Introduction

Although there is an abundant amount of magnesium (Mg2+) within human cells, and it has vital roles in numerous biological functions [1], knowledge regarding regulatory mechanisms of Mg2+ homeostasis is still minimal. Theoretically, plasma Mg2+ level is regulated within a narrow range by the synergistic actions of intestinal absorption, bone and soft tissue storage, and renal excretion [1]. Since dietary intake is the only source of Mg2+, intestinal absorption is vital for normal Mg2+ homeostasis. However, the understanding of the regulatory mechanism of intestinal Mg2+ absorption is still elusive. Principally, enterocyte epithelium absorbs Mg2+ via both saturable transcellular and non-saturable paracellular transport [1]. Approximately 90% of total intestinal Mg2+ uptake is processed through a Mg2+ channel-independent paracellular passive mechanism which exclusively occurs in the small intestine [1,2,3]. The Mg2+ channel-dependent transcellular active Mg2+ uptake plays an important role during low dietary Mg2+ intake [3].

In the small intestine, enterocyte epithelial cells are equipped with acid sensors, e.g., acid-sensing ion channel (ASIC), ovarian cancer G protein-coupled receptor 1 (OGR1), and transient receptor potential vanilloid (TRPV) that are implicated in mucosal defense by detecting mucosal protons and stimulating mucosal HCO3− secretion [4,5,6,7,8,9]. In addition to this mucosal defense, intestinal acid sensors also regulate ion transport across the enterocyte epithelium. Reiter et al. [10] reported that TRPV4 enhanced transcellular K+ transport and paracellular permeability through Ca2+ signaling in HC11 epithelial monolayers. OGR1 enhanced Mg2+ absorption in intestinal epithelium-like Caco-2 monolayers through protein kinase C (PKC) signaling [11, 12]. On the other hand, an activation of ASIC1a led to a suppression of Mg2+ absorption across Caco-2 monolayers through a Ca2+-signaling dependent pathway [12]. In addition to acid sensors, the purinergic P2Y2 receptor is also involved in mucosal acid sensing and defense of duodenocytes [8]. Activation of Gq-associated P2Y2 stimulated duodenal mucosal HCO3− secretion in a Ca2+ signaling-dependent mechanism [4]. Besides HCO3− secretion, the P2Y2 receptor also regulates Na+, K+, and Cl− transport in various epithelial tissues [13]. However, the role of P2Y on intestinal Mg2+ absorption is still unknown.

PPI-induced hypomagnesemia (PPIH) has been reported since 2006 [14,15,16,17,18,19]. Suppression of intestinal Mg2+ absorption is a major underlying mechanism of PPIH in chronic users and animal models [14, 15, 17, 19,20,21]. We previously reported that a higher apical HCO3− secretion contributed to a suppression of intestinal Mg2+ absorption in PPIH [21]. Apical acidity in the small intestine [22] required for stabilizing mineral solubility [23] and stimulates intestinal Mg2+ absorption [11, 12, 24]. An increase in luminal pH from ~ 5 to 7.8 led to a decrease in soluble Mg2+ from 79.61 to 8.71% [25]; thus, secreted HCO3− increases luminal pH and subsequently suppresses Mg2+ absorption. Omeprazole, the most common PPI, significantly enhanced HCO3− secretion in human duodenum and intestinal epithelium-like Caco-2 monolayers [12, 26]. Antagonists of mucosal HCO3− secretion markedly increased duodenal Mg2+ absorption in PPIH rats [21]. P2Y2 regulated mucosal HCO3− secretion [4], but the involvement of P2Y2 on apical HCO3− secretion and Mg2+ absorption in PPI-treated intestinal epithelium is unknown.

Therefore, the present study aimed to investigate the role of purinergic P2Y receptors in the regulation of Mg2+ absorption in normal and omeprazole-treated intestinal-like Caco-2 monolayers. Caco-2 cells express P2Y receptors, i.e., P2Y2, P2Y4, and P2Y6 [27, 28]. They are also equipped with apical HCO3− secretion transporting machineries, e.g., cystic fibrosis transmembrane conductance regulator (CFTR) and Na+-HCO3− cotransporter-1 (NBCe1), which are modulated by parathyroid hormone (PTH) and HCl [12, 29]. In addition, Caco-2 monolayers have been used as a model for studying the regulation of intestinal Mg2+ absorption [11, 12, 30, 31].

