Expression of Adenosine Receptors in Rodent Pancreas

Adenosine regulates exocrine and endocrine secretions in the pancreas. Adenosine is considered to play a role in acini-to-duct signaling in the exocrine pancreas. To identify the molecular basis of functional adenosine receptors in the exocrine pancreas, immunohistochemical analysis was performed in the rat, mouse, and guinea pig pancreas, and the secretory rate and concentration of HCO3− in pancreatic juice from the rat pancreas were measured. The A2A adenosine receptor colocalized with ezrin, an A-kinase anchoring protein, in the luminal membrane of duct cells in the mouse and guinea pig pancreas. However, a strong signal ascribed to A2B adenosine receptors was detected in insulin-positive β cells in islets of Langerhans. The A2A adenosine receptor agonist 4-[2-[[6-Amino-9-(N-ethyl-β-D-ribofuranuronamidosyl)-9H-purin-2-yl]amino]ethyl]benzenepropanoic acid (CGS 21680) stimulated HCO3−-rich fluid secretion from the rat pancreas. These results indicate that A2A adenosine receptors may be, at least in part, involved in the exocrine secretion of pancreatic duct cells via acini-to-duct signaling. The adenosine receptors may be a potential therapeutic target for cancer as well as exocrine dysfunctions of the pancreas.

A 1 , A 2A , A 2B , and A 3 adenosine receptors are expressed in the rat and mouse pancreas [8,14,15]. Real-time PCR revealed that the rank order of the adenosine mRNA level was A 2A > A 2B ≥ A 3 >> A 1 , where the level of A 2A was at least five times higher than of other receptors in the rat pancreas [8]. Immunohistochemical studies showed that A 2A and A 2B adenosine receptors were localized in the luminal membranes of rat duct cells and human duct cell lines [8,9]. Additionally, A 2A adenosine receptors were detected in the rat islets, most likely in β cells [8]. A 1 and A 2A adenosine receptors were also found in isolated α-cells from mouse islets, and 4-[2-[[6-Amino-9-(N-ethyl-β-D-ribofuranuronamidosyl)-9H-purin-2-yl]amino]ethyl]benzenepropanoic acid (CGS 21680), an A 2A adenosine receptor agonist, stimulated glucagon secretion [16]. Protein expression of A 1 adenosine receptors was identified in α-cells from the human pancreas [17]. However, the expression and function of adenosine receptors in primary human pancreatic duct cells are unknown.
A 2A and A 2B adenosine receptors generally increase cAMP levels, whereas A 1 and A 3 receptors decrease them [18]. Thus, the present study focused on functional A 2A and A 2B adenosine receptors involved in the ductal secretion of the pancreas. The results reveal that mouse and guinea pig duct cells express A 2A adenosine receptors. However, β cells in the islets express both A 2A and A 2B adenosine receptors. Furthermore, it was demonstrated that A 2A adenosine receptors contribute to exocrine secretion in the rat pancreas.

A 2A and A 2B Receptors Expressed in Pancreatic Islets of the Rat
A previous study showed that A 2A adenosine receptors were expressed in the islets of the rat pancreas [8]. In order to determine the cells that expressed A 2A and A 2B adenosine receptors, triple staining of adenosine receptors was performed with insulin and glucagon on paraffin sections of the rat pancreas. The immunofluorescence of A 2A adenosine receptors was detected in insulin-positive β cells [19] ( Figure 1A-C), but in few glucagon-positive α-cells ( Figure 1D). Additionally, a strong signal ascribed to A 2B adenosine receptors was detected specifically in insulin-positive β cells ( Figure 1E-H). Immunofluorescence was diminished with normal rabbit IgG for the isotype control ( Figure 1I-L).

A 2A and A 2B Receptors Expressed in Pancreatic Islets of the Mouse
The immunolocalization of the A 2A and A 2B adenosine receptors was examined with paraffin sections of the mouse pancreas. Immunofluorescence ascribed to the A 2A adenosine receptors was localized on the inside of cells (Figure 2A). The A 2A adenosine receptors were expressed in insulin-positive β cells and glucagon-positive α-cells ( Figure 2B-D). However, a strong signal ascribed to A 2B adenosine receptors was detected specifically in insulin-positive β cells ( Figure 2E-H). The results were consistent with the distribution of A 2B adenosine receptors in the rat pancreas. Immunofluorescence was diminished with normal rabbit IgG for the isotype control ( Figure 2I-L).

