Zinc inhibits cAMP-induced Cl⁻ secretion in intestinal epithelial cells via calcium-sensing receptor (CaSR)

Abstract

Zinc is a commonly used antidiarrheal supplement; however, its exact mechanism of action is not well understood. Calcium-sensing receptor (CaSR) is a regulator of intestinal ion transport and a therapeutic target for secretory diarrhea. CaSR is activated by various cations and here we investigated the roles of CaSR in the antidiarrheal effects of the divalent metal zinc (Zn²⁺). In human intestinal T84 cells expressing CaSR, zinc (100 μM) inhibited forskolin-induced secretory I_sc by 60% and its effect was comparable to CaSR activator cinacalcet. Zinc effect was via inhibition of apical CFTR Cl⁻ channel and basolateral K⁺ channels. In cell models, zinc was a CaSR agonist and its antisecretory effects were CaSR-dependent. Similarly, 100 μM zinc inhibited forskolin-induced secretory I_sc by 40% in wildtype mouse intestine with no antisecretory effects in intestinal epithelia-specific CaSR knockout mice (Casr^flox/flox;Vil1-cre). Zinc inhibited I_sc induced by clinically-relevant cAMP agonists (cholera toxin and vasoactive intestinal peptide) by 65% in T84 cells. Interestingly, zinc had no effect on cGMP agonists (heat-stable E. coli enterotoxin and linaclotide)-induced secretory I_sc, suggesting its antisecretory effects are specific to cAMP. The mechanisms of zinc effect in T84 cells involved intracellular Ca²⁺ release via ryanodine receptors and inhibition of cAMP synthesis. Our findings suggest that CaSR activation is a major mechanism for the antidiarrheal effects of zinc which specifically reduces cAMP levels. In addition to its use in cholera, zinc can be effective in other cAMP-mediated secretory diarrheas.

NEW & NOTEWORTHY

Zinc is a commonly used antidiarrheal supplement; however, its exact mechanisms of antisecretory effects remain unknown. In this study, we demonstrated that zinc is an agonist of the extracellular calcium-sensing receptor (CaSR) and its antidiarrheal effects are via reduced cAMP synthesis in intestinal epithelial cells. In addition to elucidating the mechanism of action of this antidiarrheal supplement, our data support the use of zinc as a simple and effective treatment option for all cAMP-mediated diarrheas.

Graphical Abstract:

INTRODUCTION

Diarrhea is a major cause of global morbidity and mortality, particularly in young children (1). Secretory diarrhea is a common type with diverse etiologies including infections, inflammation, drugs, tumors, and genetic disorders (2). Cyclic nucleotide (cAMP and cGMP) elevation in intestinal epithelial cells is the key driver of intestinal fluid loss in certain secretory diarrheas. cAMP elevation is the major pathology in cholera and vasoactive intestinal peptide-secreting tumors (VIPoma), whereas cGMP elevation is the main driver of fluid loss in traveler’s diarrhea and GUCY2C mutations (2–4). In all these diseases, elevated cyclic nucleotides result in activation of luminal membrane Cl⁻ channel CFTR (cystic fibrosis transmembrane conductance regulator) and basolateral membrane K⁺ channels which result in Cl⁻ and fluid secretion (2). In addition, inhibition of apical membrane Na⁺/H⁺ exchanger 3 (NHE3) contributes to intestinal fluid losses by impairing fluid absorption (5). Current treatment for secretory diarrhea is primarily supportive, and fluid replacement with oral rehydration solution (ORS) is the mainstay of cholera treatment (6). In addition to ORS, World Health Organization recommends oral zinc treatment in children with diarrhea (7). Despite its extensive clinical use in diarrhea, the mechanism of antidiarrheal effects of zinc are not well understood. Zinc was shown to improve intestinal epithelial barrier and immune system function, which are thought to in part contribute in its antidiarrheal effects (8). Some studies also suggested that zinc might have direct antisecretory effects in the intestine (9), however the exact mechanisms of this effect are unclear. Although some studies attributed the effects of zinc on GPR39 (also known as “zinc-sensing receptor”) which might promote Cl⁻ absorption in enterocytes (10), the mechanisms of effects of zinc on hypersecretion (the key pathology in secretory diarrheas) are largely unknown (11).

