|Home | About | Journals | Submit | Contact Us | Français|
Min Seuk Kim, acquisition of data; analysis and interpretation of data
Jeong Hee Hong, acquisition of data; analysis and interpretation of data
Qin Li, acquisition of data; analysis and interpretation of data
Dong Min Shin, drafting of the manuscript
Joel Abramowitz, produced the KO mice
Lutz Birnbaumer, produced the KO mice, drafting of the manuscript
Shmuel Muallem, drafting of the manuscript
Receptor–stimulated Ca2+ influx is a critical component of the Ca2+ signal and mediates all cellular functions regulated by Ca2+. However, excessive Ca2+ influx is highly toxic resulting in cell death, which is the nodal point in all forms of pancreatitis. Ca2+ influx is mediated by store-operated channels (SOCs). The identity and function of the native SOCs in most cells is unknown.
Here, we determine the role of deletion of Trpc3 in mice on Ca2+ signaling, exocytosis, intracellular trypsin activation and pancreatitis.
Deletion of TRPC3 reduced the receptor-stimulated and SOCs-mediated Ca2+ influx by about 50%, indicating that TRPC3 functions as SOC in vivo. The reduced Ca2+ influx in TRPC3−/− acini resulted in reduced frequency of the physiological Ca2+ oscillations and of the pathological sustained [Ca2+]i increase caused by supramaximal stimulation and by the toxins bile acids and palmitoleic acid ethyl ester. Consequently, deletion of TRPC3 shifted the dose response for receptor-stimulated exocytosis, and prevented the pathological inhibition of digestive enzyme secretion at supramaximal agonist concentrations. Accordingly, deletion of TRPC3 markedly reduced intracellular trypsin activation and excessive actin depolymerization in vitro and the severity of pancreatitis in vivo.
These findings establish the native TRPC3 as a SOC in vivo and a role for TRPC3-mediated Ca2+ influx in the pathogenesis of acute pancreatitis and suggest that TRPC3 should be considered a target for prevention of the pancreatic damage in acute pancreatitis.
Acute pancreatitis is an inflammatory, multifactorial disease of the pancreas caused by generation of toxic mediators within the pancreas, resulting in mistargeting of digestive enzymes that eventually destroy the pancreatic parenchyma 1. The pancreatic acinar cells store the harmful digestive enzymes that destroy the pancreas. It is now well established that aberrant Ca2+ signaling perturbs many functions of acinar cells and is intimately associated with all forms and models of acute pancreatitis 2.
The physiological and pathological Ca2+ signal involves IP3-mediated Ca2+ release from the endoplasmic reticulum (ER) that causes the activation of Ca2+ influx channels at the plasma membrane, the so-called store-operated channels (SOCs). At physiological stimulus intensity, the SOCs sustain the receptor-stimulated Ca2+ oscillations and determine their frequency, reload the stores with Ca2+ at the termination of cell stimulation and provide the Ca2+ required for sustaining exocytosis, gene regulation and all long term functions regulated by Ca2+ 3–5. The pathological Ca2+ signal is caused by any stimulus that results in chronic depletion of the ER Ca2+ stores and consequently uncontrolled activation of the SOC channels that causes sustained and prolonged increase in cytosolic Ca2+ ([Ca2+]i) 2, 6.
The complement of the channels mediating the SOC activity in pancreatic acinar and other secretory cells is not known. Acinar cells express the TRPC1, TRPC3 and TRPC6 7 and the newly discovered Orai channels 8. The TRPC channels (TRPCs) function as Ca2+-permeable non-selective cation channels that mediate part of the receptor stimulated Ca2+ influx in many cells 9, 10. The Orai channels are highly selective Ca2+ channels and mediate the Ca2+-release activated Ca2+ (CRAC) current 8. The Orai 11, 12 and TRPC channels 13–15 are regulated by the ER Ca2+ sensor STIM1. STIM1 transmits the ER Ca2+ load to the plasma membrane SOCs to activate them 16, 17. Although STIM1 activates the two channel types by different mechanisms 15,18, the Orai and TRPC channels appear to interact 19 and affect the activity of each other 20, 21. Moreover, interaction of each channel type with STIM1 requires the presence of the other channel and SOC requires the pore function on both channels 22.
