Physiological concentration of arginine vasopressin (AVP) induces intracellular Ca²⁺ mobilization in form of oscillation in isolated perfused rat inner medullary collecting duct (IMCD) (Yip, 2002). Confocal fluorescence microscopy revealed that each IMCD cells have their own unique oscillatory frequency and amplitude. Such Ca²⁺ mobilization is essential for the associated apical exocytosis, as intracellular Ca⁺² chelators inhibit both AVP-induced Ca²⁺ mobilization and apical exocytosis in perfused IMCD. AVP exerts its actions via binding of the V₂-receptors to stimulate adenylate cyclase and cAMP production in IMCD cells (Knepper and Inoue, 1997), the latter is mediated by adenylyl cyclase 6 (Rieg et al., 2010). It has traditionally been thought that cAMP activates the protein kinase A (PKA)-dependent signaling pathway to mediate AVP-regulated osmotic water permeability of IMCD. However, our previous study found that PKA inhibitors did not prevent AVP-induced Ca²⁺ mobilization and oscillation. Instead, the cAMP analog 8-pCPT-2′-O-Me-cAMP, which specifically activates exchange protein directly activated by cAMP (Epac) but not PKA, triggered intracellular Ca²⁺ mobilization and apical exocytosis of aquaporin-2 (AQP2) in perfused IMCD (Yip, 2006). Moreover, flash photolysis of caged cADP-ribose (an endogenous ligand of ryanodine receptors) activated Ca²⁺ oscillations resembling AVP-induced Ca²⁺ response (Yip and Sham, 2011). Previous studies showed that Ca²⁺ release from ryanodine receptors (RyRs) is essential in AVP-mediated AQP2 trafficking (Chou et al., 2000; Yip, 2002), and the process is independent of the phosphoinositol signaling pathway (Chou et al., 1998). AVP could also trigger Ca²⁺ influx via the store-operated Ca²⁺ entry (SOCE) mechanism. It was concluded that AVP-induced Ca²⁺ oscillation in IMCD is mediated by an Epac-dependent mechanism through the interplay of Ca²⁺ release from ryanodine receptors and a Ca²⁺ influx mechanism involving SOCE (Yip and Sham, 2011). Epac-induced Ca²⁺ release from RYRs-gated Ca²⁺ stores have been reported in other cell types. In cardiac myocytes, Epac-activation enhances RYR activity through protein kinase C_epsilon and Ca²⁺/calmodulain kinase II (CaMKII)-dependent phosphorylation of RYRs (Pereira et al., 2007; Oestreich et al., 2009), leading to SR Ca²⁺ leak and arrhythmia (Pereira et al., 2013; Li et al., 2017; Pereira et al., 2017). Epac-induced activation of RYRs also causes membrane hyperpolarization and relaxation of mesenteric arteries through Ca²⁺-sensitive K⁺ channel activation (Roberts et al., 2013).