Extracellular Ca²⁺-sensing receptor (CaSR) is a G protein-coupled receptor (GPCR) expressed in many tissues including parathyroid, kidney, brain, bone and gut (12). CaSR activation by approved drug cinacalcet inhibits cAMP and cGMP agonists-induced Cl⁻ secretion in human intestinal cells and mouse intestine by promoting cyclic nucleotide hydrolysis via phosphodiesterases (PDE) (13, 14). In addition to small molecules, CaSR is physiologically activated by di- and trivalent cations (15, 16). Since zinc (Zn²⁺) is a divalent metal, we postulated that CaSR activation might be responsible from its antisecretory effects in intestinal epithelial cells.

MATERIALS AND METHODS

Chemicals

All chemicals were purchased from Sigma-Aldrich (St. Louis, MO, USA), except CFTRinh-172 (MedChemExpress).

Cell culture

T84 (CCL-248) and HEK-293 (CRL-1573) were obtained from American Type Culture Collection (Gaithersburg, MD). HEK-293 cells overexpressing CaSR (HEK-CaSR) were purchased from AddexBio (San Diego, CA). Fischer rat thyroid cells stably expressing human wild-type CFTR (FRT-CFTR) were obtained from UCSF Cystic Fibrosis Drug Discovery Core Center. T84 and FRT-CFTR cells were cultured as described (13). HEK-293 and HEK-CaSR cells were cultured in DMEM medium containing a high concentration of glucose, 10%fetal bovine serum, 100 U/ml penicillin, and 100 ug/ml streptomycin. HEK CaSR cells were continuously selected by adding G418 sulfate.

Short-circuit current (I_sc) measurements

Cells were grown on snapwell inserts (12 mm diameter, 0.4 μm polyester membrane; Corning Life Sciences, Tewksbury, MA) at 37°C in 5% CO2 /95% air and used for short-circuit current experiments 5–7 days after plating. Inserts were mounted in Ussing chambers with each hemichamber containing bicarbonate-buffered Ringer’s solution (pH 7.4, in mM: 120 NaCl, 5 KCl, 1 MgCl₂, 1 CaCl₂, 10 D-glucose, 5 HEPES, and 25 NaHCO₃). Unless otherwise specified, all compounds were added to both apical and basolateral bathing solutions. The solutions were aerated with 95% O₂ /5% CO₂ and maintained at 37°C during experiments. Isc was measured using an EVC4000 multichannel voltage clamp (World Precision Instruments, Sarasota, FL, USA) via Ag/AgCl electrodes and 3 M KCl agar bridges, as described (14, 17). In parallel experiments, T84 cells were grown on permeable filters as described above, and bathed with Ringer’s solution (±100 μM ZnCl₂) for 60 min. To test the effects of zinc on epithelial barrier function, transepithelial electrical resistance (TEER) was measured using a Millicell-ERS Resistance System with dual electrode volt-ohmmeter (Millipore, Bedford, MA). Net TEER (ohms × cm²) was calculated by subtracting the resistance of cell-free media from measured resistance as described (14).

To measure apical Cl⁻ conductance, the basolateral membrane was permeabilized with 500 μg/ml amphotericin B for 30 min and 60 mM basolateral-to-apical Cl⁻ gradient was applied (13, 16). For these experiments, Ringer’s was the basolateral bathing solution (120 mM NaCl) and the apical solution contained 60 mM NaCl and 60 mM sodium gluconate. To measure basolateral membrane K⁺ conductance, the apical membrane was permeabilized with 20 μM amphotericin B for 30 min and apical-to-basolateral potassium gradient was applied (13, 14). The apical solution (pH 7.4) contained in mM: 142.5 K-gluconate, 1 CaCl₂, 1 MgCl₂, 0.43 KH₂PO₄, 0.35 Na₂HPO₄, 10 HEPES, and 10 D-glucose. In the basolateral solution (pH 7.4) 142.5 mM K-gluconate was replaced by 5.5 mM K-gluconate and 137 mM N-methylglucamine.

Animals

Wild-type, Casr^flox/flox (030647) and vil1-cre (004586) mice (all in C57BL/6 background) were obtained from Jackson Laboratories (Bar Harbor, ME). Intestinal epithelia-specific Casr knockout mice (Casr^flox/flox;Vil1-cre) were generated by crossbreeding and genotype was confirmed by PCR. Animals were bred in UCSF Laboratory Animal Resource Center and experiments were done in adherence with NIH Guide for the Care and Use of Laboratory Animals. All animal experiments were approved by IACUC. Both male and female mice (10–16 weeks old) were used in all experiments.