Because TRPC3 is prominently expressed in pancreatic and salivary gland acinar cells 7, and TRPC3 is regulated by STIM1 14, we asked whether TRPC3 contributes to the native SOCs in secretory cells and whether TRPC3-mediated Ca2+ influx contributes to aberrant Ca2+ influx responsible for pancreatitis. A definitive approach to these questions is deletion of Trpc3 in mice. Here we report that deletion of TRPC3 markedly reduced SOC activity in multiple secretory cell types, reduced the frequency of Ca2+ oscillations and consequently altered the dose response for agonist-stimulated exocytosis. Most notably, deletion of TRPC3 prevented the pathological inhibition of exocytosis observed at supramaximal receptor stimulation to reduce activation of trypsin within acinar cells and consequently the severity of acute pancreatitis in vivo. These findings establish the native TRPC3 as a SOC in vivo and a role for TRPC3-mediated Ca2+ influx in the pathogenesis of acute pancreatitis and suggest that TRPC3 should be considered a target for prevention of the pancreatic damage in acute pancreatitis.
Trpc3−/− mice were generated as detailed in 23. Acini, ducts and single pancreatic acinar cells were prepared from the pancreas or submandibular gland of WT and Trpc3−/− mice as described previously 24. The acini and cells were maintained in solution A containing (mM) 140 NaCl, 5 KCl, 1 MgCl2, 10 HEPES (pH 7.4), 10 glucose and either 1 CaCl2 or 1 EGTA and adjusted to 310 mOsm. Amylase activity was measured with the Phadebas kit (Pharmacia &Upjohn) following the manufacturer instructions. Intracellular trypsin activity was measured using the synthetic substrate, rhodamine 110-(CBZ-Ile-Pro-Arg)2 25. Acini stimulated for 60 min at 37 °C, resuspended in a solution contaning (mM) 5 HEPES, 150 NaCl, 2 EDTA, pH 7.35, 10 µM substrate and incubated for additional 20 min to allow substrate hydrolysis. Bright-field and fluorescence images were captured and results were analyzed by counting the number of fluorescent cells. For western blots, lysates were prepared as before 24 and proteins (100 µg) were probed with a 1:500 dilution of phospho-PERK and PERK and 1:1000 dilutions of LC3, tubulin or actin.
The whole cell current was measured by dialyzing the cells with a pipette solution containing (mM) 140 Cs-aspartate, 6 MgCl2, 10 BAPTA and 10 HEPES to passively deplete t ER Ca2+. The cells were perfused with a bath solution of (mM) 140 NaC1, 5 CsCl, 1 MgCl2, 10 CaCl2, 10 Hepes, 10 Glucose. After 5 min the bath solution was changed to divalent-free medium (DFM) (mM: 140 NaC1, 5 KCl, 0.5 EGTA, 10 Hepes). Current was recorded by applying 400 ms RAMPs from −100 to +100 mV from a holding potential of 0 mV at 5 sec intervals. The current recorded at −100 mV was used to calculate current density in pA/pF. The current output was filtered at 20 Hz, stored online with a Digi-Data 1200 interface and analyzed offline with pclamp 9.2 software.
Acini in solution A were stimulated with 10 nM CCK8 for 20 min at 37°C, plated on poly-L-lysine coated cover glass, rinsed with PBS and fixed with 4% paraformaldehide for 10 min. The acini were permeabilized with 0.1% Triton X-100 for 5 min and the actin was stained with 0.15 µM FITC-Phalloidin for 20 min. Luminal fluorescence intensity of confocal images was analyzed with MetaMorph.
Acini and ducts loaded with Fura-2 were perfused (37°C) with solution A. [Ca2+]i was measured at the 340 and 380 nm excitation wavelengths and the emitted light was collected by a digital camera with a cutoff filter at 510 nm and analyzed with Metafluor. Results are presented as the 340/380 ratios.
