Cannabinoid agonist WIN55,212 in vitro inhibits interleukin-6 (IL-6) and monocyte chemo-attractant protein-1 (MCP-1) release by rat pancreatic acini and in vivo induces dual effects on the course of acute pancreatitis
Department of Physiology and Pharmacology, Sapienza University of Rome, Rome, Italy
Address for Correspondence Prof. Maria Broccardo, Department of Physiology and Pharmacology, ‘V. Erspamer’– Sapienza University of Rome, P.le A. Moro, 5-00185 Rome, Italy. Tel: +39 0649912492; fax: +39 0649912487; e-mail: email@example.com
Background Cannabinoids (CBs) evoke their effects by activating the cannabinoid receptor subtypes CB1-r and CB2-r and exert anti-inflammatory effects altering chemokine and cytokine expression. Various cytokines and chemokines are produced and released by rodent pancreatic acini in acute pancreatitis. Although CB1-r and CB2-r expressed in rat exocrine pancreatic acinar cells do not modulate digestive enzyme release, whether they modulate inflammatory mediators remains unclear. We investigated the CB-r system role on exocrine pancreas in unstimulated conditions and during acute pancreatitis. Methods We evaluated in vitro and in vivo changes induced by WIN55,212 on the inflammatory variables amylasemia, pancreatic edema and morphology, and on acinar release and content of the cytokine interleukin-6 (IL-6) and chemokine monocyte chemo-attractant protein-1 (MCP-1) in untreated rats and rats with caerulein (CK)-induced pancreatitis. Key Results In the in vitro experiments, WIN55,212 (10−6 mol L−1) inhibited IL-6 and MCP-1 release from acinar cells of unstimulated rats and after CK-induced pancreatitis. In vivo, when rats were pretreated with WIN55,212 (2 mg kg−1, intraperitoneally) before experimentally-induced pancreatitis, serum amylase, pancreatic edema and IL-6 and MCP-1 acinar content diminished and pancreatic morphology improved. Conversely, when rats with experimentally-induced pancreatitis were post-treated with WIN55,212, pancreatitis worsened. Conclusions & Inferences These findings provide new evidence showing that the pancreatic CB1-r/CB2-r system modulates pro-inflammatory factor levels in rat exocrine pancreatic acinar cells. The dual, time-dependent WIN55,212-induced changes in the development and course of acute pancreatitis support the idea that the role of the endogenous CB receptor system differs according to the local inflammatory status.
Cannabinoids (CBs) are a group of compounds that were discovered in marijuana and hashish, herbal drugs derived from the female plant Cannabis sativa,1 and used to cure various medical conditions, including gastrointestinal (GI) disorders.2 Cannabinoids evoke their effects mainly by activating two specific receptor subclasses (CB-r). Cannabinoid 1 receptors (CB1-r) are widely distributed throughout the central, peripheral and enteric nervous systems in animals and in humans,3–5 and are expressed also in the immune system.6 Cannabinoid 2 receptors (CB2-r) are mainly expressed on immune cells in the periphery and modulate immune function,3–5 but are also present in the brain,7 myocardium and cardiomyoblasts,8,9 endothelial cells of various origins,10,11 and GI tract.12
A previous research showed that mice treated with tetrahydrocannabinol (THC), the principal active ingredient in marijuana, and stimulated with lipopolysaccharide (LPS) or polyinosinic–polycytidylic acid (polyI : C) produced decreased levels of type I interferons (IFN-α and IFN-β) providing the first evidence suggesting that CBs might modulate cytokine production.13,14 Cannabinoids also exert anti-inflammatory actions in various inflammatory diseases ranging from inflammatory pain (collagen-induced arthritis in mice,15 carrageenan-induced paw edema in rats 16), GI inflammatory disorders,17 myocardial infarction, stroke and atherosclerosis.18 By activating CB1-r/CB2-r, CBs may act directly on immune cell types, but they can also alter cytokine and chemokine expression, leading to a cross-signal among immune cells and playing a critical role in anti-inflammatory VS pro-inflammatory activities.19–21
Whether and how CBs alter the course of experimentally-induced acute pancreatitis remains controversial.22–25 Clinical reports describe acute pancreatitis secondary to THC abuse, even if the exact mechanism underlying the progression of acute pancreatitis in humans remains unknown.26 Acute pancreatitis is a condition characterized by an inappropriate digestive enzyme activation within pancreatic acinar cells27 that induces pancreatic damage leading first to local inflammation,28–30 and subsequently, to a release of cytokines and chemokines 31 including tumor necrosis factor α (TNF-α),32 multigenic obesity-1 (Mob-1),33–35 interleukin-6 (IL-6) and monocyte chemo-attractant protein-1 (MCP-1).35