A 2A Receptors Expressed in the Pancreatic Ducts of Mice
Our previous study showed that both A 2A and A 2B adenosine receptors localized in the luminal membranes of rat duct cells [9]. In order to distinguish between pancreatic ducts and blood vessels, triple staining of adenosine receptors was performed with ezrin and PECAM-1 (platelet endothelial cell adhesion molecule-1) on paraffin sections of the mouse pancreas, as reported previously in the rat and guinea pig [4,9]. Immunofluorescence ascribed to the A 2A adenosine receptor was colocalized with ezrin, an A-kinase anchoring protein, to the luminal membrane of the pancreatic duct ( Figure 3A-C). However, immunofluorescence ascribed to the A 2B adenosine receptor was weak in mouse ducts ( Figure 3E-G). Additionally, the signal for A 2A and A 2B adenosine receptors was detected in PECAM-1-positive endothelial cells of blood vessels ( Figure 3D,H). Immunofluorescence was diminished with normal rabbit IgG for the isotype control ( Figure 3I-L).

A 2A Receptors Expressed in the Pancreatic Ducts of Guinea Pigs
We performed immunostaining of A 2A and A 2B adenosine receptors with ezrin and PECAM-1 on paraffin sections of the guinea pig. Immunofluorescence ascribed to the A 2A adenosine receptor was colocalized with ezrin to the luminal membrane of the pancreatic duct ( Figure 4A-C). However, immunofluorescence ascribed to the A 2B adenosine receptor was weak in guinea pig ducts ( Figure 4 E-G). The signal for A 2B adenosine receptors was detected in the PECAM-1-positive endothelial cells of blood vessels ( Figure 4H). Immunofluorescence was diminished with normal rabbit IgG for the isotype control ( Figure 4I-L).

Expression of A 2A and A 2B Receptor Protein in Rat Pancreatic Ducts
We next performed Western blot analysis to examine the expression of A 2A and A 2B adenosine receptors in the rat pancreatic ducts. We detected A 2A (ADORA2A, 54 kDa) and A 2B (ADORA2B, 55 kDa) adenosine receptors in the lysates of the isolated ducts ( Figure 5; n = 2 rats). In addition, the A 2A and A 2B adenosine receptors were detected in the lysates of Capan-1, which is a human pancreas adenocarcinoma cell line, but not in HEK293 cells (n = 2).

A 2A Receptor Agonist Elicited Pancreatic Secretion in Rats
In order to demonstrate whether adenosine regulated exocrine secretion, the secretory rate and concentration of protein and HCO 3 − in pancreatic juice from the rat pancreas were measured. Specific adenosine receptor agonists were tested to identify functional adenosine receptors. The intravenous injection of CGS 21680 (20 nmol/kg body weight), an A 2A adenosine receptor agonist, significantly increased the secretory rate from 0.40 ± 0.05 in the control to 0.72 ± 0.09 µL/min after 20 min and sustained it for 20 min ( Figure 6A; n = 6 rats). The concentration of protein in pancreatic juice was decreased from 77.7 ± 8.4 to 41.2 ± 5.5 g/L after 40 min, indicating ductal secretion ( Figure 6B). In addition, the HCO 3 − concentration was increased from 38.2 ± 3.1 to 52.7 ± 5.6 mM after 40 min, indicating HCO 3 − -rich ductal secretion ( Figure 6C).

Effect of Adenosine Receptor Antagonists on Pancreatic Secretion in Rats
Cholecystokinin stimulates the release of ATP and ectonucleosides from acini into pancreatic juice [7]. Adenosine is produced by the hydrolysis of ATP in the ductal lumen. In order to demonstrate whether luminal adenosine regulated adenosine receptors, specific antagonists were used. The moderate concentration of cholecystokinin (CCK, 0.1 nmol/kg body weight) increased the secretory rate and protein concentration, as reported previously [21] ( Figure 7A

Expression of ADORA2B Associated with Poor Prognosis of Pancreatic Adenocarcinoma
In contrast to the rat pancreas, the A 2B adenosine receptors showed the highest mRNA levels among four subtypes in human pancreatic adenocarcinoma cell lines [8]. To determine whether the expression of adenosine receptors was associated with a poor prognosis of pancreatic adenocarcinoma patients, we conducted statistical analysis using The Cancer Genome Atlas (TCGA) database. Analysis of RNA-sequencing data of pancreatic adenocarcinoma (TCGA, Provisional) revealed that a low expression of ADORA2A, which encodes the A 2A adenosine receptor, was associated with poor overall survival and disease-free survival ( Figure 8A,B; log-rank P = 0.0302 and P = 0.0104, respectively). In contrast, high expression of ADORA2B (A 2B adenosine receptor) was associated with poor disease-free survival ( Figure 8D; log-rank P = 0.0125). However, the expression of ADORA1 and ADORA3 was not associated with the prognosis of pancreatic adenocarcinoma patients (log-rank P = 0.310 and P = 0.322, respectively).