Mice starved over night were injected hourly in the abdominal cavity over 4 hours with caerulein at 40 ng/g body weight. Four injections were used since eight injections induced severe pancreatitis in WT and Trpc3−/− mice. Two hours after the last injection, the mice were sacrificed to excise the pancreas and collect blood. The blood was spun down and the plasma was used to measure serum amylase. The pancreas was immediately embedded in OCT and fixed sections of 5 µm were stained with hematoxylin and eosin. Images were recorded while marinating constant exposure times and resolutions. Damaged area (edema) for each section was determined using MetaMorph. Each image was converted into a monochrome setup to set the same threshold. The damaged area was marked and converted into pixels. At least 5 randomly collected images showing tissue damage were analyzed from each pancreas and the damaged area was calculated as % of the damage in mice injected with saline.
Results are expressed as mean±s.e.m of the indicated number of observations obtained from 3–5 independent experiments and mice. Statistical significance was determined by analysis of variance.
Stimulation of pancreatic acini with high agonist concentration (0.5 mM carbachol) generated a transient increase in Ca2+ with a plateau stabilizing at about 250 nM with a half time of about 65 sec (Fig. 1a). A similar Ca2+ signal was evoked by 10 nM CCK8. Termination of cell stimulation is followed by reloading of the ER with Ca2+. The rate of Ca2+ reloading can be reliably estimated from the recovery of the CCK signal following stimulation with carbachol and inhibition with atropine 26. In wild-type (WT) acini, about 85% of the response to CCK8 was recovered after 90 sec incubation with atropine. Stimulation of WT acini with physiological concentration of 0.5 µM carbachol (not shown) or 10 pM CCK8 (Fig. 1b) induced typical Ca2+ oscillations.
Deletion of TRPC3 has multiple effects on the Ca2+ signal. Although, the initial Ca2+ increase due to Ca2+ release from the ER was not affected by deletion of TRPC3, the subsequent reduction in [Ca2+]i was faster, with a half time of about 20 sec, and [Ca2+]i stabilization at a plateau of about 150 nM (Fig. 1a). Moreover, 90 sec after termination of cell stimulation only 25% of the response to CCK8 was recovered, indicating markedly impaired Ca2+ influx in the Trpc3−/− acini. Finally, deletion of TRPC3 reduced the frequency of the CCK8-evoked Ca2+ oscillation from 8.5 to 3.5 spikes/30 min. Similar results were obtained with carbachol-evoked Ca2+ oscillations.
Reduction in receptor-stimulated Ca2+ influx in the Trpc3−/− acini raised the question of whether TRPC3 contributes to SOC. SOC activity was measured by depleting the stores with 25 µM of the SERCA inhibitor cyclopiazonic acid (CPA) in Ca2+-free media for 7.5 min and then exposing the acini to media containing 2 mM Ca2+. Deletion of TRPC3 reduced SOC activity by about 50% (Fig. 1c), indicating that the native TRPC3 functions as a SOC in pancreatic acini.
To further analyze the role of TRPC3 in SOC we measured the whole cell current under conditions that isolate the Ca2+-release activated Ca2+ (CRAC) current 5. The cells were dialyzed with 10 mM BAPTA to passively deplete the stores and exposed to 10 mM external Ca2+. As with most non-hematopoietic cells, minimal or no current was measured under these conditions. Larger CRAC current can be observed when the cells are then exposed to a divalent-free medium (DFM) 8. Such a maneuver resulted in a modestly inward rectifying current (Fig. 1d,e) that was reduced by about 40% in Trpc3−/− cells (Fig. 1f). These findings further imply a role of TRPC3 in pancreatic acinar cells SOC.
Next, we asked whether TRPC3 functions as SOC and contributes to receptor-stimulated Ca2+ influx in other cell types. We examined receptor- and CPA-induced Ca2+ influx in submandibular gland cells since they display particularly prominent SOC-mediated Ca2+ influx27. Figs. 2a,c (acini) and 2b,d (ducts) show that deletion of TRPC3 reduced agonist-stimulated and SOC-mediated Ca2+ influx Ca2+ in the two cell type. This was particularly prominent in the duct with both activities being reduced by about 70% in TRPC3−/− ducts.