Methods

Cell culture

Caco-2 cells (ATCC No. HTB-37, passage 25–40th) were grown in Dulbecco’s modified Eagle medium (DMEM) (Sigma, St. Louis, MO, USA) supplemented with 12.5% fetal bovine serum (FBS-Gold) (PAA Laboratories GmbH, Pasching, Austria), 1% l-glutamine (Gibco, Grand Island, NY, USA), 1% non-essential amino acid (Sigma), and 1% antibiotic–antimycotic solution (Gibco) and maintained in a humidified atmosphere containing 5% CO2 at 37 °C. Culture medium was changed 3 times a week. For epithelial electrical parameter measurement and Mg2+ flux studies, the monolayers were developed by seeding cells (5.0 × 105 cells/cm2) onto permeable polyester Snapwell™ inserts (12 mm diameter and 0.4 µm pore size filter) (Corning, Corning, NY, USA). After being maintained for 14 days, the Snapwell was inserted into a Ussing chamber (World Precision Instrument, Sarasota, FL, USA). For HCO3− secretion studies, the cells were plated (5.0 × 105 cells/cm2) onto permeable polyester Transwell-clear inserts (Corning) and maintained for 14 days. For western blot analysis, cells were plated (5.0 × 105 cells/well) on 6-well plates (Corning) and maintained for 14 days.

In the omeprazole-treated group, Caco-2 monolayers were grown in a culture medium containing omeprazole (Calbiochem, San Diego, CA, USA), from day 7 to day 14 of culture [31], at concentrations of 200 and 400 ng/ml that resembled those found in human plasma [32].

Bathing solutions

The physiological bathing solution contained (in mM) 118 NaCl, 4.7 KCl, 1.1 MgCl2, 1.25 CaCl2, 23 NaHCO3, 12 d-glucose, 2.5 l-glutamine, and 2 mannitol.

For the apical to basolateral total Mg2+ transport studies, the apical solution contained (in mM) 40 MgCl2, 1.25 CaCl2, 4.5 KCl, 12 d-glucose, 2.5 l-glutamine, 115 d-mannitol, and 10 HEPES, whereas the basolateral solution contained (in mM) 1.25 CaCl2, 4.5 KCl, 12 d-glucose, 2.5 l-glutamine, 250 d-mannitol, and 10 HEPES.

For apical HCO3− secretion experiments, the composition of the NaHCO3-free apical solution was as follows (in mM): 1.25 CaCl2, 4.5 KCl, 1 MgCl2, 12 d-glucose, 2.5 l-glutamine, 230 d-mannitol, and 10 HEPES; and the NaHCO3-containing basolateral solution contained (in mM) 25 NaHCO3, 1.25 CaCl2, 4.5 KCl, 1 MgCl2, 12 d-glucose, 2.5 l-glutamine, 200 d-mannitol, and 10 HEPES.

All solutions were continuously gassed with humidified 5% CO2 in 95% O2, maintained at 37 °C, pH 7.4, and had an osmolality of 290–295 mmol kg−1 water as measured by a freezing-point depression-based Fiske® micro-osmometer (model 210; Fiske® Associates, Norwood, MA, USA). All chemicals were purchased from Sigma.

Transepithelial electrical resistance

Snapwell™ inserts containing Caco-2 monolayers were rinsed gently, mounted in a Ussing chamber, and bathed on both sides with physiological bathing solution. Transepithelial potential difference (PD) and short-circuit current (Isc) were determined by Ag/AgCl electrodes and an epithelial voltage/current clamp apparatus (model ECV-4000; World Precision Instrument) as previously described [33]. Transepithelial electrical resistance (TEER) was calculated from PD and Isc by Ohm’s law.

Mg2+ transport study

Caco-2 monolayers were rinsed and mounted in a Ussing chamber as described above. After being equilibrated in physiological bathing solution for 15 min, total Mg2+ transport studies were performed by substituting the physiological bathing solution with apical and basolateral bathing solutions for Mg2+ transport. To investigate the Mg2+ channel-independent Mg2+ transport, apical sites of Caco-2 monolayers from the same passage and culture plate were pre-incubated for 10 min with Mg2+-channel inhibitor Cobalt(III)hexaammine [Co(III)hex, Table 1], which suppressed Mg2+ influx in Caco-2 epithelium and blocked Mg2+ channel-dependent Mg2+ transport. After that the apical and basolateral solutions were substituted with bathing solution for Mg2+ transport. At 30, 60, and 120 min after solution replacements, 50 μl of solution was collected from the basolateral side, as well as from the apical side. The Mg2+ concentration and the rate of Mg2+ flux were determined by the method of Thongon and Krishmanra [11]. The rate of Mg2+ channel-dependent Mg2+ transport was calculated by subtracting the rate of Mg2+ channel-independent Mg2+ transport from the rate of total Mg2+ transport. However, Co(III)hex might somehow interfered Mg2+ channel-independent paracellular Mg2+ transport, which was not demonstrated in the present study.

