Pancreatic acinar cell signalling and function exhibit an absolute requirement for activation of Gα_q

Abstract

Elevated cytoplasmic [Ca²⁺] and protein kinase C (PKC) activation are key signalling events driving secretion in pancreatic acinar cells after stimulation by the secretagogues cholecystokinin (CCK) and acetylcholine (ACh). Although both ACh and CCK binding to their cognate receptors activates G_q/11 proteins, leading to inositol 1,4,5-trisphosphate (IP₃)-mediated Ca²⁺ release and diacylglycerol (DAG)-dependent PKC activation, it has been proposed that physiological CCK stimulation bypasses this canonical pathway, instead mobilizing Ca²⁺ via production of nicotinic acid adenine dinucleotide phosphate (NAADP). We reassessed the role of G_q/11 signalling in CCK-induced responses using a bioluminescence resonance energy transfer (BRET) assay, demonstrating that both CCK1 (CCK1R) and muscarinic M3 receptors (M3R) engage G_q/11 along with other Gα subunits. Importantly YM-254890, a G_q/11 antagonist, inhibited coupling through G_q/11 but did not alter the interactions of CCK1R or M3R with other G protein families. YM-254890 eliminated CCK1R- and M3R-induced Ca²⁺ signals in isolated acinar cells. Consistent with the in vitro data, systemic CCK injection, intrinsic neural stimulation or feeding failed to elicit Ca²⁺ responses in vivo in mice pre-treated with YM-254890, indicating that physiological stimulation of Ca²⁺ signalling events requires G_q/11 activation. Additionally YM-254890 suppressed Ca²⁺-activated Cl⁻ currents, a key event underlying fluid secretion, and amylase secretion in acini after CCK or ACh stimulation. These findings establish that CCK- and ACh-induced exocrine pancreatic secretion strictly requires G_q/11 activation, leading to IP₃ generation, DAG production and downstream signalling that is essential for physiological function.

Graphical Abstract

G_q/11 activation is required for signalling and function after stimulation of pancreatic acinar cells with ACh and CCK. A, A conventional model of signalling through M3R and CCK1R in pancreatic acinar cells. ACh couples through G_q/11 to result in IP₃ and DAG) formation, which results in an elevation in [Ca²⁺] leading to fluid and protein secretion. CCK is reported to bypass this pathway but couples to the formation of NAADP, resulting in an elevation in Ca²⁺ and fluid and protein secretion. B, Using a BRET assay the coupling of CCK1R and M3R to G protein α subunits is defined together with the exquisite specificity of the G_q/11 blocker YM-254890. C, YM-254890 blocks Ca²⁺ signalling and function by both CCK and ACh, indicating that IP₃ (and DAG) generation is required after stimulation by both secretagogues. D, New model incorporating the requirement for G_q/11 signalling and function after stimulation of pancreatic acinar cells.

Introduction

The primary function of the exocrine pancreas is to produce pancreatic juice which initiates the digestion of macronutrients after feeding (Williams, 1993, 2001). Secretagogue-stimulated intracellular Ca²⁺ signals play fundamental roles in activating downstream effectors important for both the exocytosis of vesicles containing digestive enzymes and the secretion of fluid from pancreatic acinar cells. During the cephalic phase of gastrointestinal (GI) activity, acinar cells initially secrete in response to ACh released by parasympathetic vagal neural input. Subsequently during the intestinal phase, partially digested macronutrients stimulate endocrine I cells in the duodenum to secrete the peptide hormone CCK, which enters the circulation and then acts directly on acinar cells (Liddle, 1995).

The targets of ACh and CCK are M3R and CCK1R also known as CCKAR receptors, respectively (Dehaye et al., 1984; Jensen et al., 1980; Sankaran et al., 1980; Schmid et al., 1998). Both receptors are G protein coupled and act through heterotrimeric G protein α subunits, predominately the G_q/11 family (Matozaki et al., 1988; Yule et al., 1999), but signalling through other G protein α subunits has been reported (Schnefel et al., 1990; Yule et al., 1993). G_q/11 activation leads to the production of IP_3, and DAG after phospholipase C activation. As first demonstrated in pancreatic acinar cells (Streb et al., 1983), IP₃ results in Ca²⁺ release via IP₃ receptors (IP₃R) resident in the endoplasmic reticulum (ER) (Lee et al., 1997; Yule et al., 1997), whereas DAG stimulates PKC. Both an elevation in [Ca²⁺] and an increase in PKC activity are necessary to trigger exocytosis, with synergism between the two limbs of the pathway mediating digestive enzyme secretion (Williams, 2001). However an elevation in [Ca²⁺] is largely sufficient to activate the ion channels necessary for ion and fluid transport from the acinar cell (Petersen & Findlay, 1987). Consistent with the central role of this pathway for pancreatic exocrine physiology, secretagogue-stimulated Ca²⁺ signals and secretion are absent in transgenic animals engineered to lack IP₃R type 2/3 (Futatsugi et al., 2005; Giovannucci et al., 2002; Kidd & Thorn, 2000).

Although the canonical G_q/11-IP₃ signalling pathway is activated after M3R and CCK1R stimulation, the notion that inositol IP₃serves as the only second messenger driving all secretagogue-mediated Ca²⁺ signals has been challenged (Petersen, 2023). Early studies demonstrated that although both CCK and ACh stimulation led to IP₃ formation, downstream Ca²⁺ signals and amylase secretion were detectable at significantly lower secretagogue concentrations (Matozaki & Williams, 1989, 1992; Matozaki et al., 1990; Powers, Johnson et al., 1985; Powers, Saluja et al., 1985). Although this discrepancy may be attributed to the limited sensitivity of IP₃ assays and the substantial amplification within the signalling cascade, subsequent reports have revealed that ACh and CCK evoke distinct subcellular spatiotemporal Ca²⁺ signalling ‘signatures’ (Cancela et al., 2000, 2002; Menteyne et al., 2006; Osipchuk et al., 1990; Shin et al., 2001; Takano & Yule, 2023, 2024; Yamasaki et al., 2004; Yule et al., 1991). These findings suggest that secretagogue-induced signalling is not identical and may involve additional Ca²⁺-releasing messengers.

