Molecular mechanism of p-Coumaric acid in the regulation of AQP2 function through the activation of calcium-sensing receptor signaling Authors: Ines Angelini, Mariangela Centrone, Giusy Rita Caponio, Ciro Leonardo Pierri, Maria Mastrodonato, Annarita Di Mise, Giovanna Valenti, Marianna Ranieri, Grazia Tamma Published: 2026 ABSTRACT The hormone vasopressin (AVP) controls renal water reabsorption by modulating the expression and trafficking of the water channel aquaporin-2 (AQP2) through the activation of the cAMP/PKA signal transduction pathway. Previous studies revealed that Olive Leaf Extract (OLE) counteracts the vasopressin-dependent AQP2 functions by stimulating the calcium-sensing receptor (CaSR). Here, the biological activities of p-Coumaric acid, a selective polyphenol in OLE, were investigated. Stimulation of renal collecting duct MCD4 cells with p-Coumaric acid at a concentration of 1 nM caused a significant intracellular calcium release. NPS-2143, a selective CaSR antagonist, abolished this increase. Molecular docking analysis revealed that p-Coumaric acid can form binding interactions with the binding pocket of Tecalcet, a known CaSR activator, likely suggesting that p-Coumaric acid may stimulate the CaSR. Confocal analysis and immunoblotting experiments showed that p-Coumaric acid impaired the DDAVP-dependent membrane expression of AQP2 and the consequent increase of the osmotic water permeability (Pf). Additionally, Fluorescence Resonance Energy Transfer (FRET) experiments demonstrated that p-Coumaric acid prevented the DDAVP-induced cAMP generation, consequently attenuating the AQP2 phosphorylation at serine 256. Together, these findings suggest that p-Coumaric acid may antagonize the effects of vasopressin, possibly by binding to and stimulating the CaSR.Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1038/s41598-026-58679-5. Introduction The body’s water balance is tightly controlled by the antidiuretic hormone vasopressin (AVP). In renal collecting duct principal cells, AVP binds to the selective vasopressin V2 receptor (V2R), thereby activating the cAMP/PKA signal transduction pathway, which results in the phosphorylation of the water channel AQP2 at serine 256 (AQP2-pS256). This phosphorylation is considered a priming event1, resulting in the translocation of AQP2-bearing vesicles from an intracellular pool to the apical plasma membrane, where water is reabsorbed from the lumen to the serosal side via AQP3 and AQP4, which are expressed at the basolateral membrane of principal cells2. Besides serine 256, AQP2 retains other phosphorylation sites within the C-terminal domain1. Specifically, stimulation with the hormone vasopressin increased phosphorylation at serine 256, serine 264, and threonine 269, while significantly decreasing phosphorylation at serine 2611. Several studies revealed that phosphorylation at serine 269 retained AQP2 at the apical plasma membrane1,3. Recent evidence suggests that the phosphorylation of serine 261 plays a key role in modulating pS256- and pS269-mediated AQP2 apical translocation4. On the other hand, phosphorylation at serine 261 may also target AQP2 for protein degradation5,6. The molecular machinery that controls AQP2 expression, translocation, and insertion into the apical plasma membrane is quite complex and is also modulated by several intracellular and extracellular signaling molecules. Previous studies have shown that extracellular calcium disrupts forskolin-dependent AQP2 membrane localization7. Physiologically, hypercalciuria hampers vasopressin-dependent water reabsorption. Urine concentrating defects and polyuria are the main renal manifestations of hypercalcemia7. Importantly, hypercalciuria is also related to abnormal diuresis8,9. In this respect, the functional interaction between the calcium-sensing receptor (CaSR) and AQP2 in the context of a renal concentrating disorder associated with hypercalciuria has been described5,6. The CaSR is a G-protein-coupled receptor (GPCR) playing a key role in modulating calcium homeostasis. Extracellular calcium can be considered the main agonist of CaSR. However, some amino acids and small molecules, including Tecalcet (also known as NPS-R568), enhance the sensitivity of CaSR to calcium. Consequently, they are designated as allosteric activators10. Selective CaSR stimulation with NPS-R568, a specific CaSR agonist, decreased the forskolin-induced phosphorylation of AQP2 at serine 256 and hampered the osmotic water permeability11. Novel findings revealed that Goreisan (GRS), a traditional Japanese Kampo medicine, has a diuretic effect12. This is achieved by reducing AQP2 phosphorylation at serine 269, reducing AQP2 membrane expression, and reducing DDAVP-induced urine output through the activation of the CaSR12. Furthermore, in vitro and in vivo data demonstrated that Olive Leaf Extract (OLE) prevented the vasopressin-induced AQP2 trafficking through the activation of the CaSR13. However, the bioactive molecules in OLE that are responsible for these effects are still unknown. Olive leaves extract has been tested to tackle several disturbances, including metabolic syndrome, dyslipidemia, cancer, cystic fibrosis, and hypertension14–18. In patients with essential hypertension, treatment with Olea extract showed several beneficial effects. These included reductions in glycaemia, calcaemia, and blood pressure19. In MCD4 cells, the harmful effects induced by low-dose cadmium (1 µM) were counteracted by OLE20. Polyphenols are the predominant molecules