In a recent study, we provided new evidence that although CB1-r and CB2-r are expressed in rat exocrine pancreatic lobules and acinar cells, CBs inhibit only vagal stimulated amylase secretion from pancreatic lobules.36 Hence the precise role of the CB-r system in pancreatic acinar cells remains unclear. We also reported that after caerulein (CK)-induced pancreatitis, CB1-r expression in rat acinar cells remained unchanged whereas CB2-r were down-regulated. More information is needed on a possible relationship between the exocrine pancreatic acinar CB-r system and inflammatory mediators. Understanding this relationship is of importance in distinguishing the controversial pancreatic effects of cannabis use and abuse in humans, possibly identifying a new therapeutic target for managing acute pancreatitis.
We investigated the influence of the rat exocrine pancreatic acinar cell CB-r system on cytokines and chemokines mediating the local inflammatory responses. To do so we evaluated in vitro and in vivo changes induced by the synthetic non-selective CB-r agonist, WIN55,212, on the inflammatory variables amylasemia, pancreatic edema and morphology, and on acinar cell IL-6 and MCP-1 release and content in unstimulated rats and during CK-induced pancreatitis.
Materials and methods
Adult male Wistar rats (weighing 180–200 g) (Harlan Laboratories, S.r.l. Udine, Italy) were housed under conventional conditions with free access to animal chow and water. All rats were fasted and had free access to water for 24 h before each experiment. The study was conducted according to the guidelines of the Italian Ministry of University and Research (D.L.116, 27/01/92) and the European Communities Council Directive (86/609/EEC). All possible efforts were made to minimize the number of animals used and their discomfort.
Caerulein (Bachem, Germany) was dissolved in saline. WIN55,212 (Tocris, United Kingdom), a non-selective CB1-r/CB2-r agonist, for in vitro studies was dissolved in 0.025 N HCl solution and then diluted in saline, and for in vivo studies was dissolved in vehicle (TWEEN 80 : PEG : saline 5 : 5 : 90, and then diluted in saline, as required). AM251 (a selective CB1-r antagonist) and AM630 (a selective CB2-r antagonist) (Tocris, United Kingdom) were both dissolved in 100% ethanol and then diluted in saline when necessary.
Acute CK-induced pancreatitis
Acute mild pancreatitis was induced in male rats as previously described 37 by injecting CK intraperitoneally (i.p.) three times at the dose of 10 μg kg−1 with a 1-h interval between injections. Healthy control rats were i.p. injected with saline. One hour after the last injection, rats were euthanized with CO2, blood samples were collected for amylase determination, the pancreas was rapidly removed and samples were resected for use as isolated acini preparations, edema evaluation and morphological examination. Dispersed acini were then processed for functional assay and enzyme-linked immunosorbent assay (ELISA).
In vitro studies
Rat-isolated pancreatic acini preparation Isolated pancreatic acini were prepared by collagenase digestion as described by Peikin et al.38 In brief, the pancreas was injected with 5 mL of digestive solution (standard buffer containing collagenase 0.2 mg mL−1) and digested twice for 15 min at 37 °C in a Dubnoff shaking bath (Orlando Valentini, Milan, Italy; 120 oscillations min−1). After being hand shaken, fragments of tissue were removed and acini were dispersed by trituration with pipettes of decreasing diameter and washed twice with a standard solution containing 2% w/v albumin and 2 mmol L−1 CaCl2. Acini were put into 25 mL of standard solution and centrifuged for 10 s at 150 g (at 4 °C). Krebs–Ringer-HEPES buffer (pH 7.4) was used as standard solution, containing (in mmol L−1) 103 NaCl, 8 KCl, 1.2 KH2PO4, 2 glutamine, 5 glucose, 25 HEPES, 1.3 CaCl2, 0.6 MgSO4 with 1% v/v amino acid supplement, 1% w/v bovine albumin and ethylenediaminetetraacetic acid (EDTA)-free protease inhibitor cocktail (Roche Diagnostic, Milan, Italy) (1 tablet per 50 mL of extract solution). Acini were suspended in 100 mL of standard solution and preincubated for 30 min at 37 °C in a Dubnoff shaking bath.