Supporting this hypothesis pancreatic acinar cells were the first native mammalian cells in which NAADP was identified as a potent mediator of Ca²⁺ release. This messenger activates two-pore channel (TPC) family members localized to endolysosomal acidic organelles (Calcraft et al., 2009). Notably NAADP is produced in response to physiological CCK stimulation but is apparently not produced after muscarinic receptor activation by ACh (Yamasaki et al., 2005). Consistent with these data the NAADP antagonist trans-NED-19 has been shown to eliminate Ca²⁺ responses to low CCK concentrations while leaving muscarinic receptor-mediated signalling intact (Gerasimenko et al., 2015; Tsvilovskyy et al., 2024). In addition the effects of NAADP are reduced in isolated permeabilized acini prepared from TPC2 null mice (Gerasimenko et al., 2015). Although the precise biochemical pathway responsible for NAADP synthesis in pancreatic acinar cells remains undefined, studies in other cell types indicate that NAADP production is likely not directly linked to G_q/11 activation (Gu et al., 2021). In sum these findings have been interpreted to indicate that IP₃ production does not occur in response to physiological CCK concentrations and is not essential for initiating Ca²⁺ signalling events and therefore physiological function (Cancela, 2001; Menteyne et al., 2006).

As a further complexity, paradoxically, inhibitors of IP₃R and ryanodine receptors (RyRs) have been reported to eliminate NAADP-induced Ca²⁺ signals (Cancela et al., 2000). To reconcile these data a model has been developed that posits that at physiological concentrations, CCK-stimulated NAADP formation occurs in the absence of elevated levels of IP₃, resulting in limited local Ca²⁺ release. This event provides an initial trigger, which through Ca²⁺-induced Ca²⁺ release (CICR) recruits further activity through neighbouring IP₃R and RyR (Roderick & Bootman, 2003; Roderick et al., 2003). Because IP₃R activity has an absolute requirement for IP₃ binding (Alzayady et al., 2016; Foskett et al., 2007), this suggests that the local Ca²⁺ release triggered by NAADP sensitizes IP₃R to open at ambient levels of IP₃ without the necessity of CCK-induced formation of IP₃. In contrast, ACh stimulation simply results in the formation of IP₃, triggering Ca²⁺ release and CICR amplification through IP₃R and RyR (Cancela et al., 2002; Petersen, 2023). Recently it has been reported that fundamental Ca²⁺ release events mediated by TPC2 channels – termed ‘Ca²⁺ tuffs’ – do not activate IP₃Rs under basal conditions. However they enhance IP₃R activity when intracellular IP₃ levels are elevated. These findings suggest that, at least in this cell culture model, TPC2-induced Ca²⁺ release does not sensitize IP₃R to respond to increasing [Ca²⁺] at ambient, non-stimulated levels of IP₃ (Yuan et al., 2024).

As noted, although there is evidence that NAADP can release Ca²⁺ in response to CCK stimulation, the absence of a requirement for G_q/11-linked IP₃ production is difficult to reconcile with both receptors clearly coupling to G_q/11 and particularly the requirement for DAG formation and PKC activation for stimulated vesicle exocytosis. Therefore in this study we reinvestigated the role of G_q/11 signalling in response to CCK1R and M3R stimulation. Initially in a heterologous expression system, we show that CCK1R couples to Gα_q together with a wide variety of Gα subunit families, whereas M3R is less promiscuous and activates predominately Gα_q family members. We further demonstrate that an inhibitor of G_q/11, YM-254890 (Schnefel et al., 1990; Schrage et al., 2015; Voss, 2023), specifically inhibits coupling between both M3R and CCK1R and G_q/11 without altering signalling through other Gα protein families. In pancreatic acinar cells YM-254890 eliminated Ca²⁺ signals in vitro in response to both CCK and muscarinic receptor stimulation over a wide range of concentrations representing threshold through physiological stimulation and in vivo in response to nervous stimulation, CCK administration and feeding. Finally activation of both Ca²⁺-activated Cl⁻ currents, pivotal to fluid secretion and amylase secretion, was completely inhibited by YM-254890. These data are consistent with both CCK and ACh signalling through G_q/11 to increase IP₃ and DAG, and further this event is necessary to evoke physiological fluid and enzyme secretion from pancreatic acinar cells.

Results

Coupling of both M3R and CCK1R to G_q

We applied a bioluminescence resonance energy transfer (BRET)-based assay (G protein nanoBRET) to investigate the proximal events after M3R and CCK1R activation. The assay relies on the generation of a BRET signal between Venus-Gβ1γ 2 and masGRK3CT-Nluc, which occurs after cell surface receptor activation-induced dissociation of heterotrimeric G protein complexes (Fig. 1A). First in HEK293 cells expressing either M3R or CCK1R, we determined the concentration–response relationships for M3R and CCK1R coupling through G_q in live cells, as described previously (Alabdali et al., 2023). Consistent with the existing literature our in vitro findings indicated activation of the heterotrimeric G_q protein in response to M3R stimulation by Carbachol (CCh) (pEC50 = 6.83) (Fig. 1B). Similarly treatment with CCK demonstrated activation of the heterotrimeric G_q protein (pEC50 = 9.19) (Fig. 1C).

Figure 1. Concentration–response curves for M3R (muscarinic M3 receptor) and CCKR1.

A, Schematics of the G protein nanoBRET assay applied to measure specific Gα protein coupling profiles. Upon agonist interaction with the receptor and G protein dissociation, the Gβγ heterodimer fused with acceptor Venus interacts with GRK3ct fused with donor NanoLuc generating a BRET (bioluminescence resonance energy transfer) signal. B, Concentration–response curve for G_q activation by M3R in response to CCh. C, CCKR1 activation of Gα_q with application of increasing concentration of CCK. Data are expressed as mean ΔBRET ratio ± SD. N = 3, non-linear regression fit four parameters.