in the bioactive profile of the olive leaf extracts21,22. Chemical characterization of OLE, using the ultra-high performance liquid chromatography electrospray ionization tandem mass spectrometry (UHPLC-ESI-MS/MS), identified several phytocompounds such as tyrosol-related molecules, flavonoids, and coumaric-related compounds22. The biological effects of several of these phenols are well known. Oleuropein, tyrosol, hydroxytyrosol, and luteolin have antioxidant and anti-inflammatory actions23,24 and they can interfere with numerous intracellular pathways controlling cell survival and deaths25–27. p-Coumaric acid shares several biological features with other phenols in OLE, including antioxidant and anticancer properties28. Coumaric acid is a phenolic metabolite derived from cinnamic acid. Three isomeric derivatives (o-, m-, and p-Coumaric acids) have been identified. Of these, p-Coumaric acid is the most abundant form found in nature29. Compared to other phenolic acids, p-Coumaric acid exhibited a higher bioavailability30. Moreover, p-Coumaric acid has antiproliferative and anticancer action, possibly by modulating different intracellular signal pathways controlling cell death and cell cycles30. In skeletal muscle cells, p-Coumaric acid has been suggested as a potential treatment for metabolic disorders due to its ability to regulate glucose and lipid metabolism via AMPK stimulation28. In nephrotic diabetic rats, p-Coumaric acid has beneficial effects31, reduces the oxidative stress induced by high-fat diet32, and ameliorates the oxidative state in renal reperfusion injury33. Based on these features, olive leaf phenols have been assayed to tackle several diseases characterized by abnormal cell growth, oxidative stress, and inflammation. In the context of this study, functional studies were applied to evaluate the effects of p-Coumaric acid on intracellular calcium dynamics and AQP2 trafficking. Importantly, we found that p-Coumaric acid, at a concentration of 1 nM, specifically stimulates the CaSR. This impairs the vasopressin-regulated water response by inhibiting AQP2 trafficking and function. Results Effects of p-Coumaric acid on vasopressin-induced AQP2 function in MCD4 cells To evaluate the potential effects of p-Coumaric acid on the trafficking and function of AQP2 regulated by vasopressin, MCD4 renal collecting duct cells stably expressing the selective vasopressin receptor 2 (V2R) and human aquaporin-2 (AQP2) were used as an experimental model. Confocal analysis (Fig. 1) showed that incubation with p-Coumaric acid (1 nM for 45 min) partially impaired the membrane localization of AQP2 induced by stimulation with DDAVP. Treatment with p-Coumaric acid alone did not alter the intracellular localization of AQP2 compared to the control condition. To further confirm these findings, quantitative analysis has been performed in xz figures using ImageJ software (Supplementary Fig. 1). Data revealed that compared with control cells, stimulation with DDAVP significantly increased the membrane/cytosol fluorescence ratio (CTRL: 100 ± 4.5, n = 48; vs. DDAVP: 233 ± 13.30, n = 35) as expected. Whereas, treatment with p-Coumaric acid alone did not alter the AQP2 intracellular localization compared to control cells (p-Coumaric acid: 110.9 ± 4.24, n = 35). By contrast, in the presence of p-Coumaric acid, DDAVP failed in promoting AQP2 membrane localization (DDAVP + p-Coumaric acid: 117.1 ± 5.87, n = 49). Fig. 1Effects of p-Coumaric acid on AQP2 localization in MCD4 cells. Cells were treated as described in the “Materials and methods” section, stained for AQP2, and subjected to confocal laser scanning microscopy (scale bar: 5 μm). Effects of p-Coumaric acid on AQP2 localization in MCD4 cells. Cells were treated as described in the “Materials and methods” section, stained for AQP2, and subjected to confocal laser scanning microscopy (scale bar: 5 μm). Phosphorylation events tightly regulate the membrane targeting of AQP2. Specifically, stimulation with the hormone vasopressin promotes the activation of PKA, which phosphorylates AQP2 at serine 256. To test whether treatment with p-Coumaric acid affects the vasopressin-induced AQP2 phosphorylation, immunoblotting experiments were conducted (Fig. 2). Fig. 2Effects of p-Coumaric acid on AQP2 phosphorylation at serine 256 in MCD4 cells. (A) Lysates from MCD4 cells were subjected to immunoblotting using Abs specific for AQP2-pS256 and total AQP2. (B) Densitometric analysis of AQP2-pS256 bands normalized to total AQP2 bands (total AQP2) is reported in the histogram. Data are expressed as means ± Standard Error of the Mean (S.E.M.) and analyzed by One-way ANOVA followed by Tukey’s Multiple Comparisons test of six independent experiments (* p < 0.05 vs. CTRL). Original blots and stain-free blots are shown in Supplementary Fig. 2B and C. Effects of p-Coumaric acid on AQP2 phosphorylation at serine 256 in MCD4 cells. (A) Lysates from MCD4 cells were subjected to immunoblotting using Abs specific for AQP2-pS256 and total AQP2. (B) Densitometric analysis of AQP2-pS256 bands normalized to total AQP2 bands (total AQP2) is reported in the histogram. Data are expressed as means ± Standard Error of the Mean (S.E.M.) and analyzed by One-way ANOVA followed by Tukey’s Multiple Comparisons test of six independent experiments (* p < 0.05 vs. CTRL). Original blots and stain-free blots are shown in Supplementary Fig. 2B and C. Compared to untreated cells (CTRL), incubation with p-Coumaric acid significantly reduced the AQP2 phosphorylated