Table 1 Agonist, antagonist, or chelator used in the study of Mg2+ transport and HCO3− secretion across Caco-2 monolayers
Full size table

In some experiments the monolayers were pre-incubated with agonists or antagonist, as demonstrated in Table 1, for 40 min prior to performing experiments.

Measurements of HCO3 − secretion

Apical HCO3− secretion was studied by the modified method of Thongon et al. [12]. The Caco-2 monolayer was gently rinsed 3 times and incubated for 15 min in the physiological bathing solution. Then, apical and basolateral solutions were substituted with bathing solutions for HCO3− secretion. The apical membrane-bound carbonic anhydrase (CA) activity was suppressed by the selective CA IX inhibitor U-104 (Table 1). After 20 min, HCO3− secretion was stimulated by adding MRS2768 or forskolin (Table 1) and incubation proceeded for 5 min. After removal of the MRS2768- or forskolin-containing solutions, the monolayers were gently rinsed 3 times and further incubated for 25 min. Aliquots of apical solution at various time points (Fig. 5) were individually sampled. The concentration of HCO3− was immediately determined as previously described [12].

In some experiments the monolayers were pre-incubated with agonists or antagonist, as demonstrated in Table 1, for 40 min prior to performing experiments.

Western blot analysis

Western blot analysis was performed as previously described [11]. In brief, protein samples of Caco-2 cells were prepared by using Piece® Ripa Buffer (Thermo Fisher Scientific Inc., Rockford, IL, USA). Protein samples (35 μg each) were separated on 12.5% SDS-PAGE gels, and then transferred onto nitrocellulose membranes (Amersham, Buckinghamshire, UK) by electroblotting. Membranes were blocked and probed overnight at 4 °C with 1:1000 rabbit polyclonal antibodies raised against human P2Y2 receptor and CFTR (Santa Cruz Biotechnology, Santa Cruz, CA). Membranes were also re-probed with actin monoclonal antibodies (Santa Cruz Biotechnology) diluted at 1:5000. After 2 h incubation at 25 °C with goat anti-rabbit IgG-HRP-conjugated secondary antibodies (Santa Cruz Biotechnology) diluted at 1:10,000, blots were visualized by Thermo Scientific SuperSignal® West Pico Substrate (Thermo Fisher Scientific Inc.) and captured on CL-XPosure Film (Thermo Fisher Scientific Inc.). Densitometric analysis was performed using ImageJ for Mac Os X.

Statistical analysis

Results were expressed as mean ± SE. Two sets of data were compared using unpaired Student’s t-test. One-way analysis of variance (ANOVA) with Dunnett’s post test was used for comparison of multiple sets of data. The level of significance was P < 0.05. All data were analyzed by GraphPad Prism (GraphPad Software Inc., San Diego, CA, USA).

Results

Omeprazole modulated paracellular Mg2+ transport

Previous Mg2+ transport kinetic analysis demonstrated that omeprazole selectively impeded non-saturable passive Mg2+ transport but not saturable active Mg2+ transport across Caco-2 monolayers [31]. By using competitive Mg2+-channel inhibitor Co(III)hex (Fig. 1), we observed total (white bars), Mg2+ channel-independent (gray bars), and Mg2+ channel-dependent Mg2+ transport (black bars). The results showed that 200 and 400 ng/ml omeprazole significantly suppressed total (Fig. 1d) and Mg2+ channel-independent Mg2+ transport (Fig. 1e) compared to the control group. Furthermore, we had studied the effect of omeprazole on Mg2+ channel-independent Mg2+ transport. Previously, it was reported that protein kinase C activator carbachol (CCh) could increase paracellular permeability and decrease TEER in Caco-2 monolayers [34, 35]. In control (Fig. 1a), 200-ng/ml omeprazole-treated group (Fig. 1b), and 400-ng/ml omeprazole-treated group (Fig. 1c), CCh significantly increased total and Mg2+ channel-independent Mg2+ transport compared to the corresponding vehicle-treated group. The rates of Mg+ channel-dependent Mg2+ transport of all experiments were not changed (Fig. 1a–c, f).