YM-254890 selectively attenuates coupling through G_q

The macrocyclic depsipeptide, YM-254890, has been reported to inhibit cellular responses through G_q-coupled receptors by acting as a guanine nucleotide dissociation inhibitor, preventing the separation of G_q heterotrimers and thus activation of downstream effectors (Schnefel et al., 1990; Schrage et al., 2015; Voss, 2023). To test the ability of YM-254890 to inhibit G_q signalling through M3R and CCK1R, we next determined the concentration–inhibition relationships for saturating concentrations of CCh and CCK in presence of YM-254890. In cells expressing M3R YM-254890 dose dependently inhibited CCh-induced coupling to G_q (pIC50 = 5.15; Fig. 2A). Similarly YM-254890 dose dependently inhibited CCK1R response to CCK with a similar IC50 (pIC50 = 5.38; Fig. 2B). As initial evidence that YM-254890 specifically inhibits G_q-coupled events, we probed the effect of YM-254890 on the inhibition of four G-protein coupled receptors (GPCRs) primarily coupled to each of the four G protein sub-families. Agonist-induced activation of G_s-coupled dopamine D1 receptor (D1R), G_i/o-coupled dopamine D2 receptor (D2R) and G12/13-coupled lysophosphatidic acid receptor 2 (LPA2R) were not impacted by treatments with YM-254890 (Fig. 2C). However activation of G_q-coupled serotonin 2A receptor (5-HT2AR) was completely eliminated by treatment with 25 μMYM-254890 (Fig. 2C).

Figure 2. Gα protein inhibitory profile of YM-254890.

A and B, Concentration–response curves for YM-254890 on G_q activation by M3R and CCKR1 in response to application of 100 μM CCh and 1 μM CCK, respectively, measured via G protein NanoBRET assay. The pIC50 value of YM-254890 was similar between the two receptors (pIC50 = 5.15 for M3R, pIC50 = 5.38 for CCKR1). N = 3, non-linear regression fit four parameters. C, Inhibitory effect of 25 μM YM-254890 on activation of G_s by dopamine D1 receptor (D1R), GoA by dopamine 2 receptor (D2R), G₁₃ by lysophosphatidic acid receptor 2 (LPA2R) and G_q by serotonin 2A receptor (5-HT2AR). **P < 0.01. N = 3, paired t test. Data are expressed as mean ± SD.

Conceivably Ca²⁺ signals and activation of physiological responses after M3R or CCK1R stimulation could result from activating other G protein family members. Therefore we next investigated whether M3R and CCK1R can activate other G proteins by studying BRET between 12 Gα proteins, covering the 4 G protein families. Experiments were performed at saturating concentrations of the muscarinic agonist carbachol (CCh, 100 μM) and CCK (1 μM). M3R activation resulted in the dissociation of seven G protein heterotrimers across three families (G_q/11, G_i/o and G_12/13), with G_q, G₁₁ and G₁₅ coupling prominently (Fig. 3A). As previously reported CCK1R exhibited a high level of promiscuity as it stimulated the dissociation of each of the 12 Gα proteins tested (Fig. 3B). To further establish the selectivity of YM-254890 to inhibit G_q coupling between M3R and CCK1R, we examined the extent of signal inhibition by YM-254890 for each Gα protein activated by the two GPCRs. We observed that YM-254890 effectively inhibited receptor-induced G protein dissociation for G_q and G₁₁ in response to agonist stimulation of either M3R (Fig. 4A) or CCK1R (Fig. 4B). Activation of the other members of the G_q family, namely G₁₄ and G₁₅, was only partially or not significantly affected by treatments with YM-254890 (Fig. 4A,B). In sum these data indicate that M3R and CCKR1 can couple to G_q and a variety of other heterotrimeric G protein families and that YM-254890 is an exquisitely selective inhibitor of G_q/11 signalling.

Figure 3. Gα protein fingerprint of M3R (muscarinic M3 receptor) and CCKR1.

A and B, M3R and CCKR1 Gα protein activation profile upon agonist application (100 μM CCh and 1 μM CCK (cholecystokinin), respectively) measured by G protein NanoBRET assay. Twelve different Gα proteins from the 4 subfamilies were tested: G_s, G_oA, G_i1,G_i2, G_i3, G_z, G_q, G₁₁, G₁₄, G₁₅, G₁₂ and G₁₃. An average ΔBRET ratio >0.01 was considered a positive Gα protein activation. N = 3. Data are expressed as mean ± SD.

Figure 4. YM-254890 prevents activation of G_q/11 family members downstream M3R (muscarinic M3 receptor) and CCKR1 signalling.

A and B, G protein NanoBret assay testing inhibitory effect of YM-254890 (25 μM) or vehicle pre-treatment on Gα proteins downstream of M3R and CCKR1 signalling. M3R and CCKR1 were activated by the application of 100 μM CCh and 1 μM CCK, respectively. Data are expressed as mean ± SD. N = 3; **P < 0.01 and ***P < 0.001, paired t test.

Inhibition of G_q/11 by YM-254890 eliminates Ca²⁺ signals in pancreatic acinar cells in vitro

Having established the specificity of YM-254890 to inhibit G_q/11-mediated events, we next investigated the effects of YM-254890 on secretagogue-stimulated Ca²⁺ signals in acutely isolated pancreatic acini. In fura-2 AM loaded pancreatic acini, we utilized a protocol in which cells were treated twice with an identical concentration of CCh and then subsequently exposed to CCK (Fig. 5A). Essentially identical responses could be evoked to CCh at maximal concentrations that evoke a ‘peak and plateau’-type response (Fig. 5A) or at threshold concentrations that result in Ca²⁺ oscillations (Fig. 5D). After the removal of [CCh] acinar cells also responded robustly to maximal or threshold concentrations of CCK (Fig. 5A,D). In contrast exposure of cells to YM-264890 completely eliminated the initial CCh-induced increase (Fig. 5B,E), and the cells were then refractory to further stimulation by either CCh or CCK, both at maximal (Fig. 5B and pooled data Fig. 5C) and threshold (Fig. 5E and pooled data Fig. 5F) concentrations of secretagogues.

Figure 5. YM-254890 prevents agonist-induced intracellular calcium release in isolated pancreatic acinar cells.