at serine 256 in the absence or presence of DDAVP (DDAVP: 167.8 ± 14.92; p-Coumaric acid: 101.6 ± 19.15; DDAVP + p-Coumaric acid: 97.91 ± 21.73 vs. CTRL: 100.0 ± 6.58, n = 6). Treatment with p-Coumaric acid did not alter the total abundance of AQP2 (Supplementary Fig. 2A). By contrast, any relevant changes in AQP2 phosphorylation at serine 261 were instead found on the vasopressin-dependent pathway (DDAVP: 36.30 ± 7.13; p-Coumaric acid: 110.8 ± 8.52; DDAVP + p-Coumaric acid: 36.03 ± 10.30 vs. CTRL: 100.0 ± 6.75, n = 6) (Fig. 3). Fig. 3Effects of p-Coumaric acid on AQP2 phosphorylation at serine 261 in MCD4 cells. (A) Lysates from MCD4 cells were subjected to immunoblotting using Abs specific for AQP2-pS261 and total AQP2. (B) Densitometric analysis of AQP2-pS261 bands normalized to total AQP2 bands (total AQP2) is reported in the histogram. Data are expressed as means ± Standard Error of the Mean (S.E.M.) and analyzed by Kruskal-Wallis test of six independent experiments (* p < 0.05 vs. CTRL). Original blots and stain-free blots are shown in Supplementary Fig. 2B and D. Effects of p-Coumaric acid on AQP2 phosphorylation at serine 261 in MCD4 cells. (A) Lysates from MCD4 cells were subjected to immunoblotting using Abs specific for AQP2-pS261 and total AQP2. (B) Densitometric analysis of AQP2-pS261 bands normalized to total AQP2 bands (total AQP2) is reported in the histogram. Data are expressed as means ± Standard Error of the Mean (S.E.M.) and analyzed by Kruskal-Wallis test of six independent experiments (* p < 0.05 vs. CTRL). Original blots and stain-free blots are shown in Supplementary Fig. 2B and D. Therefore, to evaluate whether treatment with p-Coumaric acid is effective in regulating AQP2 trafficking by fine-tuning intracellular cAMP, fluorescence resonance energy transfer (FRET) technology was employed (Fig. 4). Fig. 4Evaluation of cAMP levels in MCD4 cells by Fluorescence Resonance Energy Transfer (FRET) analysis. MCD4 cells were transiently transfected with the Epac H187 sensor as described in the “Materials and methods” section. Histogram reports netFRET% measured in transfected MCD4 cells treated or left under basal conditions (CTRL). The reduced netFRET signal registered after DDAVP stimulation indicates a significant increase in cAMP content. Treatment with p-Coumaric acid alone did not significantly alter the intracellular cAMP levels. By contrast, p-Coumaric acid at 1 nM in combination with DDAVP impaired the DDAVP-induced reduction of netFRET. Interestingly, stimulation with 10 µM NPS-2143, a CaSR negative allosteric modulator, reduced the inhibitory effect of p-Coumaric acid on DDAVP-induced cAMP release. All data were analyzed by One-way ANOVA followed by Tukey’s multiple comparisons test and are expressed as means ± Standard Error of the Mean (S.E.M.) (* p < 0.05 vs. CTRL; *** p < 0.001 vs. CTRL; data are obtained from three independent experiments). Evaluation of cAMP levels in MCD4 cells by Fluorescence Resonance Energy Transfer (FRET) analysis. MCD4 cells were transiently transfected with the Epac H187 sensor as described in the “Materials and methods” section. Histogram reports netFRET% measured in transfected MCD4 cells treated or left under basal conditions (CTRL). The reduced netFRET signal registered after DDAVP stimulation indicates a significant increase in cAMP content. Treatment with p-Coumaric acid alone did not significantly alter the intracellular cAMP levels. By contrast, p-Coumaric acid at 1 nM in combination with DDAVP impaired the DDAVP-induced reduction of netFRET. Interestingly, stimulation with 10 µM NPS-2143, a CaSR negative allosteric modulator, reduced the inhibitory effect of p-Coumaric acid on DDAVP-induced cAMP release. All data were analyzed by One-way ANOVA followed by Tukey’s multiple comparisons test and are expressed as means ± Standard Error of the Mean (S.E.M.) (* p < 0.05 vs. CTRL; *** p < 0.001 vs. CTRL; data are obtained from three independent experiments). Compared to untreated cells, DDAVP stimulation significantly reduced the normalized netFRET signals (DDAVP = 33.25 ± 3.6%, n = 99 cells, vs. CTRL = 100.00 ± 12.65%, n = 131 cells; p < 0.0001), consistent with a significant increase in intracellular cAMP. By contrast, incubation with p-Coumaric acid impaired the DDAVP-induced reduction of netFRET measurements, consistent with a decrease of the DDAVP-dependent cAMP generation (DDAVP + p-Coumaric acid: 101.4 ± 17.04%, n = 89 cells, vs. DDAVP: 33.25 ± 3.6%, n = 99 cells, p < 0.0001). Treatment with p-Coumaric acid alone did not significantly alter the intracellular cAMP level compared with cells left under basal conditions (p-Coumaric acid: 73.14 ± 10.36%, n = 103 cells, vs. CTRL: 100.00 ± 12.65%, n = 131 cells). Previous reports have shown that the selective activation of the CaSR-dependent signaling pathway prevents the cAMP/PKA pathway induced by the hormone vasopressin11,13,34. FRET measurements showed that NPS-2143, a CaSR negative allosteric modulator, reduced the inhibitory effect of p-Coumaric on DDAVP-induced cAMP production (DDAVP + p-Coumaric acid + NPS-2143: 55.08 ± 7.13, n = 119 cells, vs. DDAVP + p-Coumaric acid: 101.4 ± 17.04%, n = 89 cells, p < 0.05). Consistent with these observations, treatment with p-Coumaric acid prevented the DDAVP-induced increase of the temporal osmotic response (indicated as 1/τ in Fig. 5) (DDAVP + p-Coumaric acid: 1.16 ± 0.03, n = 215 cells vs. DDAVP: 1.88 ± 0.06, n = 379 cells, p < 0.001). Fig. 5Effects of p-Coumaric acid on AQP2 function in MCD4 