Intestinal I_sc measurements

Jejunum of mice was excised under isoflurane anesthesia and soaked in isoosmolar solution (300 mM mannitol in water) with 10 μM indomethacin to inhibit prostaglandin production. Seromuscular layer was stripped under a dissection microscope. As done before (14, 16), mucosal segments were mounted on Ussing chambers containing Ringer’s solution (120 mM NaCl and 10 mM glucose) on the basolateral side. In apical chamber NaCl was replaced with 60 mM NaCl and 60 mM sodium gluconate, and glucose was replaced with 10 mM mannitol. I_sc was measured as described above.

CaSR activity measurements

For intracellular Ca²⁺ measurements, T84 cells were plated in 96-well, black-walled microplates (Corning), HEK-293 and HEK-CaSR cells were plated in Poly-D-Lysine-treated plates (Greiner Bio-One). Confluent cells were loaded with calcium indicator Fluo-4 NW (Invitrogen) per manufacturer’s instructions and Fluo-4 fluorescence was measured in each well continuously with a plate reader (Tecan Infinite M1000) at excitation/emission wavelengths of 495/516 nm after addition of 100 or 300 μM ZnCl₂. The experiments were done using Hank’s buffered salt solution (HBSS) as the assay buffer. In some experiments, Ca²⁺ and Mg²⁺-free HBSS was used to avoid CaSR activation by these physiological agonists. In some experiments, cells were pretreated with the PLC inhibitor U73122 (10 μM), sarcoplasmic/endoplasmic reticulum Ca²⁺-ATPase (SERCA) inhibitor thapsigargin (1 μM) or ryanodine receptor (RyR) inhibitor dantrolene (10 μM) for 10 minutes prior to addition of ZnCl₂. For quantification of IP1 (stable downstream metabolite of IP3), T84 cells were grown in 384-well opaque plates (PerkinElmer) and treated with 0–100 μM ZnCl₂ for 30 minutes. Next, cells were lysed and the IP1 concentration in each well was quantified using the IP-One Gq kit (Cisbio) according to the manufacturer’s instructions.

cAMP and cGMP measurements

T84 cells were grown in clear 24-well plates (Corning) and pretreated for 20 minutes with 100 μM ZnCl₂ or vehicle control. For cAMP measurement, cells were treated with 10 μM forskolin (for 5 minutes) and lysed by repeated freeze/thaw and centrifuged to remove cell debris. In some experiments, cells were treated with PDE inhibitor IBMX (500 μM) together with forskolin. The supernatant was assayed for cAMP using the cAMP Parameter immunoassay kit according to the manufacturer’s instructions (R&D Systems, Bio-Techne). For cGMP measurement, cells were treated with 0.1 μg/ml STa toxin (for 5 minutes) and lysed by repeated freeze/thaw and centrifuged to remove cell debris. The supernatant was assayed for cGMP using the cGMP immunoassay kit according to the manufacturer’s instructions (Cell Signaling Technology, Danvers, MA).

Statistical Analysis

Experiments with two groups were analyzed with two-tailed Student’s t-test. For experiments with three or more groups, one-way analysis of variance and post hoc Newman-Keuls multiple comparisons test was used. P value < 0.05 was considered statistically significant.

Study approval

All animal experiments were approved by IACUC.

RESULTS

Zinc inhibits CFTR-mediated Cl⁻ secretion in T84 cells

Human intestinal T84 cells are a commonly used model to study intestinal ion transport which natively express CaSR and respond to CaSR agonists (13, 16, 18). In T84 cells, zinc concentration-dependently inhibited cAMP agonist forskolin-induced secretory short-circuit current (I_sc) by up to 60%, which was comparable to the effect of small molecule CaSR activator cinacalcet (Fig. 1A and B). Since CFTR is the major cAMP-gated ion Cl⁻ channel in intestinal epithelia, we tested the roles of CFTR in zinc effect using its selective inhibitor CFTR_inh-172. Forskolin-induced secretory I_sc in T84 cells was partially reversed by CFTR_inh-172. Zinc treatment concentration-dependently reduced CFTR activity in T84 cells as suggested by reduced CFTR_inh-172 responses, and its effect was comparable to cinacalcet (Fig. 1A and C). To rule out any potential effects on barrier permeability, TEER was measured with and without 100 μM zinc treatment. Under the same conditions I_sc studies are performed, zinc had no effect on TEER (Fig. 1D). These results collectively suggest that zinc reduces CFTR-mediated Cl⁻ secretion in human intestinal epithelial cells similar to CaSR agonists.