Murine principal kidney cortical collecting duct (mpkCCD) cells are commonly used cell model used for transcriptomic and phosphoproteomic studies of AVP-signaling in kidney collecting ducts (Rinschen et al., 2010; Huling et al., 2012; Sandoval et al., 2013; Yang et al., 2022; Park et al., 2023). It is assumed that mpkCCD cell retains the feature of intact collecting duct cell in AVP-induced signaling events of AQP2 trafficking. We have demonstrated cAMP-dependent vectorial trafficking and exocytosis of AQP2 tagged with photoactivable fluorescent protein in mpkCCD cells at real time (Yip et al., 2015). It has also been shown in mpkCCD cells that Wnt5A, an endogenous ligand of the non-canonical branch of the Wnt pathway, is capable of inducing AQP2 apical expression and trafficking via basolateral Fzd receptors-mediated Ca²⁺ mobilization without activation of cAMP/PKA signal pathway (Ando et al., 2016). These observations highlight the potential of targeting Ca²⁺ pathways to ameliorate polyuria associated with nephrogenic diabetes insipidus (Mortensen et al., 2020). However, there is no information on the mechanisms underlying the dynamics of intracellular Ca²⁺ mobilization in mpkCCD cells. It is also unclear whether mpkCCD cells retain the specific properties of Ca²⁺ mobilization observed in perfused IMCD. In the present study, we sought to verify mpkCCD cells as a reliable model representing collecting duct cells for the study of the intracellular Ca²⁺ stores, the mechanisms of Ca²⁺ release, and extracellular Ca²⁺ influx. Moreover, special emphasis has been placed on the Epac-induced temporal relationship of Ca²⁺ dynamics in the cytosol, endoplasmic reticulum (ER) and mitochondria. Our results demonstrate that mpkCCD cells display similar characteristics of intracellular Ca²⁺ mobilization observed in intact cells of collecting duct, and that the Epac agonist triggered intracellular Ca²⁺ mobilization and oscillation are mediated by RyR-gated Ca²⁺ release and SOCE associated with reciprocal decrease of Ca²⁺ content in the ER. Moreover, the Epac agonist can effectively shuttle ER luminal Ca²⁺ to both the cytosol and mitochondrial matrix.

Epac-dependent Ca²⁺ mobilization was associated with AVP-induced apical exocytosis in perfused IMCD. mpkCCD cells have been used as the cell model for transcriptomic and phosphoproteomic studies of AVP-signaling in the collecting duct (Rinschen et al., 2010; Huling et al., 2012; Sandoval et al., 2013; Yang et al., 2022; Park et al., 2023). We have demonstrated that mpkCCD cells retain the characteristics of Epac-dependent Ca²⁺ mobilization as in intact IMCD cells. Taking advantage of expressing ER and mitochondrial specific biosensor proteins in mpkCCD cells, the dynamic properties and relationship between cytosolic Ca²⁺, ER luminal Ca²⁺, and mitochondrial matrix Ca²⁺ were characterized. Epac-agonist mobilized Ca²⁺ from ER Ca²⁺ stores, depleted ER luminal Ca²⁺, and activated SOCE in mpkCCD cells. The oscillation of cytosolic Ca²⁺ triggered by Epac-agonist was entrained to Ca²⁺ oscillation in mitochondrial matrix, while 180° out-of-phase to the oscillation in ER luminal Ca²⁺. These observations indicated that Epac-mediated oscillatory Ca²⁺ signaling event is an integrated process which involves interplay of luminal ER Ca²⁺ release and refill, Ca²⁺ entry and efflux in mitochondrial matrix, and extracellular Ca²⁺ entry secondary to ER Ca²⁺ depletion in mpkCCD cells.

Time series of Ca²⁺ oscillation extracted from mpkCCD cells were analyzed in frequency domain using algorithm based on Fast Fourier Transform. Frequencies of Epac-dependent oscillations were detected in the range between 0.007 and 0.1 Hz, which is similar to the frequency ranges induced by cADP-ribose, an endogenous agonist of RyRs, observed in intact IMCD cells (Yip, 2002; Yip and Sham, 2011). Moreover, the sustained Me-cAMP-mediated Ca²⁺ oscillation in mpkCCD cells is similar to those observed in perfused IMCD triggered by caged cyclic-ADP-ribose (Yip and Sham, 2011). However, the 4-CMC-induced Ca²⁺ transient is more transient in the IMCD cells. The disparity in the kinetic profile of 4-CMC-triggered Ca²⁺ oscillations could be related to the differences in the agonist sensitivity, the activation/inactivation kinetics, sensitization of Ca²⁺-induced-Ca²⁺ release of the RyRs. Nevertheless, the complete inhibition of Epac-agonist-induced Ca²⁺ oscillation with ryanodine, but not by the IP₃R-antagonist xestrospongin C, suggests that RyR is the primary Ca²⁺ source contributing to the Epac-induced Ca²⁺ oscillation in the mpkCCD cells.