cells. (A) Osmotic water permeability in MCD4 cells was measured as reported in the “Materials and methods” section. Cells were left under basal conditions (CTRL) or treated with DDAVP (100 nM for 30 min), p-Coumaric acid (1 nM for 45 min), or co-treated with p-Coumaric acid and DDAVP or NPS-2143 (10 µM). (B) The temporal osmotic response is indicated as 1/τ (expressed in sec− 1). Data are shown as bar plot reporting means ± Standard Error of the Mean (S.E.M.) followed by Tukey’s Multiple comparisons test (*** p < 0.001 vs. CTRL; data are obtained from seven independent experiments). Effects of p-Coumaric acid on AQP2 function in MCD4 cells. (A) Osmotic water permeability in MCD4 cells was measured as reported in the “Materials and methods” section. Cells were left under basal conditions (CTRL) or treated with DDAVP (100 nM for 30 min), p-Coumaric acid (1 nM for 45 min), or co-treated with p-Coumaric acid and DDAVP or NPS-2143 (10 µM). (B) The temporal osmotic response is indicated as 1/τ (expressed in sec− 1). Data are shown as bar plot reporting means ± Standard Error of the Mean (S.E.M.) followed by Tukey’s Multiple comparisons test (*** p < 0.001 vs. CTRL; data are obtained from seven independent experiments). Importantly, the inhibition of the CaSR with NPS-2143 hampered the inhibitory effect of p-Coumaric acid on DDAVP (DDAVP + p-Coumaric acid + NPS-2143: 1.58 ± 0.06, n = 451 cells, vs. DDAVP + p-Coumaric acid: 1.16 ± 0.03, n = 215 cells, p < 0.001). Effects of p-Coumaric acid on intracellular calcium changes Previous studies demonstrated that incubation of renal collecting duct MCD4 cells with OLE caused a significant intracellular calcium release and the activation of the CaSR13. Polyphenols in OLE, known for their antioxidant and anti-inflammatory properties, provide several health-beneficial actions. Among these, p-Coumaric acid is a phenolic molecule showing a high bioavailability and various biological actions35–39. Moreover, p-Coumaric acid may modulate intracellular signaling pathways by affecting calcium dynamics and transient calcium release38,40,41. To evaluate the possible role of p-Coumaric acid in the selective stimulation of the CaSR, intracellular calcium measurements were carried out using Fura-2 AM on MCD4 cells that endogenously express a functional CaSR42. Statistical analysis of the fluorescent Fura-2 ratio showed that p-Coumaric acid induced a significant intracellular calcium increase at 1 nM concentration (Fig. 6A) (p-Coumaric acid: 31.51 ± 3.21% vs. ATP: 100 ± 2.16%; p < 0.0001, n = 5). Interestingly, the increase in calcium induced by p-Coumaric acid was dependent on CaSR stimulation, given that pretreatment with NPS-2143 abolished the transient increase in calcium evoked by incubation with p-Coumaric acid (Fig. 6B), (NPS-2143 + p-Coumaric acid: 1.868 ± 0.48% vs. ATP: 100 ± 2.16%; p < 0.0001, n = 5). Moreover, in order to evaluate the positive allosteric modulator behaviour of the p-Coumaric acid, dose-response experiments were performed. As reported in Fig. 6D, p-Coumaric acid acts as a positive allosteric modulator of the CaSR, increasing the affinity of the receptor versus its natural agonist, extracellular Ca2+ (Fig. 6D), (EC50 p-Coumaric acid: 0.33 vs. EC50 in basal condition: 0.02, n = 5). Fig. 6Effects of p-Coumaric acid on intracellular calcium oscillations in MCD4 cells. MCD4 cells, loaded with Fura-2 AM as described in the “Materials and methods” section, were stimulated with p-Coumaric acid (1 nM) in the absence or in the presence of NPS-2143 (10 µM). (A) Fluorescence Fura-2 AM ratio 340/380 nm was recorded, and responses to p-Coumaric acid were compared to that obtained after stimulation with a maximal dose of the calcium-mediated agonist ATP (100 µM), that was used as an internal control (100%). (B) Fluorescence Fura-2 AM ratio 340/380 nm was recorded and responses to p-Coumaric acid/NPS-2143 were compared to that obtained after stimulation with a maximal dose of the calcium-mediated agonist ATP (100 µM) that was used as an internal control (100%). (C) Histogram shows the ΔFura-2 AM ratio percentage. Data are expressed as means ± Standard Error of the Means (S.E.M.) and analyzed by One-way ANOVA followed by Tukey’s Multiple Comparisons test (**** p < 0.0001 vs. ATP; $$$$ p < 0.0001 vs. p-Coumaric acid, data are obtained from three independent experiments). (D) Concentration effect curve for LOG of extracellular Ca2+ on the intracellular calcium in the presence or in the absence of p-Coumaric acid (1 nM) where CaSR response is expressed as change in Fura-2 AM ratio. Effects of p-Coumaric acid on intracellular calcium oscillations in MCD4 cells. MCD4 cells, loaded with Fura-2 AM as described in the “Materials and methods” section, were stimulated with p-Coumaric acid (1 nM) in the absence or in the presence of NPS-2143 (10 µM). (A) Fluorescence Fura-2 AM ratio 340/380 nm was recorded, and responses to p-Coumaric acid were compared to that obtained after stimulation with a maximal dose of the calcium-mediated agonist ATP (100 µM), that was used as an internal control (100%). (B) Fluorescence Fura-2 AM ratio 340/380 nm was recorded and responses to p-Coumaric acid/NPS-2143 were compared to that obtained after stimulation with a maximal dose of the calcium-mediated agonist ATP (100 µM) that was used as an internal control (100%). (C) Histogram shows the ΔFura-2 AM ratio percentage. Data are expressed as means ± Standard Error of the Means (S.E.M.) and analyzed by One-way ANOVA followed by Tukey’s Multiple