For the functional assays, 2 mL aliquots (about 5 × 104 cell mL−1) were put into a 25 mL flask containing the substances to be tested. Pancreatic acini were incubated with WIN55,212 at a concentration of 10−6 mol L−1 for 3 h at 37 °C in a Dubnoff shaking bath, and preincubated with AM251 (10−6 mol L−1) or AM630 (10−6 mol L−1), 10 min before the agonist.
When the experiment ended, pancreatic acini were collected into tubes and centrifuged for 4 min at 1500 g. Supernatants were rapidly frozen at −18 °C until used for ELISA.
In vivo studies
Effect of WIN55,212 on the course of acute pancreatitis For pretreatment experimental protocol, rats were randomized into six experimental groups (10 in each group) (Fig. 1A). The first group was injected i.p. with vehicle or saline or both (control group). The second group received CK (10 μg kg−1, i.p.) three times hourly, preceded by injection of vehicle. The third group (pretreated rats) received WIN55,212 (2 mg kg−1, i.p.) 40 min before the first CK dose and the fourth group received WIN55,212 followed by saline injected three times once an hour. The fifth group received AM630 (2 mg kg−1, i.p.) 70 min before the first CK dose. The sixth group received AM630 70 min and WIN55,212 40 min before the first CK dose. The first experimental series ended 1 h after the last CK injection.
For post-treatment experimental protocol, rats were randomly assigned to one of six groups (10 rats each) (Fig. 1B). The first group was injected i.p. with vehicle or saline or both (control group); the second group were injected with CK (10 μg kg−1, i.p.) three times hourly, followed by an injection of vehicle; the third group (post-treated rats) received WIN55,212 (2 mg kg−1, i.p.) 1 h after the last CK injection and the fourth group received WIN55,212 preceded by saline injected three times once an hour. The fifth group received AM251 (2 mg kg−1, i.p.) 30 min after the last CK dose. The sixth group received CK, AM251 30 min and WIN55,212 1 h after the last CK injection. The second experimental series ended 1 h after WIN55,212 was injected.
When both series ended rats were killed by CO2 inhalation and samples of blood were collected for serum amylase determination. The abdominal cavity was opened, and the pancreas was removed for histological examination, edema evaluation and acinar cell preparation. IL-6 and MCP-1 content was evaluated in acinar cells diluted in PBS and mechanically homogenized in a Teflon-glass homogenizer (Kartell, Binasco, Italy).
Pancreatic water content (edema) Pancreatic samples were weighed and then dried for 48–72 h at 60 °C and weighed again to determine pancreatic water content. The results were calculated as (wet weight – dry weight)/wet weight and are expressed as a percentage.39
Amylase activity For the quantitative determinations of serum amylase, a colorimetric test (Phadebas® Amylase Test; Magle AB, Lund, Sweden) was used.40 The serum amylase level was expressed as U 100 mL−1 of serum.
IL-6 and MCP-1 ELISA Pancreatic acinar cell supernatants (for the in vitro test) or contents (for the in vivo test) were assayed for IL-6 and MCP-1 using ELISA, according to the manufacturer’s instructions (Pierce biotechnology, Inc, Rockford, IL, USA). In the in vitro test, IL-6 and MCP-1 supernatant levels were expressed as pg μg−1protein. Protein content has been measured in each sample of acinar cell homogenate, by BCA Protein Assay (Pierce biotechnology, Inc). In the in vivo test, IL-6 and MCP-1 acinar contents were expressed as pg mL−1 of a solution containing about 5 × 104 cell mL−1.