Figure 1. Zinc inhibits cAMP-induced Cl⁻ secretion and CFTR activity in T84 cells.

A. Short-circuit current (I_sc) traces showing forskolin (10 μM) and CFTR_inh-172 (10 μM) responses in the presence of 20 min ZnCl₂ pretreatment at various concentrations. Cinacalcet (30 μM) was used as a positive control for CaSR activation. B. Summary data for forskolin response for studies in A. C. Summary data for CFTR_inh-172 response for studies in A. D. Transepithelial electrical resistance (TEER) measurements in T84 cells after 60 min bathing in Ringer solution with and without 100 μM ZnCl₂. Mean ± S.E.M., n= 5–11 experiments per group. One-way analysis of variance and post hoc Newman-Keuls multiple comparisons test, *p<0.05, **p<0.01, ***p<0.001, ns: not significant.

Zinc inhibits apical CFTR Cl⁻ channel and basolateral K⁺ channels in T84 cells

Although CFTR is the major Cl⁻ channel in intestinal epithelia, forskolin-induced secretory I_sc in T84 cells is due to combined actions of apical membrane Cl⁻ and basolateral membrane K⁺ channels (19). To selectively study the effects of zinc on apical vs. basolateral ion transport, we used selective membrane permeabilization and ion gradients. In T84 cells with basolateral membrane permeabilization and basolateral-to-apical Cl⁻ gradient, zinc inhibited forskolin-induced apical Cl⁻ secretion and CFTR activity by 65% (Fig. 2A and B). To directly study basolateral membrane K⁺ conductance we used apical membrane permeabilization and apical-to-basolateral K⁺ gradient. In this setting, zinc inhibited basolateral K⁺ conductance as suggested by 65% lower responses to forskolin and cAMP-activated K⁺ channel inhibitor BaCl₂ (Fig. 2C and D). These studies suggest that zinc inhibits both apical Cl⁻ and basolateral K⁺ conductance in human intestinal epithelial cells, similar to CaSR agonists (13, 16).

Figure 2. Zinc inhibits apical membrane Cl⁻ conductance and basolateral membrane K⁺ conductance in T84 cells.

A. I_sc traces in T84 cells showing responses to forskolin (10 μM) and CFTR_inh-172 (10 μM, apical) with basolateral membrane permeabilization (250 μg/ml amphotericin B) and 60 mM basolateral-to-apical Cl⁻ gradient. Experiments were done with and without 100 μM ZnCl₂ pre-treatment for 20 min. B. Summary data for forskolin and CFTR_inh-172 responses for studies in A. C. I_sc traces showing responses to forskolin (10 μM) and BaCl₂ (5 mM, basolateral) with apical membrane permeabilization (20 μM amphotericin B) and apical-to-basolateral K⁺ gradient. Experiments were done with and without 100 μM ZnCl₂ pre-treatment for 20 min. D. Summary data for forskolin and BaCl₂ responses for studies in C. Mean ± S.E.M., n=5–6 experiments per group, Student’s t-test, **p<0.01, ***p<0.001.