A pathological Ca2+ signal is induced by toxins, such as bile acids and the ethanol metabolites palmitoleic acid ethyl ester (POAEE), that act on pancreatic acinar cells. Reflex of bile acid into the pancreas 28 and ethanol consumption 1 are known causes of pancreatitis. Fig. 3 shows that deletion of TRPC3 has no effect of Ca2+ release by taurocholate (Fig. 3a) and POAEE (Fig. 3b). By contrast, deletion of TRPC3 reduced Ca2+ influx by about 50%. Since bile acids deplete the stores by inhibition of the SERCA pumps 28, these findings further indicate that the native TRPC3 functions as SOC.
Induction of pancreatitis in vitro and in vivo is associated with cell stress, as evident from activation of the cell stress ER kinase PERK 29, which regulates the unfolding protein response. Accordingly, Fig. 4a shows that when the acini are treated with supramaximal CCK8 or with bile acid, PERK phosphorylation is increased. A recent notable finding is that induction of pancreatitis activates autophagy and inhibition of autophagy prevents acute pancreatitis 30. Autophagy can be reliably followed by lipidation of LC3-I to form LC3-II. Fig. 4b shows that treatment with CCK8 and bile acid increased the level of LC3-II. Similar results were obtained by treating the cells with 100 µM POAEE (not shown). Hence, all stressors that increase SOC activity and induce pancreatitis activate the ER stress response and induced autophagy in pancreatic acini.
Significantly, deletion of TRPC3 reduced the rate of PERK phosphorylation (Fig. 4a) and the rate of activation of autophagy (Fig. 4b) in response to supramaximal CCK8 and to bile acids. Hence, the reduced Ca2+ influx in Trpc3−/− cells protected the cells by reducing ER stress and self-destruction by autophagy.
Exocytosis by pancreatic acinar cells is a Ca2+-triggered process. The initial phase of exocytosis is mediated by Ca2+ release from internal stores, whereas Ca2+ influx is essential to maintain exocytosis beyond the first 3–5 min 31. Since deletion of TRPC3 reduced Ca2+ influx and the frequency of Ca2+ oscillations (Fig. 1), we determined the effect of deletion of TRPC3 on exocytosis. Deletion of TRPC3 right shifted the dose response for CCK8-stimulated amylase release (Fig. 5a), as expected from the shift in the frequency of Ca2+ oscillations.
A particularly notable finding in Fig. 5a is the lack of inhibition of exocytosis at high concentrations of CCK8 in the Trpc3−/− acini. The key step in the pathogenesis of pancreatitis is mistargeting of digestive enzymes to the lysosomes via the autophagosomes. In in vitro models of pancreatitis this is manifested as inhibition of exocytosis at supramaximal agonist concentrations 1. Inhibition of exocytosis is due to excessive actin depolymerization at the terminal web that is observed at high, sustained increase in cytosolic Ca2+ 32. Indeed, direct estimation of actin depolymerization in WT and Trpc3−/− cells stimulated with the 10 nM CCK8 showed that deletion of TRPC3 reduced actin depolymerization by about 50% (Fig. 5c). Even more pronounced difference between WT and Trpc3−/− cells was observed on stimulated with 1 nM CCK8 (not shown). These findings suggest that deletion of TRPC3 reduces the severity of pancreatitis in vitro.
Mistargeting of digestive enzymes to the lysosomes results in their activation within the cells that can be followed by measuring intracellular trypsin activity 25. The lack of inhibition of exocytosis at supramaximal agonist concentration in Trpc3−/− cells predicts reduced intracellular trypsin activation in these cells. Fig. 5b shows strong intracellular trypsin activity in 45% of WT acini stimulated with 10 nM CCK8 for 1 hr. Deletion of TRPC3 reduced the number of cells with intracellular trypsin activity to 12% and most of these cells showed mild trypsin activity.