A, Representative normalized fura-2 fluorescence traces from individual acinar cells in control and in presence of 1 μM CCh or 10 pM CCK. B, Representative traces from individual cells before and after the addition of 1 μM YM-254890. C, Scatter plots of normalized fluorescence imaging data evoked by high concentrations of agonist in presence or absence of 1 μM YM-254890. Maximum normalized fluorescence values were measured and plotted for each condition listed (n = 3 mice, 5–10 cells per mouse). Black symbols represent cells that were exposed only to the indicated agonists. Red symbols represent cells that were initially stimulated by CCh and then subsequently were bathed in YM-254890 for further stimulations by CCh and then CCK. Populations were compared between normalized fluorescence values for agonist in presence and absence of YM-254890. P-values were obtained using a heteroscedastic t test. D, Representative normalized fura-2 fluorescence traces from individual acinar cells in control and in presence of 10 nM CCh or 300 fM CCK. E, Representative traces from individual cells before and after the addition of 1 mM YM-254890. F, Scatter plots of grouped fluorescence imaging data evoked by low concentrations of agonist in presence or absence of 1 μM YM-254890. Black symbols represent cells that were exposed only to agonist. Red symbols represent cells that were initially stimulated by CCh and then subsequently were bathed in YM-254890 for further stimulation by CCh and then CCK. Maximum normalized fluorescence values were measured for each condition listed (n = 3 mice, 5–10 cells per mouse). Data are expressed as mean ± SD. Populations were compared between normalized fluorescence values for agonist in presence and absence of YM-254890. P-values were obtained using a heteroscedastic t test. ****P < 0.0001. YM-254890 completely blocked the ability of individual acinar cells to release intracellular calcium irrespective of whether cells were stimulated with low or high concentrations of CCh or CCK.

Inhibition of G_q/11 by YM-254890 eliminates Ca²⁺ signals in pancreatic acinar cells in vivo

We have previously developed a platform to study Ca²⁺ signalling events in pancreatic acinar cells in vivo after physiological stimulation in a native setting. Briefly mice that express a genetically encoded Ca²⁺ indicator GCaMP6F, specifically in exocrine acinar cells, were produced, allowing the visualization of Ca²⁺ signals using multiphoton microscopy in live mice (Huang et al., 2024; Takano & Yule, 2022, 2023, 2024; Takano et al., 2021; Wahl et al., 2023). We investigated the effects of inhibiting G_q/11 on Ca²⁺ responses evoked by nervous stimulation and CCK administration to simulate physiological stimulation during the cephalic and intestinal phases of GI activity, respectively. A protocol was utilized whereby images of pancreatic acinar cells were obtained after stimulation of the intrinsic nerves to the pancreas and subsequently after injection of CCK before and after incubation of the exposed gland to YM-254890 (Fig. 6A). In the absence of exogenous stimulation small, low-frequency Ca²⁺ signals were observed (Fig. 6B, average projection images visualizing changes in the fluorescence signal during the period of imaging; Fig. 6C, kinetic plots for regions of interest (ROI) within the field of view; and Fig. 6D, pooled data summarizing cells responding, maximum change in amplitude and oscillation frequency). We have reported previously that these signals result from residual circulating CCK even in the fasted state (Takano & Yule, 2023). Subsequent stimulation of intrinsic nerves innervating the pancreas resulted in robust Ca²⁺ signals in a large proportion of the field of view (Fig. 6B, average projection image; Fig. 6C, kinetic plots; Fig. 6E, pooled data). Our previous data confirm that these signals are the result of ACh release from parasympathetic neurons as they are completely attenuated by exposure of the gland to atropine (Takano & Yule, 2023). After the nervous stimulation was terminated, the mice were then injected i.p. with 1 μg/kg of CCK, a concentration deemed physiological because it evokes Ca²⁺ signals with similar spatial and temporal characteristics observed in response to feeding (Takano & Yule, 2023). CCK stimulation again resulted in large Ca²⁺ signals in a significant portion of the acinar cells in the imaging field of view (Fig. 6B, average projection image; Fig. 6C, kinetic plots; and Fig. 6D, pooled data). The exposed gland was then bathed in a solution containing 10 μM YM-264890, and imaging was performed in the continued presence of CCK and subsequently after nervous stimulation. Strikingly Ca²⁺ signals were completely eliminated in response to either stimulus, indicating both ACh- and CCK-stimulated Ca²⁺ elevations result from G_q activation (Fig. 6B, average projection image; Fig. 6C, kinetic plot; Fig. 6D,E, and pooled data).

Figure 6. Nerve stimulation and CCK (cholecystokinin)-induced pancreatic acinar Ca²⁺ signalling require G_q activation in vivo.

A, A schematic of the experimental design. Images were taken as a pair with and without nerve stimulation during imaging. First imaging set (a, b) serves as baseline responses in fasting animals. Second set (c, d) shows CCK-induced responses. Third set (e, f) shows the effect of YM-254890. B, Representative average projection images of acinar cell Ca²⁺ activity during the period of stimulation in each condition indicated as in a–f. Scale bar: 10 μm. C, Representative kinetic plots of Ca²⁺ signals in acinar cells shown in aforementioned images. D, Summary histograms of percentage of acinar cells that exhibited Ca²⁺ signalling, the amplitude of Ca²⁺ transient and the frequencies of Ca²⁺ events at baseline (a), after CCK exposure (c) and after YM-254890 application (e). N = 4 animals, **P < 0.01, ANOVA with Tukey’s test. E, Summary histograms of acinar Ca²⁺ signalling during nerve stimulation at baseline (b), after CCK induction (d) and after YM-254890 application (f). N = 3–4 animals, **P < 0.01, ANOVA with Tukey’s test. Data are expressed as mean ± SD.

Experiments were also performed in animals that were fasted overnight and then subsequently fed by gavage 1 h prior to imaging (Fig. 7A). As we have previously reported (Takano & Yule, 2023, 2024), feeding resulted in robust oscillatory Ca²⁺ signals mediated by circulating CCK (Fig. 7B, average projection image; Fig. 7C, kinetic plots; and Fig. 7D, pooled data). Feeding-induced, together with nerve-induced, elevations in [Ca²⁺] were eliminated by exposure of the gland to 10 μM YM-264890 (Fig. 7B, average projection image; Fig. 7C, kinetic plots; and Fig. 7D,E, pooled data).

Figure 7. Feeding-induced Ca²⁺ signalling requires G_q activation in vivo.