Fig. 1
figure 1

Effect of omeprazole on Mg2+ transport across Caco-2 monolayers. The rate of Mg2+ transport across control (a), 200-ng/ml omeprazole-treated (b), and 400-ng/ml omeprazole-treated Caco-2 monolayers (c). White bar; total Mg2+ transport, gray bar; Mg2+ channel-independent Mg2+ transport, black bar; Mg2+ channel-dependent Mg2+ transport. *P < 0.05, **P < 0.01, ***P < 0.001 compared with its corresponding CCh-untreated group. (n = 6). The rate of total (d), Mg2+-channel independent (e), and Mg2+-channel dependent (f) Mg2+ transport of control and omeprazole-treated Caco-2 monolayers. *P < 0.05, **P < 0.01, ***P < 0.001 compared with its corresponding control group (n = 6)

Full size image

We also studied TEER to confirm the permeability of Caco-2 monolayers. As demonstrated in Fig. 2c, Co(III)hex, which impeded Mg2+ channel-dependent Mg2+ transport, had no effect on TEER of control and omeprazole-treated monolayers when compared to its corresponding vehicle-treated monolayers. On the other hand CCh, which increased Mg2+ channel-independent Mg2+ transport, significantly decreased TEER of control and omeprazole-treated (Fig. 2c) Caco-2 monolayers. We further observed the involvement of Ca2+-activated K+ and Ca2+-activated Cl− channels on CCh-suppressed TEER by using charybdotoxin (ChTX) and benzbromarone. ChTX and benzbromarone had no effect on CCh-suppressed TEER of control and omeprazole-treated Caco-2 monolayers. However, CCh might somehow modulate some ion channels or transports, which could affect TEER of Caco-2 monolayers. TEER of 400-ng/ml omeprazole-treated monolayers (458.56 ± 14.97 Ω cm2) was significantly higher than that of vehicle-treated control monolayers (362.31 ± 17.69 Ω cm2, P = 0.0003) (Fig. 2c). This series of experiments suggested that alteration of total Mg2+ transport across Caco-2 monolayers was the result of the modulation of Mg2+ channel-independent Mg2+ transport. These results also agreed with a previous study [31] that omeprazole exclusively suppressed nonsaturable passive Mg2+ transport in intestinal epithelium-like Caco-2 monolayers. In addition, TEER could determine the change of Mg2+ channel-independent Mg2+ transport, but not Mg2+ channel-dependent Mg2+ transport. Thus, Mg2+ channel-dependent Mg2+ transport was ignored in the rest of the experiments.

Fig. 2
figure 2

Transepithelial electrical resistance (TEER). TEER of control and 400-ng/ml omeprazole-treated Caco-2 monolayers in various experiments. *P < 0.05, **P < 0.01 compared with its corresponding vehicle-treated group (n = 6). Benzb benzbromarone

Full size image

P2Y2 receptor modulated Mg2+ transport

To observe the roles of P2Y2, P2Y4, and P2Y6 activities on Mg2+ transport across Caco-2 monolayers we incubated the monolayers with selective agonists of P2Y2, P2Y4, or P2Y6 receptors (Table 1). In control monolayers (Fig. 3a) the rate of Mg2+ transport (in nmol/h/cm2) of the P2Y2 agonist-treated group (81.14 ± 3.45), but not P2Y4 or P2Y6 agonist-treated groups, was significantly lower than that of the vehicle-treated group (138.89 ± 4.85). As demonstrated in Fig. 3b, the rate of Mg2+ transport (in nmol/h/cm2) of the P2Y2 agonist-treated group (41.94 ± 5.91) was significantly lower than that of vehicle-treated group (106.74 ± 5.12) in the 200-ng/ml omeprazole-treated condition. In 400-ng/ml omeprazole-treated monolayers (Fig. 3c), Mg2+ transport (in nmol/h/cm2) of the P2Y2 agonist-treated group (35.28 ± 2.72) was also significantly suppressed compared to the vehicle-treated group (95.24 ± 4.96). When compared to the corresponding vehicle-treated group, P2Y2 agonist suppressed the rate of Mg2+ transport by about 41.58, 60.71, and 62.96% of control, in the 200-ng/ml omeprazole-treated, and 400-ng/ml omeprazole-treated monolayers, respectively. In addition, TEER of MRS2768 treated monolayers was significantly higher than that of the corresponding vehicle-treated control (Fig. 2a) or omeprazole-treated monolayers (Fig. 2b). We further performed a western blotting study to confirm the enhancing effect of omeprazole on P2Y2 receptor activation in suppressing intestinal Mg2+ absorption. As demonstrated in Fig. 3d, 200 and 400 ng/ml omeprazole significantly enhanced P2Y2 expression when compared to the control cells.