Comparisons test (**** p < 0.0001 vs. ATP; $$$$ p < 0.0001 vs. p-Coumaric acid, data are obtained from three independent experiments). (D) Concentration effect curve for LOG of extracellular Ca2+ on the intracellular calcium in the presence or in the absence of p-Coumaric acid (1 nM) where CaSR response is expressed as change in Fura-2 AM ratio. Re-docking and docking analysis on the wild-type CaSR To evaluate whether p-Coumaric acid can directly bind and activate the CaSR, a re-docking with the subsequent docking analysis was performed. The re-docking simulation allowed us to define a specific gridbox for the human CaSR that was able to reproduce the crystallized protein–ligand (CaSR-Tecalcet) complex with high fidelity. The root mean square deviation (RMSD) between the coordinates of the ligand (Tecalcet) in the crystallized complex and the best “re-docked” pose coordinates, obtained by using AutoDock, was lower than 1 Å in both CaSR chains (i.e., the RMSD between the docked Tecalcet and Tecalcet pose present in the CaSR chain A, from the crystallized 9abs.pdb was equal to 0.82 Å, as reported in Fig. 7). Fig. 7A zoomed view of the membrane domain of CaSR chain A is reported in white cartoon crystallized in complex with Tecalcet (yellow sticks). The best pose of the re-docked Tecalcet (magenta sticks) is reported for comparative purposes, showing a RMSD of the atomic coordinates of 0.82 Å. A zoomed view of the membrane domain of CaSR chain A is reported in white cartoon crystallized in complex with Tecalcet (yellow sticks). The best pose of the re-docked Tecalcet (magenta sticks) is reported for comparative purposes, showing a RMSD of the atomic coordinates of 0.82 Å. The resulting predicted binding energy for Tecalcet calculated by AutoDock was equal to − 8.27 kcal/mol in CaSR chain A and − 7.98 in CaSR chain B. The same gridbox was used for docking p-Coumaric acid in the CaSR protein, providing the docked-pose observable in Fig. 8, which showed that p-Coumaric acid was able to establish binding interactions with a group of residues involved in the binding of the Tecalcet ligand observed in the crystallized structure. Fig. 8Docking of p-Coumaric acid to CaSR 3D protein structure (9abs.pdb), crystallized in complex with Tecalcet. (A) Side view of CaSR 3D protein dimer structure is reported in green (chain B) and white (chain A) cartoon representation, complexed with Tecalcet (yellow sticks, from the crystallized 9abs.pdb) and p-Coumaric acid (blue sticks, from docking analysis). (B) Zoomed view of the Tecalcet/p-Coumaric acid binding region in CaSR chain A (white cartoon on the right) and chain B (green cartoon on the left). (C) and (D) Residues participating the binding of Tecalcet (panel C) and p-Coumaric acid (panel D) are reported in sticks representation and labelled. Docking of p-Coumaric acid to CaSR 3D protein structure (9abs.pdb), crystallized in complex with Tecalcet. (A) Side view of CaSR 3D protein dimer structure is reported in green (chain B) and white (chain A) cartoon representation, complexed with Tecalcet (yellow sticks, from the crystallized 9abs.pdb) and p-Coumaric acid (blue sticks, from docking analysis). (B) Zoomed view of the Tecalcet/p-Coumaric acid binding region in CaSR chain A (white cartoon on the right) and chain B (green cartoon on the left). (C) and (D) Residues participating the binding of Tecalcet (panel C) and p-Coumaric acid (panel D) are reported in sticks representation and labelled. The resulting predicted binding energy calculated by AutoDock was equal to − 6.97 kcal/mol in CaSR chain A and 7.26 kcal/mol in CaSR chain B. Notably, the methoxyphenyl moiety of Tecalcet and the hydroxyphenyl moiety of p-Coumaric acid bind to a similar aromatic/hydrophobic pocket consisting of residues F668, F684, W818, I777. In addition, Tecalcet appears to assume slightly different orientations in the two chains and to form strong interactions with Y825, the acidic/hydrophilic residues E837, Q681, T780 and other hydrophobic residues like L773 in chain A or L776 and I822 in chain B. Conversely, p-Coumaric acid appears to form strong interactions with similarly located residues in both chains, with reference to the hydrophilic Q681, T780 and the acidic E837, but also with the basic R680 and the hydrophilic C781. Discussion In this study, p-Coumaric acid has been identified as a possible allosteric modulator of the calcium-sensing receptor (CaSR). CaSR is a G-protein coupled receptor (GPCR) that plays key functions in modulating intracellular calcium balance43,44. The CaSR is mainly expressed in the parathyroid glands, where it controls the release of the parathyroid hormone (PTH) that functions to fine-tune the plasma calcium levels45. However, CaSR signaling can be induced in several epithelia regardless of PTH. Several naturally occurring mutations have now been identified as causing diseases46. Most of these mutations occur in the extracellular domain (ECD), which is involved in receptor dimerization47. Gain-of-function mutations in the CaSR result in autosomal dominant hypocalcemia type 1. Conversely, loss-of-function mutations lead to familial hypocalcaemic hypercalcemia. Patients suffering from chronic kidney disease often showed activated CaSR48. Numerous studies revealed that CaSR can be targeted by a variety of endogenous and exogenous molecules that act as selective ligands, including L-amino acids, polyamines, small peptides, and phosphate, which can directly induce PTH secretion49. Under certain physiological states, these