A, A schematic of the experimental design. Images were taken as a pair with and without nerve stimulation during imaging. First imaging set (a, b) serves as baseline responses in animals that were fed before imaging. Second set (c, d) shows the effect of YM-254890. B, Representative average image projections of acinar cell Ca²⁺ signals in each condition during the period of imaging, indicated as in a–d. Scale bar: 10 μm. C, Representative kinetic plots of acinar cell Ca²⁺ signals shown in the aforementioned images. D, Summary histograms of percentage of acinar cells that exhibited Ca²⁺ signalling, amplitude of Ca²⁺ transient and frequencies of Ca²⁺ events at baseline (a) and after YM-254890 application (c). **P < 0.01, N = 4 animals, paired t test. E, Summary histograms of acinar Ca²⁺ signalling during nerve stimulation at baseline (b) and after YM-254890 application (d). N = 4 animals, *P < 0.05, paired t test. Data are expressed as mean ± SD.

ACh and CCK stimulation of fluid secretion and exocytosis requires G_q activation

To confirm that Ca²⁺ signals required for the stimulation of fluid secretion and exocytosis are dependent on G_q activation, we performed a series of experiments to assess the effect of YM-254890 on secretagogue-stimulated enzyme secretion and ion channel activation. Ca²⁺-dependent activation of a Cl⁻ channel, likely TMEM16a, which is localized on the apical plasma membrane (PM) of acinar cells, is essential for fluid secretion from acinar cells (Giovannucci et al., 2002; Park et al., 2001; Petersen, 1992). Ca²⁺-activated Cl⁻ currents were monitored in acutely isolated pancreatic acinar cells by whole-cell patch clamp where the Ca²⁺ in the internal pipette solution was weakly buffered with N-(2-hydroxyethyl) ethylenediamine-N,N^’,N^’-triacetic acid (HEDTA) at 100 nM to allow changes in Ca²⁺ from an appropriate basal value after secretagogue stimulation. Cl⁻ currents were monitored after voltage excursions from a holding potential of −50 mV to more positive potentials. Acinar cells were exposed twice to CCh and then stimulated with CCK. Stimulation resulted in robust Cl currents, which were maintained throughout the period of secretagogue exposure (Fig. 8A, circles, and Fig. 8B,D). Next patch-clamped acinar cells were first treated with CCh to confirm the viability of the cell and then exposed to YM-264890. YM-264890 completely eliminated the activation of Cl⁻ currents after restimulation with either CCh or CCK (Fig. 8A, squares, and Fig. 8C,D).

Figure 8. YM-254890 attenuates whole-cell macroscopic calcium-dependent TMEM16A Cl^– currents induced by CCh and CCK (cholecystokinin).

A, Time course of average Cl⁻ current densities evoked by an initial CCh treatment to establish current magnitudes, followed by a second stimulation by CCh and subsequent CCK application, with and without the addition of 1 μM YM-254890. Black circles represent current densities without YM-254890. Black squares represent current densities before the addition of 1 μM YM-254890, whereas red squares represent current densities in presence of YM-254890 and the indicated agonist. B, Representative Cl⁻ current traces evoked by 1 μM CCh and subsequent 10 pM CCK stimulations when cells were held at −50 mV and stepped from −80 to 120 mV in 20 mV increments (inset shows voltage stepping protocol). C, Representative Cl⁻ current traces before and after the addition of YM-254890 in presence of the agonists indicated. D, Grouped current–voltage relationships elicited by agonists in presence and absence of 1 μM YM-254890 (n = 3 mice, 3 cells per mouse). Data are expressed as mean ± SD. Black symbols represent current densities without YM-254890 evoked by either CCh or CCK. Red solid squares represent current densities before the addition of 1 μM YM-254890, whereas red open symbols represent current densities in presence of YM-254890 and the indicated agonist. TMEM16A currents were completely absent in presence of 1 μM YM-254890 irrespective of whether stimulated by CCh or CCK.

Finally we assessed the G_q/11 dependency of secretagogue-stimulated exocytosis by monitoring the secretion of α-amylase, the enzyme responsible for the initial digestion of carbohydrates. Aliquots of acutely isolated mouse pancreatic acinar cells were incubated with increasing concentrations of either CCh or CCK, and secreted α-amylase was assayed in the bathing medium (Williams et al., 1978). Both CCh and CCK incubation resulted in a characteristic biphasic concentration–secretion relationship. Notably exocytosis of amylase was completely inhibited by YM-264890 at all concentrations of both CCh and CCK (Fig. 9A,B). Moreover amylase activity in all samples treated with YM-264890 and secretagogues was not significantly different from basal secretion measured in the absence of secretagogues (one-way ANOVA, P = 0.144 (Fig. 9A); P = 0.96 (Fig. 9B)). In sum these data demonstrate that M3R and CCK1R couple to G_q/11 to increase in [Ca²⁺], likely as a result of IP₃ formation, and that is required for the physiological function of pancreatic acinar cells.

Figure 9. YM-254890 inhibits stimulated amylase secretion in isolated acinar cells.

Isolated mouse acini were pre-treated with DMSO or 1 μM YM-254890 for 10 min and then stimulated with A, CCK or B, CCh at the indicated concentrations for 30 min. Amylase release into the medium is expressed as a percentage of total cellular amylase. N = 3, data are mean ± SD. ****P < 0.0001 and ***P < 0.001, non-paired t test. # indicates that basal secretion was not significantly different from ± secretagogue + YM 254890 one-way ANOVA, P = 0.144 (CCK (cholecystokinin) + YM vs. control); P = 0.962 (CCh + YM vs. control).

