HT-29 cells transfected with or without the halide sensor YFP-H148Q/I152L fluorescence protein were cultured in RPMI 1640 medium (Sigma Chemical Co, St. Louis, MO. U.S.A.) supplemented with 8% fetal bovine serum, 2 mM L-glutamine, 100 U/ml penicillin, and 100 μg/ml streptomycin in a 37°C incubator with 5% CO₂ and 95% humidity.

Fisher rat thyroid (FRT) cells stably transfected with ANO1 were cultured in Coon's modified F12 medium (Sigma Chemical Co, St. Louis, MO. U.S.A.) supplemented with 10% fetal bovine serum, 2 mM L-glutamine, 100 U/ml penicillin, and 100 μg/ml streptomycin in a 37°C incubator with 5% CO₂ and 95% humidity.

Male ICR mice (8–10 weeks) were fed with a chow diet and kept under specific pathogen-free conditions at Dalian Medical University (Permit Number: SCXK liao 2008-0002, 2013-0003).

Indomethacin and amiloride were all purchased from Sigma (Sigma Chemical Co, St. Louis, MO. U.S.A.). ATP and NaI were purchased from Sangon (Sangon Biotech Co., Ltd, Shanghai). Carbachol (CCh) was obtained from EDM chemicals, Inc. (San Diego, CA). E_act and CaCC_inh-A01 were purchased from Chembest (Chembest research laboratories Co., Ltd, Shanghai). Amphotericin B was bought from Solarbio (Solarbio Science and Technology Co., Ltd.). Quercetin was obtained from Shanghai Tauto Biotech Co., LTD (Shanghai, China). CFTR_inh-172 was synthesized as described by Garcia et al. (2009). T16A_inh-A01 was provided by Dr. A.S.Verkman (University of California, San Francisco, USA).

Fluorescence assays were performed using a FLUOstar Optima fluorescence plate reader (BMG Labtechnologies, Durham, NC) equipped with syringe pumps and fixed excitation/emission (500 ± 10 nm/535 ± 15 nm) filters (Chroma, Brattleboro, Vernont, USA). YFP-expressing HT-29 cells were plated in 96-well black-walled clear bottom microplates (Costar, Corning, NY, USA) and incubated for 48 h until confluent. The cells were washed three times with PBS (200 μl/wash) and 50 μl of the PBS was retained after the final wash. Quercetin was then added to the cells at different concentrations, and after 10 min of incubation the absorbance of the plate was at 535 nm. For chloride channels-mediated iodide influx assay, the fluorescence of the plate was recorded continuously (200 ms/point) for 2 s (baseline), and 28 s after adding 120 μl PBS, in which the chloride in it was replaced with iodide. Cells treated with PBS only were set as negative control, whereas those treated with 100 μM ATP plus 100 μM CCh were set as positive control.

Snapwell inserts containing HT-29 or ANO1-expressing FRT cells were mounted in an Ussing chamber system (Physiologic Instruments, San Diego, CA). For HT-29 cells, both hemi-chambers contained a standard bathing solution (in mM: 119 NaCl, 1.2 CaCl₂, 0.6 KH₂PO₄, 2.4 K₂HPO₄, 1.2 MgCl₂, 21 NaHCO₃, 10 D-glucose). For FRT cells, the basolateral chamber was filled with 5 ml of HCO₃⁻ buffered solution (in mM: 120 NaCl, 5 KCl, 1 MgCl₂, 1 CaCl₂, 5 HEPES, 25 NaHCO₃, 10 D-glucose) and half-Cl⁻ solution was used by replacing NaCl with Na-gluconate in the apical side. The basolateral membrane was permeabilized with 250 μg/ml amphotericin B before the addition of E_act and quercetin. All the cells were bathed for 15 min (to ensure a stable baseline before adding the agonist) at 37°C and aerated with 95% O₂/5% CO₂ to maintain a pH of 7.4. Short-circuit current was recorded continuously using a VCC MC6 multi-channel voltage-current clamp (World Precision Instruments, Sarasota, FL) with Ag/AgCl electrodes and 3M KCl agar bridges.

ICR mice were anesthetized with sodium pentobarbital and then sacrificed by cervical dislocation. The ileum and colon were removed and washed with ice-cold Krebs bicarbonate solution (in mM: 120 NaCl, 5 KCl, 2.5 CaCl₂, 1.2 KH₂PO₄, 1.2 MgCl₂, 25 NaHCO₃, 10 D-glucose). The ileum and colon were opened along the mesenteric border and mounted in Ussing chambers after stripping off the muscularis. Chambers were filled with Krebs bicarbonate solution containing 10 μM indomethacin (to prevent prostaglandin generation) and bubbled with 95% O₂/5% CO₂ at 37°C. Amiloride (10 μM) was added to the mucosal side to inhibit the activity of epithelial Na⁺ channel (ENaC).