ligands can stimulate or interfere with CaSR-dependent signaling, allowing the receptor to function as a complex molecular hub that receives a wide variety of incoming signals that ultimately lead to a specific cellular response. Cryo-electron microscopy (cryo-EM) analysis identified several CaSR ligands such as L-tryptophan10,50. Ultra-performance liquid chromatography-mass spectrometry (UPLC-MS) identified novel potentially bioactive molecules isolated from Rhizoma drynariae (RD), acting as natural CaSR ligands carrying multiple mutations in the ECD. Among these natural molecules, coumaric acid 4-O-β-D-glucopyranoside was detected47. Whether p-Coumaric acid can also interact with and modulate the activity of the wild-type human CaSR remains unclear. Here, functional measurements showed that administration of p-Coumaric acid at very low concentrations (1 nM) induced a transient but significant intracellular calcium release in renal MCD4 cells, which endogenously expressed the CaSR. Importantly, the intracellular calcium response induced by p-Coumaric acid was abolished by pre-treatment with NPS-2143, a CaSR negative allosteric modulator, likely indicating that p-Coumaric acid might be proposed as a natural allosteric modulator for the CaSR. In this respect, docking simulation analysis was applied to wild-type CaSR. Importantly, to propose a possible binding region for p-Coumaric acid in the CaSR, justifying the observed allosteric activation, we explored the Tecalcet binding region by docking to estimate the ability to bind p-Coumaric acid. This choice was motivated by the structural similarity between the methoxyphenyl moiety of Tecalcet and the hydroxyphenyl moiety of p-Coumaric acid. Our analysis revealed that p-Coumaric acid binds with comparable affinity to both chains of the investigated CaSR, with values broadly similar to those calculated for Tecalcet. Together, these findings propose p-Coumaric acid as a possible natural allosteric modulator for the CaSR, which could be useful in further research. In renal collecting duct cells, selective activation of the CaSR dramatically reduced the phosphorylation of AQP2 at serine 256, thus limiting the renal water reabsorption11,51. Consistent with these findings, we found that short-term treatment with p-Coumaric acid significantly decreased the phosphorylation of AQP2 at serine 256 thus preventing the intracellular trafficking and fusion of AQP2-bearing vesicles to the apical plasma membrane. In contrast, p-Coumaric acid did not affect the decrease in AQP2 phosphorylation at serine 261 induced by DDAVP, which suggests that its effect only targets cAMP/PKA-dependent AQP2 phosphorylation at S256. Phosphorylation at serine 256 depends on cAMP/PKA signaling2,52. FRET studies demonstrated that p-Coumaric acid impaired the DDAVP-induced cAMP generation, thereby reducing the vasopressin-induced AQP2 expression at the apical membrane. In accordance with these findings, the compromised AQP2 translocation to the cell membrane induced by p-Coumaric acid corresponds to a substantial reduction in the osmotic water permeability that was partially restored by inhibiting the p-Coumaric acid-induced CaSR activation, using NPS-2143. This observation, indeed, highlights that the inhibitory action of p-Coumaric acid on vasopressin-dependent water permeability depends on the CaSR activation elicited by p-Coumaric acid. However, it cannot be excluded that, in addition to its effects on CaSR signaling, p-Coumaric acid may control other intracellular signaling pathways that converge to regulate osmotic water permeability. Additionally, in the brain, p-Coumaric acid mitigates oxidative stress, inflammation, and apoptosis53 and activates the PKA-CREB-Brain Derived Neurotrophic Factor (BDNF) signaling pathway54. To conclude, this study identified for the first time p-Coumaric acid as an allosteric modulator in OLE that possibly can directly bind to and activate the CaSR. We therefore propose that p-Coumaric acid, by activating the CaSR signaling, may exert beneficial effects in the alleviation of disorders characterized by abnormal renal reabsorption. Materials and methods Reagents and antibodies p-Coumaric acid (CAS No. 501-98-4) was bought from D.B.A. (DBA Italia s.r.l., Segrate, MI). Calcein-AM and Fura-2 AM were bought from Invitrogen (Thermo Fisher Scientific, Waltham, MA, USA). DDAVP was purchased by Sigma-Aldrich (St Louis, MO, USA). NPS-2143 hydrochloride was from Tocris Bioscience (Bio-Techne SRL, 20128 Milano, Italy). Aquaporin-2 (AQP2) was detected using specific antibodies (C-tail Ab) raised against a synthetic peptide corresponding to the last 15 C-terminal amino acids of human AQP2. The AQP2 antibody was bought from Santa Cruz Biotechnology (Texas, USA). The AQP2-pS256 antibody was purchased from abcam (Cambridge, UK). Anti-phospho-AQP2 (Ser261) antibody was from Novus Biologicals (Littleton, CO, USA). Secondary goat anti-rabbit horseradish peroxidase–coupled antibodies were obtained from Merck (Merck KGaA, Darmstadt, Germany). TransFectin Lipid Reagent and Clarity Western ECL Substrates were from Bio-Rad (Bio-Rad Laboratories, Inc., Hercules, CA, USA). Cell culture and treatments MCD4 cells derived from mouse cortical collecting duct M-1 cells55 that originated from mice transgenic for the early region of Simian virus 40 and purchased from the European Collection of Authenticated Cell Culture (ECACC, https://www.culturecollections.org.uk/nop/product/m-1). M-1 cells were then stably transfected with a