BCA Protein Assay BCA Protein Assay is a detergent-compatible formulation based on bicinchoninic (BCA) for the colorimetric and quantitation of total protein.
Morphological examination For conventional light microscopy pancreatic, specimens were fixed in 10% neutral phosphate-buffered formalin, dehydrated and embedded in paraffin. 5 μm thick sections were cut and stained with haematoxylin-eosin and periodic acid-Schiff.
Other pancreatic specimens (small samples) were fixed in a 2.5% purified glutaraldehyde solution in 0.1 mol L−1 cacodylate buffer at pH 7.3 and postfixed in 1.33% osmium tetroxide. Tissue was dehydrated in increasing concentrations of ethanol and embedded in epoxy resin (Araldite). Semi-thin sections (0.980 μm thick) were stained with azure II/methylene blue.
Light microscopy images were captured with a camera equipped with a Zeiss AxioSkop 40 lens, using MRGrab 1.0 software (Carl Zeiss, American Laboratory Trading Groton, CT, USA). Images obtained from semi-thin sections were elaborated using Photoshop® Elements® 2.0 software (Adobe Systems, San Josè, CA, USA). Original magnification ×40.
Statistic All values are expressed as means ± SEM. Data were analyzed with the Statistic Software Package (anova and post hoc Student’s t-test). P-values <0.05 were considered to indicate statistical significance.
In vitro studies
Effect of WIN55,212 on IL-6 and MCP-1 release from pancreatic acini of rat in unstimulated conditions and after CK-induced pancreatitis Rat pancreatic acini released both pro-inflammatory factors studied, IL-6 and MCP-1 (Fig. 2). WIN55,212, at the concentration of 10−6 mol L−1, significantly reduces IL-6 and MCP-1 amounts in the supernatant, in unstimulated condition (IL-6 by 46% and MCP-1 by 33%) (Fig. 2A, B) and after CK-induced pancreatitis (IL-6 by 53% and MCP-1 by 55%) (Fig. 2C, D).
Treatment with AM251 and AM630 alone, at the lowest concentration (10−6 mol L−1) used to maintain the molar ratio agonist : antagonist 1 : 1, increased IL-6 and MCP-1 release (data not shown), confirming that both antagonists show an intrinsic inverse agonist activity 41,42 that prevents the possibility to characterize the exact site of action of WIN55,212 on this function.
In vivo studies
Effect of WIN55,212 on serum amylase, pancreatic water content, IL-6 and MCP-1 acinar content in rats in unstimulated conditions and during acute pancreatitis WIN55,212 (2 mg kg−1, i.p.) injected before and after three saline injections left serum amylase, pancreatic water content (Table 1), and IL-6 and MCP-1 acinar content all unchanged compared with values in control rats (Fig. 3A–D).
Table 1. Serum amylase level and pancreatic water content of rats in unstimulated condition, after CK-induced pancreatitis and in WIN55,212 pretreatment and post-treatment protocols
Serum amylase (U 100 mL−1)
Pancreatic water content (% of wet weight)
*P < 0.01 VS saline (10 rats for each), †P < 0.01 VS CK-injected group (10 rats for each), ‡P < 0.01 VS WIN55,212 + CK-injected group (10 rats for each), §P < 0.01 VS CK + WIN55,212-injected group (10 rats for each).
394 ± 92.8
67 ± 2.8
WIN55,212 (2 mg kg−1, i.p.) + saline
428 ± 93.9
66 ± 1.2
Saline + WIN55,212
400 ± 78.6
64 ± 3.2
CK (10 μg kg−1, i.p.)
1900 ± 50.2*
76 ± 1.2*
WIN55,212 + CK
840 ± 80.3†
68 ± 3.3†
AM630 (2 mg kg−1, i.p.) + vehicle + CK
1800 ± 80.4*
78 ± 1.2*
AM630 + WIN55,212 + CK
1408 ± 86†‡
72 ± 1.5†‡
CK + WIN55,212
3990 ± 119.2†
84 ± 1.2†
CK + AM251 (2 mg kg−1, i.p.) + vehicle
1860 ± 100.1*
75 ± 1.5*
CK + AM251 + WIN55,212
2100 ± 90†§
79 ± 1.7†§
When rats were injected i.p. three times with CK (10 μg kg−1, i.p.), the secretagogue increased significantly serum amylase (about 5 fold), pancreatic water content (by 13%) (Table 1) and MCP-1 and IL-6 levels (Fig. 3A–D) compared with values in unstimulated control rats, confirming that CK induced acute mild pancreatitis.