Fig. 3
figure 3

The effect of P2Y receptor agonists on Mg2+ transport across Caco-2 monolayers. The rate of Mg2+ transport across control (a), 200-ng/ml omeprazole-treated (b), and 400-ng/ml omeprazole-treated Caco-2 monolayers (c) with agonist of P2Y2 receptor MRS2768, P2Y4 receptor MRS4062, and P2Y6 receptor MRS2693 pre-incubations. Representative immunoblotting and densitometric analysis of P2Y2 expression in control and omeprazole-treated Caco-2 cells (d). ***P < 0.001 compared with its vehicle-treated group (n = 6)

Full size image

Signaling pathway of P2Y2 receptor activation suppressed Mg2+ transport

This series of experiments aimed to observe the underlying mechanism by which P2Y2 receptor activation mediated the suppression of intestinal Mg2+ absorption. In control (Fig. 4a) and omeprazole-treated Caco-2 monolayers (Fig. 4b), MRS2768 significantly suppressed the rate of total Mg2+ transport. P2Y2 receptor antagonist, PLC antagonist, IP3 receptor antagonist, and intracellular Ca2+ chelator markedly normalized the effect of MRS2768 on Mg2+ transport. However, the antagonist of PKC, MEK1/2, PI3K, PKA, or voltage-gated Ca2+ channel had no effect on MRS2768-suppressed Mg2+ transport in either control or omeprazole-treated Caco-2 monolayers (Fig. 4a–b). In the TEER study, P2Y2 receptor antagonist, PLC antagonist, IP3 receptor antagonist, and intracellular Ca2+ chelator also normalized the effect of MRS2768-increased TEER of control (Fig. 2a) and omeprazole-treated (Fig. 2b) monolayers. These results suggested that P2Y2 receptor activation mediated the suppression of intestinal Mg2+ absorption through PLC, IP3 receptor, and intracellular Ca2+ signaling pathway.

Fig. 4
figure 4

The signaling pathway of P2Y receptor activation suppressed Mg2+ transport across Caco-2 monolayers. The rate of Mg2+ transport across control (a) and 400-ng/ml omeprazole-treated Caco-2 monolayers (b) with agonist or antagonist pre-incubations. ***P < 0.001 compared with its vehicle-treated group (n = 6)

Full size image

Contribution of HCO3 − secretion on P2Y2 receptor activation suppressed Mg2+ transport

Previously, we reported the contribution of mucosal HCO3− secretion on omeprazole-suppressed duodenal Mg2+ absorption in PPIH rats [21]. Then, we further studied the contribution of HCO3− secretion on P2Y2 receptor activation-suppressed Mg2+ transport in Caco-2 monolayers. As demonstrated in Fig. 5a and b, P2Y2 agonist significantly suppressed Mg2+ transport in control and omeprazole-treated monolayers. The antagonist of NBCe1, CFTR, and CA could relieve the inhibitory effect of P2Y2 activation on Mg2+ transport across control and omeprazole-treated monolayers (Fig. 5a–b). By using HCO3−-free bathing solution in both apical and basolateral sites, the P2Y2 agonist had no effect on Mg2+ transport in normal Caco-2 monolayers (Fig. 5a). In omeprazole-treated monolayers (Fig. 5b), the HCO3−-free condition significantly increased Mg2+ transport when compared to its corresponding vehicle-treated group. Under the HCO3−-free condition, the P2Y2 agonist also had no effect on Mg2+ transport in omeprazole-treated Caco-2 monolayers (Fig. 5b). Moreover, the antagonist of CFTR, CA, and NBCe1 also normalized the effect of P2Y2 activation-increased TEER of control (Fig. 2a) and omeprazole-treated (Fig. 2b) Caco-2 monolayers. Therefore, apical HCO3− secretion was involved in P2Y2 receptor activation-suppressed Mg2+ transport.