Discussion
In the present study, it was demonstrated that the A 2A adenosine receptor contributed to exocrine secretion in the rodent pancreas. This conclusion was based on the following major results: the A 2A adenosine receptor agonist stimulated a HCO 3 − -rich fluid secretion from the rat pancreas ( Figure 6A-C); the A 2A adenosine receptor colocalized with ezrin in the luminal membrane of duct cells in the mouse and guinea pig pancreas (Figures 3 and 4), as reported in rats previously. The conclusion is consistent with a pharmacological study in the dog pancreas [10].

Compartmentalization of cAMP Signaling in the Luminal Regions of Pancreatic Duct Cells
In pancreatic ducts, adenosine is produced by the hydrolysis of ATP that is secreted from acini by CCK stimulation [6]. Adenosine binds to adenosine receptors on the luminal membrane of duct cells and stimulates ductal secretion [22]. In the present study, A 2A and A 2B adenosine receptor antagonists showed a tendency to suppress 25% of the secretory rate by CCK stimulation (Figure 7), indicating acini-to-duct signaling. A 2A and A 2B adenosine receptors primarily signal via G s proteins, resulting in the activation of adenylyl cyclase, an increase in cAMP production, activation of a membrane-associated isoform of protein kinase A (type II PKA), and subsequent activation of cAMP-activated Cl − channels (CFTR) [23][24][25]. Previous studies reported that ezrin forms the scaffold for type II PKA and cAMP signaling compartments, including the A 2B adenosine receptor, adenylyl cyclase, and CFTR [26][27][28]. The membrane-associated adenylyl cyclase isoforms (AC3, AC4, AC6, AC7, and AC9) were found to be expressed in the mouse pancreas [29]. Furthermore, AC6 was shown to play a critical role in cAMP/PKA-mediated signaling in pancreatic exocrine cells. The AC6 physically and functionally associates with CFTR at the apical surface of intestinal epithelial cells [30]. A recent study reported that the stimulation of A 2A adenosine receptors activates AC6, which is bound to A-kinase-anchoring protein (AKAP79/150), to synthesize cAMP that is used by PKA and phosphodiesterase 3A (PDE3A) in hepatocytes [31]. Further studies are required to clarify whether A 2A adenosine receptors associate with AC6 and lead to the compartmentalization of cAMP signaling in the luminal regions of pancreatic duct cells.

Role of A 2A and A 2B Adenosine Receptors in Pancreas
In accordance with the present results, previous studies demonstrated that A 2A adenosine receptors regulated anion secretion in the gerbil middle ear epithelium [32], rabbit distal bright convoluted tubule cell line [33], and mouse colonic epithelia [34]. In addition to epithelial transport, the A 2A adenosine receptor is known to be involved in the endocrine pancreas [13]. In the present study, A 2A and A 2B adenosine receptors were detected in insulin-positive β cells (Figures 1 and 2). A previous study demonstrated that A 2A adenosine receptors stimulated insulin secretion on mouse islets [14]. However, the A 2B adenosine receptor antagonists were shown to increase plasma insulin levels in rats [35]. Future studies are required to clarify the intracellular signaling in pancreatic β cells.

Composition of Adenosine Receptor Subtype in Duct Cells
Several limitations of this study should be acknowledged. The protein level of A 2A and A 2B adenosine receptors could not be quantified by immunohistochemical analysis. The effects of CGS 21680 and BAY 60-6583 were small compared to those of secretin in in vivo experiments ( Figure 6). Only the maximum dose of the chemical was tested. In the present study, ductal secretion was stimulated significantly more by CGS 21680 than BAY 60-6583 at the same concentration, indicating that A 2A adenosine receptors were dominant in rat duct cells, whereas BAY 60-6583 showed a tendency to promote HCO 3 − -rich fluid secretion ( Figure 6E,F). Thus, we cannot rule out the existence of A 2B adenosine receptors in pancreatic duct cells. Previous studies postulated that A 2A and A 2B adenosine receptors form functional heterooligomers [36]. The affinities of the A 2A adenosine receptors for adenosine or CGS 21680 were shown to be decreased on co-expression with A 2B adenosine receptors in recombinant cells [37]. Meanwhile, the affinity of A 2B adenosine receptors for BAY 60-6583 was not altered by co-expression with A 2A adenosine receptors. Future studies are needed in order to establish the presence of the heterooligomer in pancreatic duct cells and its functional relevance.