plasmid encoding wild-type human AQP256 and the chimeric V2R-Rluc using the pantropic retroviral expression system by Clontech57. Cells were grown in a 1:1 mixture of Dulbecco’s modified Eagle’s medium and F-12 supplemented with 5% (v/v) fetal bovine serum, 1% (v/v) L-glutamine and 1% penicillin/streptomycin, 5 µM dexamethasone, 400 µg/ml G418 (for AQP2 resistance) and 1 µg/ml puromycin (for V2R-Rluc resistance), in a humidified atmosphere of 5% CO2 at 37 °C. MCD4 cells were left under basal condition (CTRL) or treated with p-Coumaric acid at 1 nM for 45 min, DDAVP at 100 nM for 30 min or co-treated with p-Coumaric acid and DDAVP or NPS-2143 at 10 µM for 10 min. Confocal microscopy Confocal studies were performed as previously described58. MCD4 cells grown on cell culture polyethylene terephthalate (PET) inserts to reproduce a polarized epithelial phenotype. Starting from day 7, transepithelial electrical resistance (TEER) was measured daily, and experiments were performed only when TEER values exceeded 600 Ω·cm². Cells were stimulated with DDAVP (100 nM for 45 min) in the presence or the absence of p-Coumaric acid (1 nM for 45 min). After fixation (4% paraformaldehyde), MCD4 cells were permeabilized with 0.1% Triton X-100 in PBS for 5 min. Cells were treated with 1% bovine serum albumin (BSA) in PBS and then incubated with the AQP2 antibodies overnight at 4 °C. Alexa Fluor 555 conjugated secondary antibodies were applied for 1 h at 37 °C. Images were captured with a confocal laser-scanning fluorescence microscope Leica TCS SP2 (Leica Microsystems, Heerbrugg, Switzerland). xz figures were quantitatively analyzed using ImageJ as shown in Supplementary Fig. 1. Briefly, F-actin staining (red channel) was used to visualize cell boundaries and the epithelial architecture. The apical membrane was identified as the luminal-facing surface of the polarized epithelial monolayer in xz sections. AQP2 fluorescence (green channel) was quantified by measuring the mean fluorescence intensity within a narrow ROI positioned at the luminal cell surface and comparing it with the intensity measured in an adjacent cytosolic compartment. The membrane/cytosol ratio was calculated after background subtraction. Cell lysates and immunoblotting Cells were grown on 60 mm Ø dishes and lysed in RIPA buffer containing 150 mM NaCl, 10 mM Tris-HCl pH 7.2, 0.1% SDS, 1.0% Triton X-100, 1% sodium deoxycholate and 5 mM EDTA in the presence of proteases (1 mM PMSF, 20 µM leupeptin, and 2 µg/ml pepstatin A) and phosphatases (10 mM NaF and 1 mM sodium orthovanadate) inhibitors. After sonication (60 kHz for 5 s), lysates were centrifuged at 12,000 x g for 10 min. The supernatants were collected and used for immunoblotting studies. According to the manufacturer’s protocols, SDS-PAGE was performed on 12% polyacrylamide gels (Bio-Rad Laboratories, Inc., Hercules, CA, USA). Protein bands were transferred onto the PVDF Transfer membrane (Thermo Fisher Scientific, Waltham, MA, United States) and subjected to immunoblotting using specific antibodies. To improve the sensitivity and specificity of the bands, the EveryBlot blocking buffer was applied (Bio-Rad Laboratories, Milan, Italy). Proteins were detected using Clarity Western ECL Substrate with the ChemiDoc System gels (Bio-Rad Laboratories, Milan, Italy). Stain-free technology gels were used to normalize specific bands to total protein content. Densitometric analysis of the bands was carried out with Image Lab software (Bio-Rad Laboratories, Milan, Italy). Fluorescence resonance energy transfer measurements (FRET) To detect changes in intracellular cAMP, FRET experiments were performed as already described13. MCD4 cells were transfected with the FRET-based sensors Epac H187 (Addgene plasmid # 170348; http://n2t.net/addgene:170348; RRID: Addgene_170348) that was kindly provided by Kees Jalink. The cells were transiently transfected using TransFectin lipid reagent in accordance with the manufacturer’s instructions (Bio-Rad Laboratories, Inc., Hercules, CA, USA). Visualization of ECFP- and/or EYFP-expressing cells and detection of FRET was performed using an inverted TE2000-S microscope (Nikon Eclipse microscope, Tokyo, Japan). Each image was corrected and analyzed as previously shown13,59. Water permeability assay Osmotic water permeability was measured as previously described60. Briefly, MCD4 cells were grown on 25 mm Ø glass coverslips. Cells were left untreated or stimulated with DDAVP (100 nM for 45 min). Alternatively, cells were incubated with p-Coumaric acid in the absence or the presence of DDAVP (100 nM for 45 min). Cells were loaded with 10 µM membrane-permeable Calcein-AM for 45 min at 37 °C, 5% CO2 in DMEM. Cytosolic fluorescent Calcein displays a concentration-dependent quenching by proteins or salts. Measured changes in fluorescence were directly proportional to changes in cell volume. Therefore, hypertonic solutions led to a rapid drop in the Calcein fluorescence due to the quenching of the fluorophore. Measurements were carried out using an inverted TE2000-S microscope (Nikon Eclipse Microscope, Tokyo, Japan) equipped for single-cell fluorescence measurements and imaging analysis. The Calcein-AM-loaded samples were excited at 490 nm. Fluorescence measurements, following isosmotic (140 mM NaCl, 1 mM MgCl2, 5 mM KCl, 1 mM CaCl2, 5 mM Glucose,10 mM HEPES, pH 7.4) or hyperosmotic (isosmotic solution added with 135 mM Mannitol) solutions, were obtained using Metafluor software (Molecular Devices, MDS Analytical Technologies, Toronto, Canada). Time