Fig. 5
figure 5

Contribution of HCO3− secretion on P2Y receptor activation suppressed Mg2+ transport across Caco-2 monolayers. The rate of Mg2+ transport across control (a), and 400-ng/ml omeprazole-treated Caco-2 monolayers (b) with agonist or antagonist pre-incubations. *P < 0.05, ***P < 0.001 compared with its vehicle-treated group (n = 6)

Full size image

P2Y2 receptor activation stimulated HCO3 − secretion

Previously, it was reported that P2Y2 stimulated mucosal HCO3− secretion in rat duodenum [4]. The present study observed the effect of P2Y2 receptor activation on apical HCO3− secretion in Caco-2 monolayers. Since forskolin stimulated duodenum HCO3− secretion in mice [36] and apical HCO3− secretion in Caco-2 monolayers [29], we used forskolin as positive control for the stimulation of apical HCO3− secretion. Our results showed that P2Y2 agonist MRS2768 and forskolin significantly increased the rate of apical HCO3− secretion by control (Fig. 6a, b), 200-ng/ml omeprazole-treated (Fig. 6c, d), and 400-ng/ml omeprazole-treated (Fig. 6e, f) monolayers. Under the HCO3−-free condition, the rate of basal, forskolin-stimulated, and MRS2768-stimulated HCO3− secretions were suppressed in control and omeprazole-treated Caco-2 monolayers (Fig. 6b, d, f). In the pre-stimulating condition (Fig. 6g), the basal HCO3− secretions (in µmol/h/cm2) of 200-ng/ml omeprazole-treated (4.45 ± 0.46) and 400-ng/ml omeprazole-treated (5.19 ± 0.49) monolayers were significantly higher than that of the control monolayers (2.49 ± 0.41). The rate of peak HCO3− secretions (in µmol/h/cm2) in forskolin-stimulated and MRS2768-stimulated conditions (Fig. 6h) of 200-ng/ml omeprazole-treated (12.48 ± 0.46 and 13.01 ± 0.57, respectively) and 400-ng/ml omeprazole-treated (14.12 ± 0.44 and 14.59 ± 0.48, respectively) monolayers were significantly higher than those of the control monolayers (9.16 ± 0.68 and 8.66 ± 0.57, respectively). We further observed the expression of CFTR protein in omeprazole-treated monolayers. As demonstrated in Fig. 7a, omeprazole had no effect on CFTR protein expression in Caco-2 monolayers.

Fig. 6
figure 6

P2Y2 receptor agonist stimulated HCO3− secretion. Time course and rate of apical HCO3− secretion by control (a, b), 200-ng/ml omeprazole-treated (c, d), and 400-ng/ml omeprazole-treated Caco-2 monolayers (e, f) that were induced by forskolin or MRS2768. Basal (at 20 min; g) and peak forskolin- or MRS2768-stimulated HCO3− secretion (at 33 min; h) by control or omeprazole-exposed Caco-2 monolayers. *P < 0.05, **P < 0.01, ***P < 0.001 compared with the corresponding control group. †P < 0.05, ††P < 0.01, †††P < 0.001 compared with the pre-treated control group (n = 6)

Full size image
Fig. 7
figure 7

The signaling pathway of P2Y2 receptor agonist stimulated HCO3− secretion. Representative immunoblotting and densitometric analysis of CFTR expression in control and omeprazole-treated Caco-2 cells (a). The rate of peak MRS2768-stimulated HCO3− secretion by control (b), and 400-ng/ml omeprazole-treated Caco-2 monolayers (c) that were induced by agonist or antagonist. The pH of apical culture media of Caco-2 monolayers at 24 h after culture media change (d). *P < 0.05, **P < 0.01, ***P < 0.001 compared with the corresponding vehicle group (n = 6)