In rats pretreated with WIN55,212 (2 mg kg−1, i.p.) (40 min before inducing pancreatitis) all the tested pancreatic variables decreased significantly: serum amylase (−56%), pancreatic water content (−18%); and pancreatic acinar content (IL-6, 23% and MCP-1, 30%) compared with values in CK-treated rats (Fig. 3A, B). Treatment with AM630 (30 min before agonist), that injected alone did not modify CK-induced pancreatitis, significantly reversed the protective effect of WIN55,212 on the development of acute pancreatitis (Table 1 and Fig. 3A, B).
Conversely, when rats were post-treated with WIN55,212 (2 mg kg−1, i.p.) (1 h after induced pancreatitis) all the tested pancreatic variables increased: serum amylase by about 2.1-fold, pancreatic water content by 10% (Table 1) and pancreatic acinar IL-6 content by 27% and MCP-1 content by 35% from values in CK-treated rats (Fig. 3C, D). AM251, that injected alone 30 min before WIN55,212 did not affect pancreatitis, partially but significantly reversed the deleterious effect of the CB agonist on the course of pancreatitis (Table 1 and Fig. 3C, D).
Morphological examination The morphological exami-nation in semi-thin sections of pancreatic acinar cells from saline-treated control rats (Fig. 4A) showed vesicular nuclei with prominent nucleoli and normal, fair chromatin. Their cytoplasm contained numerous zymogen granules. Conversely, sections from CK-treated rats (Fig. 4B) showed distinct ultrastructural changes: the pancreatic parenchyma showed interacinar spaces with evident edema and vascular congestion. In many pancreatic cells the cytoplasm contained apoptotic bodies with emarginated chromatin that sometimes made them hard to distinguish from cells in the surrounding cytoplasm. These pancreatic acinar cells contained fewer zymogen granules, numerous intracytoplasmic microvacuoles and some macrovacuoles. In sections from WIN55,212 pretreated rats (Fig. 4C), no apoptotic bodies were visible in the pancreatic parenchyma and cytoplasmic zymogen granules were only slightly reduced in number. Cytoplasmatic micro-vacuolization were also slightly reduced and macro-vacuolization absent. Some nuclei contained emarginated chromatin. In contrast, in sections from WIN55,212 post-treated rats (Fig. 4D), light microscopy identified apoptotic bodies and a notable reduction in cytoplasmic zymogen granules. They also showed numerous intracytoplasmic macrovacuoles. Acinar cell nuclei had emarginated chromatin. The intracellular spaces showed pathological changes, edema and vascular congestion.
By evaluating in vitro and in vivo experiments we provide new evidence supporting the modulatory role of the CB system on exocrine pancreas in unstimulated rats and during CK-induced mild acute pancreatitis.