course fluorescence data after perfusing cells with hyper- or isosmotic solutions were recorded every 3 s and over about 3 min. The time constant of cell shrinkage because of the hypertonic stimulus was obtained by fitting data with an exponential function13. Intracellular calcium measurements Changes in intracellular calcium measurements were evaluated, as already showed51. Briefly, MCD4 cells were loaded with 4 µM Fura-2 AM for 15 min at 37 °C in the culture medium. Ringer’s solution was used to perfuse cells during the experiment, containing 140 mM NaCl, 5 mM KCl, 1 mM MgCl2, 10 mM HEPES, 5 mM Glucose, 1.8 mM CaCl2 (pH 7.4). In fluorescence measurements, the coverslips with dye-loaded cells were mounted in a perfusion chamber (FCS2 Closed Chamber System, Bioptechs, Butler, U.S.A.). Measurements were carried out using an inverted TE2000-S microscope (Nikon Eclipse Microscope, Tokyo, Japan). The ratio of 340 and 380 nm fluorescence intensities was plotted and calculated as a fluorescence change. Fluorescence signals measured in stimulated cells with p-Coumaric acid (1 nM) or NPS-2143 (10 µM) were compared to those detected in the same cell type obtained after stimulation with ATP (100 µM), a calcium-mediated agonist used as an internal positive control. For allosteric modulator experiments fluorescence signals were measured in cells stimulated with different extracellular Ca2+ concentration (1, 2.5, 5 and 10 mM) in the absence and in the presence of 1 nM p-Coumaric acid. Molecular modeling investigations The atomic coordinates of the CaSR dimer were taken from 9asb.pdb61, showing a CaSR dimer hosting a cyclomethyltryptophan ligand coordinating a Ca2+ ion plus other two Ca2+ ions for each monomer above and below the cyclomethyltryptophan-Ca2+ binding region, for each monomer61. In addition, the positive modulator Tecalcet was crystallized in the CaSR allosteric binding region61. The binding pocket of the positive modulator Tecalcet62,63 (https://pubchem.ncbi.nlm.nih.gov/compound/158797) was thus explored for investigating a possible binding of the p-Coumaric acid in both chains of CaSR dimer. With this aim, the CaSR wildtype protein was prepared for the docking simulation converting CaSR protein “.pdb” file in a “.pdbqt” file by using the MGLTools of AutoDock64–66. The generated “.pdbqt” file contained gasteiger charges for every atom present in the original CaSR “.pdb” file. Similarly, Tecalcet and p-Coumaric acid (“.mol2” files, energetically minimized) to be tested in the docking analysis were converted in “.pdbqt” files by using AutoDock64–66. In order to set-up a validated docking protocol for the docking analysis to be performed with p-Coumaric acid, firstly we performed a re-docking simulation by using AutoDock 4.2, on the WT CaSR deprived of Tecalcet, by varying gridbox size and center, aiming to reproduce the same Tecalcet binding pose obtained in the reference CaSR-Tecalcet crystallized structure (9asb.pdb), according to previously validated protocols67,68. For each re-docking simulation the parameter “rmstol” was set to 1.5 Å; the parameter “ga_pop_size” was set to 150, the parameter “ga_num_evals” was set to 2.5 × 106, the parameter “ga_num_generations” was set to 2.7 × 104. The number of top individuals to survive to the next generation was set to 5 and a ranked cluster analysis of the 50 ligand poses generated was performed. For details about the significance of the above reported parameter names see Morris et al.64. The gridbox that better reproduced the pose of the wild-type CaSR in complex with Tecalcet (according to the procedure above described) consists of 24 × 24 × 32 gridpoints in the x–y–z Cartesian space centred on the point of coordinates 203.879 (x), 237.425 (y), 212.302 (z), for CaSR chain A and 204.305 (x), 212.848 (y), 205.052 (z) for CaSR chain B. The spacing, namely the distance between adjacent gridpoints within the gridbox, was set for this simulation to 0.385 Å. Obtaining a good re-docked conformation (i.e., a pose of the Tecalcet-ligand within the Tecalcet binding region very similar to the Tecalcet pose observed in the crystallized complex) allows establishing residues most likely directly involved in binding interactions with Tecalcet. The re-docked conformation and the reference Tecalcet ligand have been used as a reference system, e.g. to quantify the affinity of p-Coumaric acid, by using as a discriminant and positive control, the relative binding energy calculated for the best re-docked CaSR-Tecalcet protein-ligand complex conformation. Thus, we used the same gridbox for docking p-Coumaric acid in the Tecalcet binding pocket. The Lamarckian Genetic Algorithm, as previously described64–68, was used both for re-docking Tecalcet and docking p-Coumaric acid. For the docking of the p-Coumaric acid the same docking parameters were used, namely the parameter “ga_pop_size” was set to 150, the parameter “ga_num_evals” was set to 10 × 106, the parameter “ga_num_generations” was set to 3 × 104. At the end of the simulation, we explored by visual inspection the obtained binding poses for comparative purposes. Statistical analysis The one-way analysis of variance (ANOVA) followed by Tukey’s multiple comparisons test was applied. In alternative, the non-parametric Kruskal-Wallis test was applied. Values are shown as means ± Standard Error of the Mean (S.E.M.). The difference of p < 0.05 was considered statistically significant. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1. Supplementary Material 1.