Full size image

Signaling pathway of P2Y2 receptor activation stimulated HCO3 − secretion

Previously, it was found that P2Y2 activation enhanced mucosal HCO3− secretion through PLC, IP3 receptor, and intracellular Ca2+ signaling pathway [4]. This series of experiments, therefore, showed the underlying signaling transduction pathway of P2Y2 receptor activation that mediated the stimulation of apical HCO3− secretion in Caco-2 monolayers. In control (Fig. 7b) and omeprazole-treated monolayers (Fig. 7c), the rate of peak MRS2768-stimulated HCO3− secretion was significantly suppressed in the monolayers treated with the antagonist of P2Y2 receptor, PLC, PI3K receptor, CFTR, NBCe1, and CA, as well as intracellular Ca2+ chelator. These results showed that P2Y2 receptor activation mediated the stimulation of HCO3− secretion through PLC, IP3 receptor, and intracellular Ca2+ signaling pathway. We further observed the pH of apical culture media of Caco-2 monolayers at 24 h after culture media change. MRS2768 and 400 ng/ml omeprazole significantly increased apical pH compared to vehicle treated group (Fig. 7d). CFTR and NBCe1 antagonists significantly abolished MRS2768 and omeprazole effects on apical pH.

Discussion

Intestinal Mg2+ absorption can be processed through saturable transcellular and non-saturable paracellular mechanisms. Transcellular Mg2+ transport is an active process that requires the activity of transient receptor potential melastatin 6 (TRPM6), TRPM7, and basolateral Na+/Mg2+ exchanger [1, 37,38,39] or other transport pathways. Paracellular Mg2+ transport is a passive mechanism modulated by the tight junction associated Claudin (Cldn) [11, 40]. However, the regulatory mechanism of intestinal Mg2+ absorption is largely unknown.

Our previous study demonstrated that intestinal-associated proton sensors ASIC1a and OGR1 could modulate intestinal-like Mg2+ transport across Caco-2 monolayers [12]. OGR1 activation increased Mg2+ transport across Caco-2 monolayers while ASIC1a activation decreased it. In the present study we focused on the role of the purinergic P2Y1 receptor family, which was regulating trans-epithelial Na+, K+, and Cl− transport [13], on Mg2+ transport across Caco-2 monolayers. The Gq-coupled P2Y1 receptor family is composed of P2Y1, P2Y2, P2Y4, P2Y6 and P2Y11 receptors. However, Caco-2 monolayers expressed P2Y2, P2Y4, and P2Y6 [27], which are exclusively localized in the apical membrane [28]. Our results suggested that only P2Y2 was involved in the modulation of Mg2+ transport across Caco-2 monolayers. Generally, P2Y2 receptors regulate epithelial ion transport through Gq-dependent pathways which activate PLC and stimulate intracellular Ca2+ mobilization [4, 13]. In agreement with our results, P2Y2 receptor activation suppressed Mg2+ transport through PLC, IP3 receptor, and the intracellular Ca2+ signaling pathway. Li et al. [41] reported the inhibitory role of P2Y2 receptor on Cldn-1 expression. Purinergic P2Y receptor agonist adenosine triphosphate (ATP) rapidly suppressed epithelial paracellular permeability and increased epithelial TEER [42]. In addition, purinergic P2 receptor agonist also suppressed the activity of TRPM6 and TRPM7 [43, 44]. However, the role of P2Y2 on Cldn, TRPM6, and TRPM7 expressions and functions required further studies.

Duodenal mucosal bicarbonate secretion (DMBS) is the critical process of duodenal defense against intermittent duodenal epithelial exposure to a luminal acidic environment (pH < 2). Luminal H+ is the potent activator of DMBS by stimulating a duodenal associated acid sensor, e.g., ASIC1a [5]. In addition, luminal uridine triphosphate, a P2Y2 agonist, also stimulates DMBS [4]. Previously, apical HCO3− secretion had been observed in Caco-2 monolayers [12, 29]. Laohapitakworn et al. [29] reported that PTH rapidly stimulated CFTR-, CA-, and NBCe1-mediated apical HCO3− secretion in Caco-2 monolayers. Our group demonstrated that activation of ASIC1a stimulated an apical HCO3− secretion CFTR-dependent mechanism [12]. In the present study we reported the activation of P2Y2 receptor stimulating CFTR-, CA-, and NBCe1-mediated apical HCO3− secretion in Caco-2 monolayers. Our results agreed with the finding of a previous report [4] that showed P2Y2 receptor activation mediated the stimulation of intestinal HCO3− secretion through PLC, IP3 receptor, and the intracellular Ca2+ signaling pathway.