ICR mice were starved for 24 h and killed immediately. The small intestines of the animals were then removed, and three closed mid-ileum loops (length 15–20 mm) were generated with sutures. The closed loops were soaked in chambers containing Krebs bicarbonate solution containing quercetin (200 μM) with or without CaCC_inh-A01 (100 μM). The solution was bubbled with 95% O₂/5% CO₂ at 37°C during the whole process. After 4 h, the ileum loops were removed, and the loop length and weight were measured to quantify the net fluid secretion. Fluid secretion was measured separately as loop weight/length ratio.

ICR mice were starved for 24 h and then intraperitoneally administered 100 μl of saline, 200 μM quercetin, 50 μM E_act or 200 μM quercetin plus 50 μM E_act. Fifteen minutes later, the animals were orally administered 10% activated charcoal (diluted in 5% gum arabic) over a 30-min period. The animals were then sacrificed and the small intestines were removed. The peristaltic index was calculated as the percentage of distance that the activated charcoal had traveled relative to the total length of small intestine.

All data were shown by representative traces or expressed as means ± SEs from four to six independent experiments. One-way or two-way ANOVA followed by Dunnett's multiple comparison test was used to compare test and control values, and differences between test and control values were considered statistically significant at the P < 0.01 and P < 0.05 levels.

All animals in this study were handled in accordance with the recommendations of “Guide for the Care and Use of Laboratory Animals of the National Institutes of Health,” and experimental protocol was approved by the Liaoning Normal University Committee on Animal Research. All surgical procedures were performed under sodium pentobarbital anesthesia to minimize suffering.

Phenotype-based fluorescence quenching test was conducted with HT-29 cells expressing halide sensor fluorescence protein to evaluate the dose-response relationship, kinetics and reversible effect of quercetin on the activation of CaCC. HT-29 cells treated with quercetin exhibited an increase in I⁻ influx, and the increase was dose-dependent, yielding an EC₅₀ value of ~37 μM (Figure 1A). Maximal activation was obtained with 200 μM quercetin as seen from the maximum fluorescence quenching in Figure 1B. Carbachol and ATP elevated the intracellular calcium concentration by combining with muscarinic and purinergic receptors, resulting in the activation of chloride channels. The effect exerted by 200 μM quercetin on chloride channel was similar to that produced by a mixture of ATP and carbachol. The activation of chloride channels by 100 μM quercetin was rapid reaching a maximum after 4 min (Figure 1C). Activation of chloride channels by quercetin was reversible, since it was fully abolished 8 min after the removal of quercetin (Figure 1D). These results suggested that quercetin could activate Cl⁻ transport in HT-29 cells.

To confirm the activation of CaCC by quercetin, short-circuit current was further measured in HT-29 cells. Since quercetin also acts as a CFTR chloride channel activator (Pyle et al., 2010; Zhang et al., 2011), 20 μM CFTR_inh-172 was added to the bath solution before administration of quercetin to eliminate the influence of CFTR-mediated Cl⁻ current. The result showed that quercetin in the apical side of HT-29 monolayers activated the short-circuit currents in a dose-dependent manner. The activation effect could be abolished by the CaCC-specific inhibitor CaCC_inh-A01 (30 μM) (Figure 2A). In addition, basolateral application of quercetin also activated CaCC-mediated short-circuit current, although this was much less potent than that produced by apical side application (Figure 2B). These results suggested that quercetin can activate both CFTR and CaCC mediated Cl⁻ transport in HT-29 cells.

ANO1 is a bona fide calcium-activated chloride channel, thus Cl⁻ current measurement was performed in stably transfected FRT cells expressing ANO1 to investigate the effect of quercetin on ANO1 Cl⁻ channel activity. Surprisingly, quercetin did not activate ANO1-mediated Cl⁻ current in the stably transfected FRT cells. On the contrary, after CaCC current was induced by 10 μM of the ANO1 agonist E_act, quercetin exhibited inhibitory activity. When added to the apical side of FRT monolayers, quercetin slightly inhibited the current at low concentrations (10–20 μM) and almost completely inhibited the Cl⁻ current at 200 μM (Figure 3A). The remaining current could be inhibited by the specific inhibitor T16A_inh-A01. Basolateral administration of quercetin resulted in partial inhibition of E_act-induced short-circuit current (~57%) (Figure 3B). The inverse effects of quercetin on ANO1 in transfected FRT cells and CaCC current in HT29 cells suggested the existence of calcium-activated chloride channels in the intestinal epithelial cells that are different from ANO1.

Short-circuit current experiment was further performed to investigate whether quercetin act as a blocker of ANO1 chloride channel activity in ANO1-expressed FRT cells. Pre-treatment of both the apical and basolateral sides of FRT cells with 100 μM quercetin partially blocked the Cl⁻ current induced by E_act, yielding an inhibition rate of ~20 and 65%, respectively for apical and basolateral sides compared to the control (Figure 4).