Antisecretory effect of zinc is dependent on CaSR in cells and mouse intestine

To study the roles of CaSR in zinc effect, we initially used FRT-CFTR cells that have robust CFTR currents, but do not express CaSR or respond to CaSR agonists (13, 16, 20). In this setting, zinc had no effect on forskolin or CFTR_inh-172 responses suggesting that zinc does not have a direct CFTR inhibitory effect (Fig. 3A and B). GPR39 is proposed as the “zinc-sensing receptor” responsible from some of its biological effects (21). To test whether GPR39 activation has any roles in antisecretory effects of zinc, we used its selective agonist GPR39-C3 in T84 cells, which natively express GPR39 (22). Unlike zinc and CaSR agonists, GPR39-C3 pretreatment had no effects on forskolin or CFTR_inh-172-induced I_sc changes in T84 cells (Fig. 3C and D), which suggests that antisecretory effects of zinc is independent of GPR39. To test the effects of zinc on CaSR directly, we used HEK cells transfected with CaSR (HEK-CaSR). Zinc treatment resulted in intracellular Ca²⁺ elevation in HEK-CaSR (Fig. 4A), which is the known signaling pathway for CaSR activation in these cells (23, 24). Zinc treatment had no effect on intracellular Ca²⁺ in non-transfected HEK-293 cells which do not express CaSR (23) even at a high concentration of 300 μM (Fig. 4A). These experiments were done using regular assay buffer containing Ca²⁺ and Mg²⁺, which are known physiological CaSR agonists. To directly test the effects of zinc on CaSR activity in the absence of any other agonists, we performed similar experiments in Ca²⁺ and Mg²⁺-free assay buffer. In this setting, zinc treatment concentration-dependently elevated intracellular Ca²⁺ and activated CaSR with an EC₅₀ of ~100 μM (Fig. 4B). To test whether zinc has any allosteric activating effects of on CaSR, we tested Ca²⁺-induced CaSR activation in HEK-CaSR cells in the presence and absence of zinc. Extracellular Ca²⁺ activated CaSR with similar potency in the presence and absence of extracellular zinc (Fig. 4C). These results collectively suggest that zinc is a direct CaSR agonist and its antisecretory effects are via CaSR activation in intestinal epithelial cells. To further determine the role of CaSR in zinc effect, we did I_sc studies in mouse intestine. Pretreatment of jejunal mucosa with 100 μM zinc decreased forksolin-induced I_sc change by 40% in wild-type mice (Fig. 5A). Zinc had no antisecretory effects in intestinal epithelia-specific CaSR knockout mice (Casr^flox/flox;Vil1-cre, Fig. 5B) which further suggests that its antisecretory effects are via CaSR activation in the intestine.

Figure 3. Antisecretory effect of zinc is CaSR-dependent.

A. I_sc traces showing forskolin (10 μM) and CFTR_inh-172 (10 μM) responses in FRT-CFTR cells with basolateral membrane permeabilization (500 μg/ml amphotericin B) and 60 mM basolateral-to-apical Cl⁻ gradient. Experiments were done with and without 100 μM ZnCl₂ pre-treatment for 20 min. B. Summary data for forskolin and CFTR_inh-172 responses for studies in A. C. I_sc traces showing forskolin (10 μM) and CFTR_inh-172 (10 μM) responses with and without GPR39-C3 pretreatment (10 μM for 20 min) in T84 cells. D. Summary data for forskolin and CFTR_inh-172 responses for studies in C. Mean ± S.E.M., n=5–6 experiments per group, Student’s t-test, ns: not significant.

Figure 4. Zinc is a CaSR agonist.

A. (left) Representative intracellular Ca⁺² traces measured by Fluo-4 fluorescence and in HEK-CaSR and HEK-293 cells with ZnCl₂ (100–300 μM) treatment. Experiments were done in HBSS which contains 1.2 mM Ca²⁺ and 0.9 mM Mg²⁺. Summary data for Fluo-4 fluorescence change is presented on the right. B. Zinc concentration-response for intracellular Ca⁺² elevation in HEK-CaSR cells. Experiments were done in Ca²⁺ and Mg²⁺-free HBSS. C. Extracellular Ca²⁺ concentration-response curves for CaSR activation in the absence and presence of 100 μΜ ZnCl₂ in the assay buffer (Ca²⁺ and Mg²⁺-free HBSS). Mean ± S.E.M., n=4–8 experiments per group, Student’s t-test, **p<0.01, ***p<0.001.

Figure 5. Antisecretory effect of zinc in mouse intestine is CaSR-dependent.

A. I_sc traces and summary data showing responses to 10 μM forskolin in jejunal mucosa of wildtype mice with and without 100 μM ZnCl₂ pre-treatment for 20 min. B. I_sc traces and summary data showing responses to 10 μM forskolin in jejunal mucosa of intestinal epithelia-specific CaSR knock out mice (Vil1-Cre; Casr-flox) with and without 100 μM ZnCl₂ pre-treatment for 20 min. Mean ± S.E.M., n=7–9 experiments per group, Student’s t-test, *p<0.05, ns: not significant.