Direct evidence for reduced severity of pancreatitis in the Trpc3−/− mice was obtained with the caerulein model of pancreatitis. The severity of pancreatitis is evaluated by the extent of pancreatic edema and by mistargeting of amylase to the circulation. Figs. 6a and 6b and the summary in Fig. 6c show that deletion of TRPC3 reduced the edema by about 50%. Similarly, serum amylase reported about 50% reduction in the severity of pancreatitis in TRPC3−/− animals (Fig. 6d). Importantly, the reduction in serum amylase and the severity of pancreatitis was not due to reduced level of amylase in the Trpc3−/− pancreas. Total amylase content in WT and Trpc3−/− pancreas was 3.07±0.23 and 2.98± 0.15 mU/µl, respectively.
The present study reports that TRPC3 functions as SOC in vivo to mediate significant portion of the receptor-stimulated Ca2+ influx in exocrine secretory cells; that the TRPC3-mediated Ca2+ influx affects the frequency of Ca2+ oscillations; and that excessive Ca2+ influx by TRPC3 during supramaximal receptor stimulation is toxic to acinar cells and is responsible in part to the cell stress and damage that occur in pancreatitis. Therefore, inhibition of acinar cell TRPC3 and other Ca2+ influx channels may be considered as a modality to control and reduce the severity of pancreatitis.
It is still a matter of debate of whether and which of the TRPCs function as SOCs. Gating of TRPCs by STIM1 13–15 and interaction of TRPCs with Orai1 19–22, 33, 34 provide compelling evidence for the function of TRPCs as SOCs. Moreover, our recent work indicate that the function of both Orai1 and TRPCs is required for the native SOC 22 and the Orai1-TRPC complex mediate both SOC and receptor-stimulated Ca2+ influx 35. Nevertheless, in spite of all these findings, a recent report claims that TRPCs, including TRPC3, do not function as SOCs 36. The current findings add significantly to this topic by showing that the native TRPC3 functions as SOC in vivo. This is particularly important, since the behavior of TRPC3 is affected by its expression level 14, 37. Thus, the native TRPC3 functions as SOC, and as such significantly contributes to receptor-stimulated Ca2+ influx in secretory glands acinar and duct cells. Interestingly, the native TRPC1 functions as SOC in salivary glands acinar and duct cells 38. In model systems TRPC1 and TRPC3 are assembled into a heteromultimer by STIM1 and the function of TRPC3 as SOC requires TRPC1 14. Together, these results suggest that native TRPC1 and TRPC3 may require the activity of each other to function as Ca2+ influx channels.
The contribution of TRPC3 to receptor-stimulated Ca2+ influx is also evident from the reduced frequency of Ca2+ oscillations in Trpc3−/− cells, providing independent evidence that sustained Ca2+ oscillations requires refilling of Ca2+ stores by Ca2+ influx on a spike per spike basis. Through control of the frequency of Ca2+ oscillations TRPC3 determines the exocytotic response of acinar cells. The oscillatory Ca2+ signal that controls exocytosis is initiated by Ca2+ release from the ER at the apical pole 39, 40. The apical pole of pancreatic acini is enriched with all components of the Ca2+ signaling complex 4, including TRPC3 7. Localization of TRPC3 at the apical pole can explain its profound effect on the frequency of the Ca2+ oscillations and exocytosis.
It is now well established that sustained [Ca2+]i increase is responsible for the cell damage occurring in various modes of pancreatitis 6. However, there is no information on the molecular nature of the pathway mediating the sustained [Ca2+]i increase. The present work shows that TRPC3 mediates a significant fraction of the sustained [Ca2+]i increase evoked by the toxins supramaximal stimulation, bile acids and POAEE. In addition, deletion of TRPC3 prevented the pathological inhibition of exocytosis and intracellular trypsin activation in the in vitro model of pancreatitis and reduced the severity of pancreatitis in mice. Hence, it is clear that overactivation of TRPC3 by pathological depletion of ER Ca2+ mediates significant portion of the pathological Ca2+ influx associated with pancreatitis. Complete characterization of all the Ca2+ influx pathways in acinar cells should further clarify the role of Ca2+ influx channels in pancreatitis and other diseases of secretory cells and may provide viable molecular targets for developing drugs to reduce the severity of these diseases.
Grant Support: This work was supported by NIH grants DE12309 and DK38938 and the Ruth S. Harrell Professorship in Medical Research to S.M. and by the Intramural Research Program of the NIH (Z01-ES-101684) to L.B.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.