It has been established that AVP regulates AQP2 shuttling through a cAMP-dependent pathway, and PKA has been considered as the only effector protein of cAMP for mediating AVP-regulated water permeability in kidney collecting ducts (Knepper and Inoue, 1997). It is now known that Epac is an important effector protein of cAMP. Epac1 and Epac2 isoforms are expressed in collecting duct and mpkCCD cells (Li et al., 2008; Kortenoeven et al., 2012). Our previous study found that Epac activation, but not PKA activation, mimics AVP in triggering Ca²⁺ mobilization and oscillations and induces apical shuttling of AQP2 in perfused IMCD (Yip, 2006). In the present study, PKA specific cAMP analog (6-Bnz-cAMP/AM) did not trigger Ca²⁺ oscillations in mpkCCD cells, while Epac-specific cAMP (Me-cAMP/AM) triggered long lasting Ca²⁺ oscillations. Moreover, long-term regulation of AQP2 by AVP in mpkCCD cells is mediated by Epac but not by PKA (Kortenoeven et al., 2012). Such evidence is consistent with an Epac-dependent signal pathway for regulation of collecting duct water permeability. It is also consistent that mice lacking Epac1 or Epac2 showed impaired urinary concentration ability and augmented urinary excretion of Na⁺ and urea (Cherezova et al., 2019). However, no defects in AVP-induced Ca²⁺ signaling in split-opened collecting ducts or changes in AQP2 protein abundance were observed. The urinary concentrating defect might be caused by reduced expression of the Na⁺/H⁺ exchanger isoform 3 (NHE3) (Cherezova et al., 2019). An inhibitory effect of Epac1 on NHE3 activity was previously shown in opossum kidney cells and mouse kidney slices (Honegger et al., 2006). A recent study reported compromised tight junctions in the collecting duct of Epac1 knockout mice in conjunction with a reduced papillary osmolarity (Sivertsen Asrud et al., 2020), suggesting Epac1 is involved in regulating paracellular permeability in the collecting duct. The reason for the discrepancies in the two knock-out mice studies is unclear but could be related to mouse dietary conditions or the existence of a different microbiome between different institutions.

Our data also show that Epac-agonist triggered Ca²⁺ oscillations resemble those activated by a RyR agonist in mpkCCD cells, suggesting that Epac-agonist mobilizes ER Ca²⁺ in mpkCCD cells via RyRs. Organelle-specific Ca²⁺-sensitive biosensors expressed in mpkCCD cells showed that the Epac agonist triggered synchronized Ca²⁺ spikes in cytosol and mitochondrial matrix, which are temporally correlated with reciprocal changes in ER luminal Ca²⁺. These observations indicated that agonist-induced Epac activation not only triggered release of Ca²⁺ into cytosol, but also transferred Ca²⁺ from ER to mitochondria.