Zinc inhibits Cl⁻ secretion induced by cAMP agonists, but not cGMP agonists in T84 cells

Next, we tested the effects of zinc on Cl⁻ secretion induced by clinically relevant secretagogues. Similar to the experiments with forskolin, zinc pretreatment inhibited secretory I_sc induced by other cAMP agonists cholera toxin (Fig. 6A and B) and VIP (Fig. 6C and D) by 65%. 100 μM zinc treatment similarly inhibited CFTR_inh-172 responses in both settings (Fig. 6). In certain secretory diarrheas such as Traveler’s diarrhea, cytoplasmic cGMP elevation is the key driver of intestinal Cl⁻ secretion. To test the antisecretory efficacy of zinc in cGMP-mediated diarrheas, we used heat-stable E. coli enterotoxin (STa) as the cGMP agonist. Interestingly, zinc treatment had no effects on I_sc changes induced by STa-toxin or CFTR_inh-172 in this setting (Fig. 7A and B). Similarly, zinc treatment had no effect on secretory I_sc and CFTR_inh-172 response when guanylate cyclase-C receptor agonist linaclotide was used as the cGMP agonist (Fig. 7C and D). These results suggest that the antisecretory effects of zinc are specific for cAMP agonists.

Figure 6. Zinc inhibits Cl⁻ secretion induced by cAMP agonists.

A. I_sc traces showing responses to 1 μg/ml cholera toxin and 10 μM CFTR_inh-172 with and without 100 μM ZnCl₂ pre-treatment for 20 min. B. Summary data for cholera toxin and CFTR_inh-172 responses for studies in A. C. I_sc traces showing responses to 10 nM vasoactive intestinal peptide (VIP) and 10 μM CFTRinh-172 with and without 100 μM ZnCl₂ pre-treatment for 20 min. D. Summary data for VIP and CFTR_inh-172 responses for studies in C. Mean ± S.E.M., n=5–6 experiments per group, Student’s t-test, *p<0.05, **p<0.01, ***p<0.001.

Figure 7. Zinc has no effect on Cl⁻ secretion induced by cGMP agonists.

A. I_sc traces showing responses to 0.1 μg/ml heat-stable E. coli enterotoxin (STa toxin) and 10 μM CFTR_inh-172 with and without 100 μM ZnCl₂ pre-treatment for 20 min. B. Summary data for STa toxin and CFTR_inh-172 responses for studies in A. C. I_sc traces showing responses to 1 μM linaclotide and 10 μM CFTR_inh-172 with and without 100 μM ZnCl₂ pre-treatment for 20 min. D. Summary data for linaclotide and CFTR_inh-172 responses for studies in C. Mean ± S.E.M., n=6 experiments per group, Student’s t-test, ns: not significant.

Zinc inhibits cAMP synthesis and stimulates intracellular Ca²⁺ release via RyR

We previously showed that CaSR activation by other agonists (cinacalcet and Mg²⁺) stimulates hydrolysis of cyclic nucleotides via PDE in intestinal epithelial cells. Consistent with this mechanism, cinacalcet and Mg²⁺ were effective in both cAMP and cGMP-mediated diarrhea models (13, 16). To test the roles of this signaling mechanism in zinc effect, we measured cAMP levels. In T84 cells, forskolin treatment resulted in marked elevation in cellular cAMP, which was largely reduced by zinc pretreatment (Fig. 8A) consistent with its antisecretory effects in the presence of cAMP agonists. Interestingly, co-treatment with PDE inhibitor IBMX did not reverse the effect of zinc on cAMP, suggesting that zinc does not stimulate PDE activity and it likely inhibits cAMP synthesis. Consistent with lack of its antisecretory effects, zinc pretreatment had no effects on STa toxin-induced cGMP elevation in T84 cells (Fig. 8B). These results suggest that zinc activates a distinct signaling pathway to specifically reduce cAMP synthesis in intestinal epithelial cells.

Figure 8. Zinc inhibits cAMP synthesis and stimulates intracellular Ca²⁺ release by RyR.