The underlying mechanism of PPI-suppressed intestinal Mg2+ uptake is still unclear. A previous mathematically simulated study suggested that only a 1% reduction of intestinal Mg2+ absorption could induce 80% Mg2+ depletion within 1 year of PPIs used [45]. Previous studies proposed that PPIs mainly affected colonic Mg2+ absorption in PPIH mice [2, 20]. Our group reported that PPIs impeded duodenal Mg2+ absorption in PPIH rats [21]. We hypothesized that a higher luminal HCO3− secretion could lead to a suppression of small intestinal Mg2+ absorption in PPIH [12, 21]. Omeprazole markedly enhanced HCO3− secretion in human duodenum and intestinal epithelium-like Caco-2 monolayers [12, 26]. Secreted HCO3− increased luminal pH and probably decreased Mg2+ solubility, since luminal soluble Mg2+ decreased from 79.61 to 8.71% when luminal pH increased from ~ 5 to 7.8 [25]. Therefore, antagonists of CFTR and NBCe1 significantly increased duodenal Mg2+ absorption in PPIH rats [21]. These findings agreed with the present study that omeprazole induced basal and peak-stimulated HCO3− secretion. Interestingly, in the HCO3−-free condition the rate of Mg2+ transport markedly increased in omeprazole-treated Caco-2 monolayers, suggesting the inhibitory role of secreted HCO3− on intestinal Mg2+ absorption. However, luminal Mg2+ solubility and precipitation in the PPI-treated animal model requires further study.

The present study showed higher P2Y2 expression in omeprazole-treated Caco-2 monolayers. Thus, P2Y2-activated HCO3− secretion was also significantly higher in omeprazole-treated monolayers. These results explained the higher degree of suppression of Mg2+ absorption in P2Y2-activated 200-ng/ml omeprazole-treated (60.71%) and 400-ng/ml omeprazole-treated monolayers (62.96%) in comparison to control monolayers (41.58%). Moreover, P2Y2 activation enhanced HCO3− secretion and suppressed Mg2+ absorption were also mediated by PLC, IP3 receptor, intracellular Ca2+ mobilization, CFTR, CA, and NBCe1. In addition to higher HCO3− secretion, another possible mechanism of omeprazole-suppressed Mg2+ absorption is phosphatidylinositol 4,5-bisphosphate (PIP2)-mediated TRPM6 function. Since TRPM6 function required an interaction with membrane-associated PIP2, hydrolysis of PIP2 through activation of the Gq-protein coupled PLC-dependent pathway fully inactivated TRPM6 channels [46]. The higher Gq-associated P2Y2 expression and function might have induced PIP2 degradation which then inactivated TRPM6 channels in omeprazole-treated Caco-2 monolayers. However, our results suggested that Mg2+ channel-dependent Mg2+ absorption could not be involved in omeprazole-suppressed Mg2+ transport across Caco-2 monolayers.

Our results in the present study agreed with the previous study [8] that omeprazole suppressed Mg2+ channel-independent, but not Mg2+ channel-dependent, Mg2+ transport across Caco-2 monolayers. There are two possible answers to explain why omeprazole had no effect on Mg2+ channel-dependent Mg2+ transport across Caco-2 monolayers. Since omeprazole has no effect on TRPM6 expression in Caco-2 cells [21], Mg2+ channel-dependent Mg2+ transport was maintained in the same fraction. Regarding our recent results, from 100% of total Mg2+ transport, the percentages of Mg2+ channel-dependent Mg2+ transport were 18.55, 19.02, and 18.58% in control, 200-ng/ml omeprazole-treated, and 400-ng/ml omeprazole-treated monolayers. On the other hand, our recent method may not be sensitive enough to detect a very small change of Mg2+ channel-dependent Mg2+ transport of omeprazole-treated Caco-2 monolayers.

In conclusion, the present study reported the role of P2Y2 function on the modulation of intestinal Mg2+ absorption. P2Y2 agonist enhanced HCO3− secretion and suppressed Mg2+ transport through the activation of PLC, IP3 receptor, intracellular Ca2+ mobilization, CFTR, CA, and NBCe1. Inhibition of HCO3− secretion could restore Mg2+ transport in P2Y2 agonist-treated monolayers. The higher P2Y2 expression was found in omeprazole-treated Caco-2 monolayers. Therefore, the higher degree of HCO3− secretion and Mg2+ transport suppression was demonstrated in P2Y2-activated omeprazole-treated Caco-2 monolayers. Our results propose an inhibitory role of P2Y2 on intestinal Mg2+ absorption.