The resting [Ca²⁺] of mitochondrial matrix is comparable to the resting cytosolic [Ca²⁺]. Mitochondrial Ca²⁺ uptake takes place at specialized microdomains where cytosolic [Ca²⁺] are high. Such microdomains are localized in the mitochondria-associated membranes (MAMs), where the endoplasmic reticulum membrane is within 10–30 nm from the outer mitochondrial membrane (Hajnoczky et al., 2002). Ca²⁺ released from ER enters the intermembrane space through voltage-dependent anion channels (VDACs) localized in the outer membrane, and then enters mitochondrial matrix through the mitochondrial Ca²⁺ uniporter (MCU) of the inner membrane (Patergnani et al., 2011; Giorgi et al., 2015). RyR agonist (4-CMC) triggered similar Ca²⁺ transfer from ER to mitochondria as Epac-agonist suggested that the Epac-induced ER-mitochondrial Ca²⁺ transfer is mediated by RyRs in MAMs of mpkCCD cells. IP₃Rs and RyRs have been localized in MAMs (Hajnoczky et al., 2002; Chen et al., 2012; Bartok et al., 2019), but their distribution in mpkCCD cells or native renal collecting duct cells is unclear. The current study explored on this knowledge gap. Exogenous Bt₃-Ins(1,3,5)P₃/AM, which effectively activated cytosolic Ca²⁺ oscillation, only triggered a decrease in ER luminal Ca²⁺ but not concomitantly increased mitochondria Ca²⁺ in mpkCCD cells. It is possible that IP₃Rs are less efficacious in facilitating ER-mitochondrial Ca²⁺ transfer in mpkCCD cells; or the cell permeant Bt₃-Ins(1,3,5)P₃/AM, had a poor access to the IP₃Rs in MAMs. To test the latter hypothesis, the native agonist ATP was applied to increase endogenous IP₃ abundance in mpkCCD cells. ATP triggered synchronized increase in cytosolic [Ca²⁺] with concomitant elevation in mitochondrial matrix [Ca²⁺] and depletion ER luminal Ca²⁺. ATP-induced Ca²⁺ transfer between ER and mitochondria was also visualized in cells co-expressed with ER and mitochondrial Ca²⁺ biosensors. These observations indicated that IP₃Rs is capable of mediating Ca²⁺ transfer between ER and mitochondria in mpkCCD cells. However, the kinetics of ATP-induced cytosolic, endoplasmic, and mitochondrial Ca²⁺ responses are distinctly different from those induced by Epac or RyR agonists, distinguishing the Ca²⁺ signals activated by the two different agonists induced signaling pathways. Our studies also support that this mechanism is at play in vivo and it is noteworthy that 5 µM ATP also triggered Ca²⁺ mobilization and oscillations in isolated perfused IMCD (Figure 5E), confirming that purinergic receptor mediated Ca²⁺ mobilization is present in both intact IMCD cells and mpkCCD cells. Consistent with this, studies in acutely isolated connecting tubule/collecting duct of mice support that an acute increase in cytosolic [Ca²⁺] inhibits ENaC activity (Mamenko et al., 2011) mediated by P2Y₂ receptor activation (Pochynyuk et al., 2008).

Mitochondrial Ca²⁺dynamics plays important roles in intracellular Ca²⁺signaling, cell metabolism, cell survival, and other cell-type specific functions (Rizzuto et al., 2012). As described above MCU supports cytoplasmic Ca²⁺ oscillations, SOCE and Ca²⁺-dependent gene expression in response to receptor-mediated stimulation (Samanta et al., 2014). It has been proposed that mitochondrial Ca²⁺ shuttling via MCU sustains the cytosolic Ca²⁺ signal by preventing Ca²⁺-dependent inactivation of IP₃Rs and store-operated CRAC channels (Yoast et al., 2021). ER-mitochondrial Ca²⁺ transfer also stimulates Ca²⁺-sensitive dehydrogenases and respiratory chain components to promote oxidative phosphorylation, ATP, and ROS production (Territo et al., 2000; Territo et al., 2001; Hou et al., 2013). The physiological implications for the ER-mitochondrial Ca²⁺ transfer in the regulation of cellular functions in renal tubular cells remain to be determined.

In conclusion, mpkCCD cells retained all reported features of Epac-induced Ca²⁺ mobilization observed in isolated perfused IMCD. The temporal relationship between cytosolic Ca²⁺, ER luminal Ca²⁺, and mitochondrial matrix Ca²⁺ activated by Epac and RyR-agonists are highly compatible, but is distinctly different from those induced by IP₃R stimulation. Furthermore, we have provided the first characterization of ER-mitochondrial Ca²⁺ transfer in mpkCCD cell, which can be used as a renal cell model to address novel questions of how mitochondrial Ca²⁺ regulates cytosolic Ca²⁺ signals, inter-organellar Ca²⁺ signaling, and other renal tubular functions.