A. cAMP concentration in T84 cell lysates with 10 μM forskolin (±500 μM IBMX, phosphodiesterase inhibitor) treatment in the presence of 20 min pretreatment with and without 100 μM ZnCl₂. B. cGMP concentration in T84 cell lysates with 0.1 μg/ml heat-stable E. coli enterotoxin (STa toxin) treatment in the presence of 20 min pretreatment with and without 100 μM ZnCl₂. C. Changes in intracellular Ca²⁺ measured by Fluo-4 NW fluorescence in T84 cells with 300 μM ZnCl₂ or vehicle treatment. In some experiments, T84 cells were pretreated with PLC inhibitor U73122 (10 μM), sarco/endoplasmic reticulum Ca²⁺-ATPase (SERCA) inhibitor thapsigargin (1 μM) or ryanodine receptor inhibitor dantrolene (10 μM) for 10 min before Zn²⁺ addition. D. IP1 (stable downstream metabolite of IP3) accumulation after 30 min of treatment with 100 μM ZnCl₂ or vehicle.

CaSR activation by other agonists (cinacalcet and Mg²⁺) result in Ca²⁺ release from intracellular stores via PLC and IP3 receptors in intestinal epithelial cells (13, 16). To study the roles of this signaling mechanism in zinc effect, we performed studies using intracellular Ca²⁺ indicator Fluo-4 NW. Similar to other CaSR agonists, zinc treatment resulted in intracellular Ca²⁺ elevation in T84 cells. The source of this Ca²⁺ response was intracellular stores as suggested by abolishment of zinc effect with sarco/endoplasmic reticulum Ca²⁺-ATPase (SERCA) inhibitor thapsigargin pretreatment (Fig. 8C). Interestingly, PLC inhibitor U73122 had minimal effect on zinc-induced intracellular Ca²⁺ release. Similarly, zinc treatment had no effects on IP1 (stable downstream metabolite of IP3) levels in T84 cells (Fig. 8D), suggesting that zinc effect is independent of PLC-IP3 pathway. Ryanodine receptor (RyR) is another major pathway for intracellular Ca²⁺ release (25) and pretreatment of T84 cells with RyR inhibitor dantrolene essentially abolished zinc-induced intracellular Ca²⁺ elevation (Fig. 8C). Since some adenylate cyclase isoforms are known to be inhibited by Ca²⁺ (26), this mechanism can potentially explain the effects of zinc on reducing cAMP synthesis. These results collectively suggest that activation of CaSR by zinc stimulates a unique signaling pathway in intestinal epithelial cells which inhibits cAMP agonists-induced Cl⁻ secretion.

DISCUSSION

Here we showed that zinc has profound antisecretory effects in human intestinal epithelial cells and mouse intestine. Several lines of evidence suggest that antisecretory effects of zinc effect are via activation of CaSR. In addition to its current use in cholera, our findings suggest that zinc can be an effective targeted treatment for other forms of cAMP-mediated secretory diarrheas.

Although ORS is the mainstay for secretory diarrhea treatment, the World Health Organization also recommends oral zinc treatment for 10–14 days in children with diarrhea (7). Randomized controlled trials showed that zinc can reduce stool weight and diarrhea duration in children with acute diarrhea and cholera (27, 28). Based on these, a zinc-fortified ORS was developed which appears to have greater efficacy than traditional ORS in pediatric diarrhea patients which further supports utility of zinc in diarrhea (29). The efficacy of zinc in diarrhea have traditionally been attributed to its effects on epithelial barrier and immune system, since zinc deficiency can impair these processes (8). However, some studies suggested the possibility of direct antisecretory effects of zinc. Zinc was shown to inhibit basolateral cAMP-activated K⁺ channel activity in rat intestine (9) and human colonic crypts (30). However, the exact mechanisms of this observed phenomenon remained unclear. Here we showed that zinc inhibits all cAMP-gated ion channels in intestinal epithelia including apical CFTR Cl⁻ channel and basolateral K⁺ channels. Importantly, we found that reduced cAMP levels via CaSR activation is the underlying mechanisms of the antisecretory effects of zinc.