Discussion

Pancreatic acinar cells secrete fluid and enzymes essential for macronutrient digestion after feeding. Extensive research has established that Ca²⁺ signalling plays a crucial role in stimulus–secretion coupling in exocrine acinar cells. Stimulation by the primary secretagogues, ACh and CCK, elicits increases in intracellular [Ca²⁺] with distinct spatial and temporal characteristics that are necessary for physiological fluid and protein secretion (Matozaki et al., 1990; Takano & Yule, 2023). Early studies demonstrated that both secretagogues promote the production of IP₃ and DAG through phosphatidylinositol 4,5-bisphosphate (PIP₂) hydrolysis, downstream of canonical G_q signalling pathway (Matozaki & Williams, 1989; Matozaki et al., 1990; Yule et al., 1999). This classical bifurcating pathway subsequently leads to Ca²⁺ release and PKC activation. Consistent with this mechanism Ca²⁺ has been shown to be essential for fluid secretion, whereas the elevation of Ca²⁺ and PKC activity plays a well-established role in the exocytosis of secretory granules containing zymogens and digestive enzymes. Nevertheless there is evidence that NAADP exclusively produced after CCK stimulation acting on TPC2 (Yamasaki et al., 2004, 2005) provides the only agonist-stimulated signal which satisfies the Ca²⁺-elevating requirement for the action of CCK in acinar cells (Petersen, 2023; Tsvilovskyy et al., 2024). In this mechanistic framework CCK stimulation does not necessitate an obligatory increase in [IP₃], and any PKC activation must therefore either involve an atypical PKC isoform or be mediated by DAG produced through pathways independent of PIP₂ hydrolysis. Although atypical PKCs are expressed in acinar cells (Satoh et al., 2004), the exocytosis of secretory vesicles is induced by phorbol ester exposure, which does not activate atypical PKCs, and in addition DAG is produced after hydrolysis of PIP₂ in response to CCK stimulation (Matozaki & Williams, 1989). Due to the apparent discrepancies in the existing literature, the present study re-evaluated the requirement for increased G_q activity and the consequent increase in [IP₃] and [DAG] in the regulation of pancreatic exocrine signalling and function.

CCK1R and M3R are classical class A ‘rhodopsin-like’ G protein-coupled receptors which have been shown to couple to a variety of G protein classes in addition to G_q. Using a BRET assay in a heterologous expression system, we initially obtained a fingerprint of the G protein family members engaged after receptor activation. Subsequently the specificity of a widely used G_q antagonist YM-254890 to inhibit coupling through this pathway was assessed with a goal of subsequently using the inhibitor to probe G_q involvement in pancreatic signalling and function. This assay of membrane-proximal signalling established that CCK1R and M3R activate a variety of G proteins but importantly that YM-254890 is exquisitely selective at antagonizing only the coupling to G_q/11 after stimulation of either receptor, indicating this agent is a powerful, specific tool to investigate the role of G_q/11 in native cells.

As anticipated in isolated acinar cells, the G_q/11 antagonist completely eliminated Ca²⁺ signalling after activation of the M3R across a range of [CCh] representing threshold to maximal stimulation in vitro. These data support the notion that activation of G_q/11 is both necessary and sufficient for Ca²⁺ signalling events stimulated by ACh. The prevailing model posits that at physiological [CCK], Ca²⁺ signalling is independent of an increase in [IP₃]. However both Ca²⁺ oscillations induced by threshold [CCK] and responses to maximal [CCK] were similarly eliminated by YM-254890. These findings demonstrate that M3R- and CCK1R-mediated signalling requires G_q/11 activation at all concentrations of secretagogue and strongly suggest that an increase in [IP₃] is necessary for eliciting Ca²⁺ signals through both receptors. Because CCK1R stimulation activates a broad array of G proteins, it is tempting to speculate that additional signals may function to define the specific characteristics of the CCK-stimulated Ca²⁺ signal, in addition to activating downstream effectors important for pancreatic physiology (Giovannucci et al., 2000; LeBeau et al., 1999; Yule et al., 1991).

It could be argued that Ca²⁺ signalling events measured in isolated cells may not fully recapitulate physiological function and processes in vivo (Takano & Yule, 2023, 2024). To further substantiate the critical role of G_q/11 signalling, we conducted intravital imaging in live mice, employing stimuli that mimic intrinsic exocrine pancreatic activation across different phases of digestion. Consistent with the in vitro findings, pancreatic exposure to YM-254890 completely eliminated Ca²⁺ signalling events induced by endogenous neural input, systemic CCK administration and, importantly, postprandial stimulation. This latter observation unequivocally establishes that physiological levels of CCK stimulated by feeding require G_q/11 activation to generate Ca²⁺ signals in pancreatic acinar cells. To definitively establish the necessity of G_q/11 activation in physiological pancreatic function, we examined the impact of YM-254890 on secretagogue-induced activation of Ca²⁺-activated Cl⁻ channels (Kidd & Thorn, 2000), serving as a proxy for fluid secretion, and on amylase release as an indicator of exocytosis. Our findings demonstrate that the complete abrogation of both physiological responses upon G_q/11 inhibition underscores its essential role in exocrine pancreatic secretion. Specifically G_q/11 activation downstream of CCK1R and M3R stimulation, leading to elevated IP₃ and DAG levels, is indispensable for both fluid and protein secretion.

Taken together the simplest interpretation of our data is that activation of G_q/11 is necessary and sufficient for signalling and function through both CCK1R and M3R. Nevertheless our data do not strictly exclude a contributory role for NAADP in modulating signalling induced by CCK. For example it is formally possible that production of NAADP is downstream of G_q/11 activation. Moreover given the substantial data which suggests the involvement of NAADP in signals generated through CCK1R, it remains likely that NAADP plays a role in shaping the IP₃-induced signals. Our findings align with a model in which physiological stimulation of M3R and CCK1R leads to G_q/11 activation, subsequent production of IP₃ and DAG and Ca²⁺ release via IP₃Rs, accompanied by enhanced PKC activity. Additionally CCK1R activation induces NAADP generation (Yamasaki et al., 2004, 2005), facilitating Ca²⁺ release through lysosomal TPC2 channels in close proximity to the apical ER. This localized Ca²⁺ release may further sensitize IP₃Rs (Yuan et al., 2024), amplifying and shaping Ca²⁺ signals and thus contributing to the unique spatial and temporal dynamics of CCK-evoked Ca²⁺ responses in pancreatic acinar cells (Calcraft et al., 2009; Cancela et al., 2002; Petersen, 2023; Takano & Yule, 2023; Yule et al., 1991). Future important questions include defining the biochemical pathway responsible for NAADP production and investigating how the spatial relationship between TPC2 and IP₃R channels influences the characteristics of Ca²⁺ signals in physiological and pathophysiological conditions in the exocrine pancreas.

Materials and methods

DNA plasmids and chemicals

The plasmids encoding human M3R (66250) and CCK1R (66236) were from Addgene.