The distinctive finding from our in vitro experiments is that by activating pancreatic CB receptors, expressed on rat acinar cells but unable to control digestive enzyme secretion,36 the synthetic non-selective CB-r agonist WIN55,212 decreases release of the two pro-inflammatory factors studied, IL-6 and MCP-1, in the unstimulated condition and after CK-stimulated pancreatitis. Our in vitro data, confirming previous works reporting the anti-inflammatory activity of CBs in several models of acute inflammation,19–21 extend this role to the exocrine pancreas. Interesting and recent studies have shown that resident pancreatic cells, like acinar cells, have an important role in leukocyte attraction via chemokine and cytokine expression and secretion. Hence, besides being a site for premature activation of proteases in the pancreas, acinar cells have a dynamic, pathophysiological role in the development of pancreatitis.43
Our in vitro experiments showing unexpected IL-6 and MCP-1 pancreatic acinar cell secretion in unstimulated conditions, strongly support a previous report that cytokine and chemokine production by isolated pancreatic acinar cells may depend partly on stress related to isolation and culture.35 Conversely, increased pancreatic pro-inflammatory factor mRNA levels after CK-induced pancreatitis in rats confirm that cytokines and chemokines play a critical role in the development of this pathological condition.32
Precisely how WIN55,212 reduced cytokine and chemokine secretion under unstimulated and pathological conditions remains unclear. In our previous study using immunofluorescence and Western blot analysis, we found that rat pancreatic acinar cells express CB1-r and CB2-r and that in rats with mild CK-induced pancreatitis CB2-r were down-regulated whereas CB1-r expression under both conditions remained unchanged.36 That cells co-express two receptor subtypes controlling the same function is not surprising,6 given that in mast cells expressing CB1-r and CB2-r, CB1-r activation reduces inflammatory mediator secretion, whereas CB2-r-mediated signaling pathways inhibit their nuclear transcription and synthesis.6 Collectively these in vitro findings suggest that WIN55,212 probably reduces cytokine and chemokine release under unstimulated and pathological conditions by activating the unmodified CB1-r population, known to inhibit the secretion of inflammatory mediators. Further studies with CB1-r/CB2-r selective agonists and antagonists without intrinsic activity will be necessary to certainly characterize the receptor subtype involved in the control of these actions.
The findings from our in vivo experiments show that WIN55,212 injected peripherally in the rat has a dual protective or deleterious effect on CK-induced mild acute pancreatitis, depending on when the CB agonist is injected. When WIN55,212 is injected before CK-induced pancreatitis, it improves pancreatitis, and does so by preventing morphological pancreatic damage, pancreatic edema, hyperamylasemia and the increased IL-6 and MCP-1 acinar content. Evidence that WIN55,212-pretreatment protected rats against CK-induced pancreatitis came also from our ultrastructural findings showing less severe pancreatic acinar cell damage in WIN55,212-pretreated rats than in CK-treated rats.
Various pathophysiological mechanisms might explain how WIN55,212 reduced serum amylase in pretreated rats. In our experimental model of rat CK-induced pancreatitis, CK activates vagal pathways,44 first increasing then gradually decreasing and ultimately blocking pancreatic amylase release, thereby activating intracellular zymogen, thus provoking cell damage, digestive amylase outflow in the blood and, finally, pancreatitis.45 Convincing observations show that the peripheral CB system inhibits cholinergic excitatory transmission through prejunctional receptors located on myenteric cholinergic neurons.46,47 Hence we conjecture that pretreatment with the CB agonist could at least in part reduce the vagal CK-induced amylase increase by inhibiting cholinergic transmission (anticholinergic activity).
When we investigated the cytokines and chemokines mediating the local inflammatory responses to acute pancreatitis, we found that in pretreated rats WIN55,212 decreased IL-6 and MCP-1 acinar cell content, thus preventing acute CK-induced pancreatitis from developing. In the pretreatment experimental protocol, when WIN55,212 is injected, CB2-r, whose signaling pathways inhibit protein synthesis, are fully expressed on acinar cells and available for activation.36 We therefore suggest that WIN55,212 injected before experimentally-induced acute pancreatitis exerts its protective effect mainly by activating the CB2-r transduction signaling pathway. This claim is supported by showing that WIN55,212 protective effect is significantly reversed by AM630, a selective CB2-r antagonist. In vitro studies show that CK-induced pancreatitis rapidly and strongly activates pancreatic acinar cell nuclear factor-kappa B (NF-κB),48–50 a factor known to play a major role in regulating the expression of many inflammatory mediators, cytokine induction, growth factors and adhesion molecules.51 Ample evidence now shows that inhibiting NF-κB activation induces beneficial effects on experimental pancreatitis 52,53 and that esocannabinoids and endocannabinoids induce their anti-inflammatory effects by inhibiting NF-κB activation and NF-κB-dependent pathways, in various inflammatory models.54–56 Collectively, these findings suggest that WIN55,212 protects against rat CK-induced pancreatitis by activating the CB2-r transduction signaling pathway, thereby inhibiting NF-κB activation, as reported in different models of tissue injury.57,58 Our findings may open the way for using WIN55,212 to prevent local and systemic inflammatory effects in patients at risk of acute pancreatitis.