CaCCs have been shown to exist in mouse intestinal epithelia, and therefore the efficacy of quercetin was tested ex vivo in isolated mouse intestinal mucosa by Ussing chamber short-circuit assay. The results showed that the short-circuit currents in mouse ileal and colonic epithelia on serosal side were increased by quercetin in a dose-dependent manner, reaching a peak at 200 μM in both cases (Figures 5A,B). I_sc stimulated by the same concentration of quercetin was slightly higher in mouse ileal mucosa than in colonic mucosa. As expected, the current was slightly inhibited by 20 μM T16A_inh-A01 (inhibition rate ≤ 20%), but was partially inhibited by 100 μM CaCC_inh-A01 (inhibition rate ≥ 40%) while the remaining current was abolished by 100 μM CFTR_inh-172. The inhibitory effect of CaCC_inh-A01 on CFTR chloride channel activity in mouse ileal mucosa was also measured, and CaCC_inh-A01 did not appear to inhibit forskolin-induced CFTR-mediated Cl⁻ current at 10–100 μM (Figures 5C,D). These results further confirmed that quercetin activated a CaCC current different from ANO1 in mouse intestinal epithelia.

Next, we investigated the components of quercetin-induced short-circuit current in mouse ileum. CaCC can be activated by intracellular Ca²⁺, and Ca²⁺ influx by calcium channels is a main route to enhance intracellular calcium concentration. In order to prove whether L-type calcium channels were involved in quercetin-induced CaCC activation, 10 μM of an L-type calcium channel inhibitor (nifedipine) was added to the serosal incubation solution. Compared with DMSO control (Figure 6A), the short-circuit current induced by 200 μM quercetin was reduced by 41% after pre-treatment with nifedipine (Figure 6B). The result suggested that L-type calcium channel was involved in the I_sc generated by quercetin. We also analyzed the effect of CaCC_inh-A01 and CFTR_inh-172 on quercetin-induced I_sc after inhibition of the L-type calcium channel. Quercetin-stimulated short-circuit current was further decreased by ~6.3 and 43% after the addition of nifedipine plus CaCC_inh-A01 (100 μM) and nifedipine plus CFTR_inh-172 (100 μM), respectively (Figures 6C,D). No significant difference was observed between the nifedipine group and nifedipine plus CaCC_inh-A01 group (Figures 6B,C). These results indicated that Ca²⁺ influx through L-type calcium channel was an important way to activate intestinal epithelial CaCC. As CFTR_inh-172 could further inhibit the short-circuit current by 25% (Figure 6D), it suggested the existence of CFTR-mediated Cl⁻ current. However, CaCC_inh-A01 or CFTR_inh-172 could only inhibit part of the I_sc induced by quercetin. CaCC_inh-A01 produced a slightly higher inhibition rate (~52%) (Figure 6F) than CFTR_inh-172 (~44%) (Figure 6G). These results were consistent with previous findings. In addition, pre-treatment with nifedipine, CaCC_inh-A01 and CFTR_inh-172 almost completely abolished quercetin-induced short-circuit current in mouse ileal tissues (Figure 6E), but there remained a small amount of short-circuit current after the addition of CaCC_inh-A01 and CFTR_inh-172 (Figure 6H). Figure 6I summarizes the result of I_sc from the different treatments. The results suggested that quercetin-induced short-circuit currents may involve other Ca²⁺-activated channels such as Ca²⁺-activated K⁺ channel.

Transepithelial Cl⁻ secretion can be promoted by activated chloride channel, as well as the accumulation of Cl⁻ in the epithelial cells. Therefore, it was important to determine whether transporters and ion channels in the basolateral membrane would participate in quercetin-activated short-circuit currents. Ouabain applied at 1 mM can inhibit 61% of the quercetin (200 μM)-activated short-circuit current when applied to mouse ileum on the serosal side, whereas bumetanide and clotrimazole given at 100 and 50 μM, respectively, completely abolished the short-circuit currents stimulated by quercetin (Figures 7A,B). These results suggested that Na⁺/K⁺-ATPase, NKCC, and K⁺ channels could participate in quercetin-induced Cl⁻ currents.

The effect of quercetin on intestinal fluid secretion was investigated by closed-loop experiments. Figure 8 shows the ileal closed-loops incubated with 200 μM quercetin, and 200 μM quercetin plus 100 μM CaCC_inh-A01. Krebs bicarbonate solution was used as a control in this test. Quercetin significantly increased fluid secretion in the closed loop, which was reduced by CaCC_inh-A01. The results revealed that quercetin induced intestinal fluid secretion via activation of CaCC chloride channel activity.

ANO1 is expressed in the interstitial cells of Cajal in the intestine where it modulates smooth muscle contraction (Huang et al., 2009; Hwang et al., 2009; Ferrera et al., 2010). Thus, intestinal motility measurement was performed in mice to evaluate the efficacy of quercetin in vivo. Intraperitoneal administration of 50 μM of the ANO1 activator E_act accelerated intestinal peristalsis, with peristaltic index of 72 ± 5.5% compared to saline control of 56 ± 6.9%. As shown in Figure 9A, quercetin (200 μM) significantly inhibited the basal and E_act-stimulated intestinal peristalsis. Figure 9B compares the peristaltic indexes from the different treatments. These results demonstrated that quercetin slowed intestinal peristalsis by inhibiting ANO1 activity.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.