Plasmids encoding D1R, D2R, LPA2R and 5-HT2AR were subcloned from the PRESTO-Tango library (Addgene kit no. 1000000068, provided by Dr Bryan Roth, University of North Carolina, NC) into pcDNA3.1 plasmids removing V2-tail, TEV site and tTA sequences. BRET sensor constructs encoding Gβ1-Venus^156–239 and Gγ 2-Venus^1–155 were provided by Dr Nevin A. Lambert (Augusta University, Augusta, GA). Gα proteins and masGRK3CT-Nluc constructs were provided by Dr Kirill A. Martemyanov (UF Scripps Biomedical Research, Jupiter, FL). YM-254890 was purchased from Tocris, Pittsburgh, U.S.A (7352/1) and dissolved in DMSO at 10 mM, CCh was purchased from Sigma-Aldrich, St Louis, MO, USA (C4382) and CCK was purchased from Thermo Fisher, Waltham, MA, USA (11661).

Cell cultures and transfections

Human embryonic kidney HEK293T/17 cells (RRID:CVCL_1926) were obtained from the American Type Culture Collection (CRL-11268) and grown in Dulbecco’s modified Eagle’s medium (Gibco, Norristown, PA, USA, 10567–014) enriched with 10% fetal bovine serum (FBS, Biowest, Bradenton, FL, USA, 291G23), non-essential amino acids (Gibco, 11140–050), penicillin (100 units/ml), streptomycin (100 μg/ml) (Gibco, 15140–122) and amphotericin B (250 μg/ml) (Thermo-Fisher, 15290–018). Cells were grown at 37°C in a humidified incubator in a 5% CO₂ atmosphere. Cells were regularly checked for mycoplasma contamination. Two million cells were plated in each well of six-well plates in antibiotic-free medium and 10% dialysed FBS (Biowest, 347G18-D) for 4 h, and then transfected with DNA plasmid (2.5 μg) and polyethylenimine (7.5 μl) in the ratio 1:3 (Polysciences, Warrington, PA USA 23966). The following optimized amount of each DNA plasmid was co-transfected: 208 ng of GPCR, 833 ng of Gα protein, 208 ng of Gβ1-Venus^156–239, 208 ng of Gγ 2-Venus^1–155 and 13 ng of masGRK3CT-Nluc; 208 ng of DNA plasmids encoding chaperones Ric8B (for G_s) or Ric8A (for G_q and G₁₁) was also included. The empty vector pcDNA3.1 was used to normalize the total amount of transfected plasmids. After transfection cells were incubated for 16 h before use.

G protein NanoBRET assay

GPCR activation in live cells was measured as a BRET signal between Venus-Gβ1γ 2 and masGRK3CT-Nluc as previously described (Alabdali et al., 2023). The day after transfection cells were briefly washed with PBS and mechanically harvested using a gentle stream of PBS. Cells were collected in 1.5 ml tubes and centrifuged for 5 min at 500 g. The pelleted cells were resuspended in 500 μl of BRET buffer (PBS supplemented with 0.5 mM MgCl₂ and 0.1% glucose), and 25 μl of cells in suspension was transferred to 96-well flat-bottomed white micro-plates (Greiner Bio-One, Monroe, NC, USA). Freshly prepared NanoLuc substrate Hikarazine-103 (Coutant et al., 2019, 2020) diluted 1:250 in BRET buffer was applied to the cells. BRET measurements were obtained using a POLARstar Omega microplate reader (BMG Labtech, Ortenberg, Germany) at room temperature. The BRET signal was calculated as the ratio of light emitted by Gβ1γ 2-Venus (collected using a 535/30 emission filter) to that emitted by masGRK3CT-Nluc (475/30 filter). Plated cells were incubated with YM-254890 or a vehicle and with the NanoLuc substrate. Baseline recordings were performed for 5 min. After 5 min a vehicle or agonists were applied, and recordings were continued for 3 min. In concentration–response experiments 30 μl of cells per well was plated and mixed with NanoLuc substrate and YM-254890 or a vehicle to the indicated final concentrations. Light emission (535/475) was recorded for 5 min before a vehicle or an agonist was added, and recordings continued for another 3 min. The BRET ratio of the last 10 s of baseline recordings was averaged to obtain a baseline value for each condition. This baseline value was subtracted from the maximal BRET ratio obtained after vehicle or agonist application to calculate the BRET ratio. A threshold of 0.01 was decided a priori to define Gα protein activation.

Pancreatic acinar cell isolation

Animals were housed with free access to food and water, unless fasted overnight for some in vivo experiments as stated. Animals were housed at 19–21°C on a 12-h light–dark cycle. Pancreatic acini were obtained from 16-week-old wild-type (C57BL6NTac) male mice. Acinar cells were isolated as previously described (Messenger et al., 2013). The animals were killed by exposure to CO₂ and then by cervical dislocation after unconsciousness. In brief the pancreas was dissected and injected with collagenase (Sigma-Aldrich, C6885), and then incubated in a 37°C water-bath with shaking (180 rpm). Acini were further separated by gentle pipetting, filtering through a 150 μM nylon mesh and layering over a 4% bovine serum albumin (BSA, Alfa Aesar, J65966) solution.

In vitro Ca²⁺ imaging

Acinar cells were loaded with the Ca²⁺ indicator fura-2 AM (2 mM) by incubating in imaging buffer (10 mM HEPES, 1.26 mM Ca²⁺, 137 mM NaCl, 4.7 mM KCl, 5.5 mM glucose, 1 mM Na₂HPO₄ and 0.56 mM MgCl₂, at pH 7.4) for 30 min. Isolated pancreatic acinar cells were seeded onto coverslips and then perfused with imaging buffer and stimulated with the agonist. Ca²⁺ imaging was performed using an inverted epifluorescence Nikon microscope with a 40× oil-immersion objective (Numerical Aperature (NA)=1.3). Loaded cells were excited alternately at 340 and 380 nm using a Poly IV monochromator (TILL Photonics, Grafelfing, Germany), and emission was monitored at 530 nm, as previously described (Arige et al., 2022). Images were captured every 500 ms with an exposure of 10 ms and 4 × 4 binning using a digital camera driven by TILL Photonics software and exported to Microsoft Excel and analysed using Prism (GraphPad, Boston, MA, USA).