When we injected WIN55,212 after CK-induced pancreatitis, it had the opposite effect: namely it adversely affected the severity of the disease, and all the tested biochemical, histological and ultrastructural variables related to acute pancreatitis worsened. In rats with CK-induced pancreatitis post-treated with WIN55,212, serum amylase levels probably increased because WIN55,212 directly inhibited pancreatic amylase secretion.36 In our previous study we showed that WIN55,212 decreases amylase secretion from pancreatic lobules only under vagal stimulation,36 a condition existing after CK-induced pancreatitis. Thus, WIN55,212 injected after CK-induced pancreatitis by causing pancreatic acinar cells to retain and prematurely activate zymogens, could aggravate CK-induced pancreatic damage and ultimately increase amylase outflow, (hyperamylasemia), even though we cannot exclude other precipitating events.59,60 This biological mechanism receives support from our in vitro observation indicating that, after CK-induced pancreatitis, WIN55,212 increases amylase outflow from isolated pancreatic lobules indirectly, by worsening the pancreatic damage arising when presynaptic CB-r activation inhibits secretion (personal communication).
In rats post-treated with WIN55,212 after CK-induced pancreatitis, MCP-1 and IL-6 acinar content increases further. Because this experimental condition downregulates pancreatic acinar cell CB2-r,36 the lack of a CB2-r mediated negative signal on pancreatic synthesis of inflammatory mediators could imply that WIN55,212 increases MCP-1 and IL-6 acinar content thus aggravating acute pancreatitis. Accordingly, a previous study shows that the CB-r antagonists injected after inducing acute pancreatitis prolong rat survival thus confirming CB system involvement in worsening acute pancreatitis.22
Conversely, our in vivo results argue against a previous report from Dembinski et al.23 showing that pretreatment with the natural, endogenous CB agonist anandamide before inducing acute pancreatitis worsens inflammation, whereas post-treatment improves the disease course. The discrepancy between the two studies could depend on differences in experimental design including dose and number of CK injections needed to induce pancreatitis, and dose and type of CB agonist used. Discrepancies could also arise because the effects induced by synthetic CB agonists sometimes differ from those induced by the natural ligand.61–63
An explanation on the opposing results of the pre- and post-treatment with the CB agonist on pancreatitis course could be provided by the recent evidence of opposing roles of the CB1 and CB2 receptors in some forms of tissue injury.64 One can speculate therefore that WIN55,212 in the initial acute phase of pancreatitis exerts its beneficial effects through CB2-r activation, whilst during pancreatitis it worsens inflammatory response by promoting oxidative stress, inflammation and tissue injury through CB1-r (CB2-r being down-regulated). This hypothesis is confirmed as AM251, a selective CB1-r antagonist, reversed the hurtful effect of WIN55,212 post-treatment. On the other hand, we can not exclude that WIN55,212 pro-inflammatory effect during pancreatitis is due to CB1-r and CB2-r expressed on macrophages recruited by the inflammatory response.65,66
In conclusion, our in vitro study shows that under unstimulated conditions and after CK-induced pancreatitis, CB-r and predominantly CB1-r, expressed in rat isolated pancreatic acinar cells inhibit release of the pro-inflammatory factors IL-6 and MCP-1.
Preventive CB-r system stimulation before CK-induced pancreatitis through in vivo peripherally-injected WIN55,212 activates protective mechanisms thus reducing some variables reflecting pancreatic damage, whereas stimulating the CB-r system in vivo after inducing mild pancreatitis aggravates the severity of disease. The dual time-dependent effect of WIN55,212 on the course of CK-induced pancreatitis seems related to an endocannabinoid system mechanism predominantly involving changes in CB receptor levels depending on the inflammatory status in the exocrine pancreas. The bidirectional role of the CB system in regulating experimental-induced pancreatitis we describe here in rats, gives new information on the heretofore contrasting interactions between cannabis use/abuse and pancreas in humans, thus offering a new therapeutic target for managing acute pancreatitis.
The authors wish to thank Claudio Munari for his technical assistance.