Here we found that zinc has profound antisecretory effects at 100 μM. To the best of our knowledge, there are no studies reporting stool or intestinal fluid zinc concentrations in diarrhea patients on zinc treatment, thus it is difficult to directly ascertain translational relevance of our findings. An earlier clinical study in pediatric cholera patients showed an average of 1.6–1.8 kg per day watery stool output (27). Assuming the WHO recommended daily zinc dose of 20 mg (~300 μmoles) (31) and 50% intestinal absorption (32), the zinc concentration in stool of cholera patients is predicted to be 85–95 μM. Thus, it is plausible that the zinc doses currently used in clinic may yield sufficient zinc levels in intestinal fluid for CaSR activation and antisecretory effects. However, we acknowledge that these calculations are indirect, and clinical studies quantifying stool zinc concentration are needed to test validity of these estimates. A previous study showed that during acute diarrhea in infants, fecal zinc loss is increased by 2.5 folds and plasma zinc concentration is reduced by 50% compared to healthy controls (33). It is plausible that this acute zinc loss might be exacerbating diarrhea via reduced CaSR activity in the intestinal epithelia. However, future clinical studies quantifying zinc concentrations in intestinal fluid and plasma of diarrhea patients on zinc treatment are needed to better understand clinical relevance of our findings.

Our findings suggest that antisecretory effect of zinc is dependent on CaSR in human intestinal epithelial cells and mouse intestine. Some earlier studies proposed GPR39 as the “zinc-sensing receptor” responsible from its biological actions (21, 34). Here we showed that, unlike zinc, selective GPR39 agonist has no antisecretory effects in human intestinal T84 cells. Previous studies in transfected cell models also showed that GPR39 activation results in elevated cAMP (35), contrary to zinc effect found here which is reduced cAMP. These collectively suggest that the antisecretory effects of zinc is independent of GPR39 in the intestinal epithelial cells. Although zinc was listed as a potential CaSR ligand (36), there has been limited research in this subject and an earlier study claimed that zinc does not activate CaSR (21). However, several lines of evidence presented herein including experiments in transfected cell models and knockout mice suggest that zinc is a CaSR agonist with antidiarrheal effects similar to other CaSR activators.

Another interesting finding of this study is that zinc inhibits Cl⁻ secretion induced by cAMP agonists with no effect on cGMP agonists-induced Cl⁻ secretion. These findings are in agreement with a previous study which reported that zinc inhibits cholera toxin-induced secretion but not STa toxin-induced secretion in human intestinal Caco-2 cells (37). We recently showed that CaSR agonists cinacalcet and Mg²⁺ inhibit Cl⁻ secretion induced by both cAMP and cGMP agonists (13, 14, 16). Cinacalcet and Mg²⁺ increases intracellular Ca²⁺ via Gq-PLC-IP3 pathway, which leads to activation of PDE and hydrolysis of cyclic nucleotides, which explains their efficacy in both cAMP and cGMP-mediated diarrhea models. Here we found that zinc also elevates intracellular Ca²⁺ via CaSR activation, however its effects are through RyR and not PLC. Consistent with its selective efficacy in cAMP agonist-induced secretion, zinc reduced cAMP levels by inhibiting its synthesis without any effects on cGMP. Since some adenylate cyclase isoforms are known to be inhibited by calcium (26), this mechanism can potentially explain the effects of zinc on reducing cAMP synthesis. These results collectively suggest a potential distinct signaling pathway when CaSR is activated by zinc. Consistent with this, CaSR activation in certain settings and cell types is known to reduce cAMP synthesis (38). Future studies elucidating the mechanisms of CaSR activation by different agonists might be informative for better understanding of intestinal ion transport and its modulation by CaSR. Some earlier studies using membrane preparations and recombinant enzymes suggested that zinc can inhibit adenylate cyclase at micromolar concentrations (39, 40). However, it is unlikely that the zinc effect observed in our study is due to direct inhibition of adenylate cyclase, since free zinc levels in cytoplasm are at tightly kept at pico-nanomolar range and higher levels of cytoplasmic zinc are toxic (41).

In conclusion, our findings suggest that CaSR activation in intestinal epithelial cells is a key mechanism for antidiarrheal effects of zinc. In addition to its use in cholera, our findings suggest that zinc can be used in other forms of cAMP-mediated secretory diarrheas including VIPoma and bile acid diarrhea.

DATA AVAILABILITY

Source data for this study will be available at [doi: 10.17632/8kjkd6g2jj.1] following an embargo from the date of publication (no more than 3 months) to allow for commercialization of research findings.