In vivo Ca²⁺ imaging

Mist1^CreERT2+/– × GCaMP6f^+/– transgenic mice served as the experimental subjects for Ca²⁺ imaging of the pancreas in vivo. Mist1^CreERT2+/– × GCaMP6f^+/– transgenic mice were produced by crossing GCaMP6f^flox mice (Jackson Laboratory, Jax 028865) with Mist1^CreERT2 mice (Jackson Laboratory, Jax 029228, a gift from Dr Catherine Ovitt, University of Rochester). Tamoxifen (Sigma-Aldrich, T5648, 0.25 mg/g of body weight) was administered to the mice via oral gavage for three consecutive days at least 1 week prior to imaging to excise the loxP sites flanking the STOP codon allowing expression of the Ca²⁺ indicator within the exocrine pancreas. The mice were fasted overnight, and for imaging of feeding-induced activity the mice were gavage fed with 0.3 ml of nutrition gel (Bio-Serv Nutra Gel) 15 min prior to the surgical procedure. The mice were anaesthetized i.p. with an injection of ketamine (60–100 mg) and xylazine (10–20mg), with the degree of anaesthesia confirmed by a toe pinch. The gland was exposed, as described previously (Takano & Yule, 2022; Takano et al., 2021). The immobilized gland was secured within the holder using a cover glass and maintained in Hank’s salt solution (HBSS). Ca²⁺ imaging was conducted in vivo via two-photon microscopy using an Olympus FVMPE-RS system equipped with an Insight X3 pulsed laser (Spectra-Physics) utilizing a heated (OKOLab COL2532) 25× water-immersion lens (Olympus XLPlan NA = 1.05 W MP). GCaMP6f was excited at 950 nm, and emission was recorded between 495 and 540 nm, with images captured at 0.5-s intervals for 3-min durations with or without the presence of 5 Hz electrical stimulation applied to the gland through bipolar electrode wires. After the experiment the animal was killed by an overdose of ketamine/xylazine. Statistical analyses were performed using two-way ANOVA with multiple comparisons, as indicated in the figure legends.

Patch clamp electrophysiology

Acinar cells were allowed to adhere to Cell-Tak (Radnor, PA, USA)-coated glass coverslips for 15 min before experimentation. Coverslips were transferred to a chamber containing an extracellular bath solution (155 mM tetraethylammonium chloride to block K⁺ channels, 2 mM CaCl₂, 1 mM MgCl₂, 10 mM HEPES, pH 7.2). Cl⁻ currents in individual cells were measured in the whole-cell patch clamp configuration using pClamp 9 and an Axopatch 200B amplifier (Molecular Devices). Recordings were sampled at 2 kHz and filtered at 1 kHz. Pipette resistances were 3–5 MΩ, and seal resistances were >1 GΩ. Pipette solutions (pH 7.2) containing 60 mM tetraethylammonium chloride, 90 mM tetraethylammonium glutamate, 10 mM HEPES, 1 mM HEDTA and 20 μM CaCl₂ were used to mimic physiological buffering and basal [Ca²⁺]_i conditions (~100 nM Ca²⁺). Free [Ca²⁺] was estimated using MaxChelator freeware. Agonists were directly perfused onto individual cells using a multibarrel perfusion pipette. Chloride currents after agonist application were monitored with a single voltage step to 80 mV from a holding potential of −80 mV every second until current magnitudes reached a plateau. Current–voltage relationships were obtained in 20 mV incremental steps between −80 and 120 mV from a holding potential of −50 mV.

Amylase secretion assay

Acini were maintained and treated at 37°C in Dulbecco’s modified Eagle’s medium (Thermo Fisher, 21063–029) supplemented with 0.1% BSA and 0.01% soybean trypsin inhibitor (Thermo Fisher, 17075–029). Isolated acini were pre-treated with 0.05% DMSO (Sigma, 276855) or 1 μM YM-254890 (Tocris Biosciences, 568580-02-9) for 10 min, and then incubated for 30 min with CCK-8 (Echelon Biosciences, Salt Lake City, UT, USA 471–47) or carbachol (Sigma-Aldrich, PHR1511) as indicated. Amylase released into the medium was measured using Phadebas Amylase Assay tablets (Magle Life Sciences, Cambridge, MA, USA, 1302) as previously described (Messenger et al., 2013). Amylase secretion was expressed as amylase absorbance units in the medium normalized to total cellular amylase present prior to the start of secretory stimulation.

Supplementary Material

Additional supporting information can be found online in the Supporting Information section at the end of the HTML view of the article. Supporting information files available:

Key points.

• An increase in cytoplasmic Ca²⁺ and PKC activity after CCK and ACh stimulation following feeding is a pivotal event in the activation of fluid secretion and exocytosis from pancreatic acinar cells.

• In contrast to ACh, it has been suggested that at physiological concentrations, CCK stimulation results in the production of nicotinic acid dinucleotide adenine phosphate, without activating the canonical G_q/11 pathway, and the production of inositol 1,4,5,-trisphosphate (IP₃) and diacylglycerol (DAG).

• After having established that YM-254890 is an exquisitely selective G_q/11 inhibitor, we show that Ca²⁺ signals stimulated in vitro and in vivo in response to both M3R and CCK1R stimulation are completely inhibited by YM-254890.

• YM-254890 completely abrogates Ca²⁺-activated Cl⁻ current activation, pivotal for fluid secretion together with amylase secretion stimulated by both M3R and CCK1R activation.

• We conclude that ACh and CCK stimulation results in Gq/11 activation, an increase in IP₃ and DAG, and this event is fundamentally important for exocrine function.

Biography

Takahiro Takano completed his PhD in cell biology at New York Medical College and is currently an assistant professor in the Department of Pharmacology and Physiology at the University of Rochester Medical Center. Since 2019 his research has focused on establishing methods to directly measure Ca²⁺ dynamics in exocrine glands of live animals under physiological conditions utilizing multiphoton microscopy and genetically encoded calcium probes expressed in intact tissue. Luca Franchini has completed his PhD in pharmacological, experimental and clinical sciences at the University of Milan in 2021, focusing on hippocampal synaptic plasticity and NMDA receptor pharmacology. Since then he has been a post-doctoral fellow at the University of Rochester in Cesare Orlandi’s laboratory, investigating orphan G protein-coupled receptor signalling and uncovering their endogenous ligands.

Data availability statement

All original raw data (image files and analysis) are retained on a University of Rochester server, and access is available on request.