A Potential Therapeutic Target for Pancreatic Cancer
In contrast to the present results, we demonstrated that adenosine regulated anion secretion via A 2B adenosine receptors in the luminal membrane of Capan-1, which is a human pancreas adenocarcinoma cell line [9]. The A 2B adenosine receptors had the highest mRNA level among four subtypes in human pancreatic adenocarcinoma cell lines (PANC-1 and CFPAC-1) [8]. In silico analysis of TCGA data showed that a high mRNA expression of ADORA2B was associated with a poor disease-free survival of pancreatic adenocarcinoma patients ( Figure 8D). However, a low expression of ADORA2A was associated with poor prognosis (Figure 8A,B). Similarly, the expression of A 2B adenosine receptors was highest in prostate cancer cells and bladder urothelial carcinoma [38,39]. Moreover, mRNA and the protein expression of A 2B adenosine receptors were consistently upregulated in bladder urothelial and colorectal carcinoma compared with normal tissues [39,40]. The high expression of A 2B adenosine receptors was associated with a poor prognosis in bladder urothelial carcinoma and breast cancer patients [39,41]. The expression of A 2B adenosine receptors was increased in colon and breast cancer cells by hypoxia, suggesting a potential therapeutic target for cancer [40,42]. Indeed, selective A 2B adenosine receptor antagonists inhibited the proliferation of prostate, colon, and breast cancer cells [38,40,43]. The blockade of A 2B adenosine receptors was shown to increase the sensitivity of mouse GL261 glioma cells to the chemotherapeutic drug temozolomide [44]. In addition, the selective antagonist improved the survival of mice bearing metastatic breast tumors [41]. Future studies are needed in order to verify the presence of A 2B adenosine receptors in primary pancreatic tumors and confirm their pathophysiological functions related to proliferation, metastasis, and chemoresistance in pancreatic cancer.

Western Immunoblotting
Male Sprague-Dawley rats were sacrificed by cervical dislocation. Pancreatic ducts were isolated by enzymatic digestion and microdissection from the pancreas, as previously described [1].  Table 1). The reaction was visualized with a secondary antibody labeled with alkaline phosphatase (Promega, Madison, WI, USA).

In Vivo Collection of Pancreatic Secretion
Male Sprague-Dawley rats were left overnight without food, but with access to water. The rats were anesthetized with isoflurane, pentobarbital (25 mg/kg body weight i.p.), and ethyl carbamate (1 g/kg b.w.) to maintain a stable depth of anesthesia without awareness episodes. The animals were placed on a heating table to maintain body temperature. The surgical technique used to collect pancreatic juice from the rat with a slight modification is described elsewhere [20,45]. The abdomen was opened by a midline incision. The common pancreatic bile duct was cannulated, and pancreatic juice was collected in a silicone tube (CP-N 0.5×1; Shin-Etsu Polymer, Tokyo, Japan). In order to collect pancreatic juice, free of bile, the proximal end of the bile duct was ligated and cut. For intravenous injections, a polyethylene catheter was placed in one of the external jugular veins. Sample volumes were determined by the length of pancreatic juice in the silicone tube. Samples were collected into 1 mL of saline, and the protein concentration was measured at OD280 with a spectrophotometer (BioPhotometer; Eppendorf, Hamburg, Germany). The samples were equilibrated with 5% CO 2 overnight and pH values were measured with a pH meter (B-71X; Hobiba, Kyoto, Japan) at 37 • C in a CO 2 incubator. The HCO 3 − concentration was calculated with the Henderson-Hasselbalch equation and determined with the calibration curve method.

In Silico Analysis of TCGA Data
Statistical analysis was performed on data from The Cancer Genome Atlas (TCGA) database using the online cBioPortal for Cancer Genomics platform (http://www.cbioportal.org) [46,47]. A total of 179 cases were selected, corresponding to mRNA data (RNA Seq V2 RSEM) from Pancreatic adenocarcinoma patients (TCGA, Provisional). A cohort of pancreatic adenocarcinoma patients was divided into those with high (Z-score >1.5) and low expressions of mRNA. Kaplan-Meier analysis with the log-rank test was performed to estimate the overall survival and disease/progression-free survival.

Statistics
Means ± SEM of the number n of experiments are shown. Significance of the differences was analyzed by the one-way analysis of variance, with p < 0.05 indicating significance. Data were analyzed in Igor or Microsoft Excel.

Conclusions
The results showed that A 2A adenosine receptors may be, at least in part, involved in the exocrine secretion of pancreatic duct cells via acini-to-duct signaling. The A 2A and A 2B adenosine receptors may be a potential therapeutic target for cancer as well as exocrine dysfunctions of the pancreas.

Conflicts of Interest:
The author declares no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.