only confirms the pivotal role of [Ca 2ϩ ] cyt as primary signaling in transepithelial HCO 3 Ϫ secretion but also further supports our notion that intestinal HCO 3 Ϫ and Cl Ϫ secretion can be triggered independently by different signaling/mechanisms (13).
What is the physiological relevance of this study? Food nutrients, such as dietary calcium, spermine, and L-amino acids, are CaSR activators that regulate gastric acid secretion, intestinal fluid, and ion transports. Here we confirmed a novel physiological role of these nutrients, namely DBS stimulation, and elucidated the underlying mechanisms. Because the DBS is critical for duodenal mucosal protection, these dietary CaSR modulators may also be involved in this physiological process through the [Ca 2ϩ ] cyt -IK Ca -CFTR cascade (Fig. 10). Because CaSR-mediated Ca 2ϩ signaling can also stimulate Ca 2ϩ -activated chloride channel-dependent epithelial secretion that is independent of the CFTR, these CaSR modulators might be used to restore fluid secretion defects in cystic fibrosis disease (60). Moreover, understanding whether different cell signaling triggers distinct intestinal epithelial ion secretion is important for the development of better drugs that can specifically target either intestinal the HCO 3 Ϫ or Cl Ϫ secretion pathway. The medications that specifically trigger intestinal HCO 3 Ϫ secretion to protect gastrointestinal tract would not increase Cl Ϫ secretion, which might induce diarrhea. Also, the medications that specifically inhibit intestinal Cl Ϫ secretion to treat diarrhea would not reduce HCO 3 Ϫ secretion, which might induce gastrointestinal injury.

CONCLUSION
On the basis of this study, we conclude that dietary calcium and spermine could activate the CaSR in duodenal epithelial cells to specifically trigger Ca 2ϩ -dependent DBS that protects mucosa, CaSR activation-induced Ca 2ϩ entry through ROC is critical to trigger DBS, and Ca 2ϩ signaling regulates DBS, likely through activation of IK Ca and CFTR channels. This study not only reveals that [Ca 2ϩ ] cyt signaling is critical for CaSR-induced DBS but also provides novel insights into the molecular mechanisms of [Ca 2ϩ ] cyt signaling-mediated transepithelial HCO 3 Ϫ secretion.