Nicotinic stimulation of splenic T cells is protective in endoscopic retrograde cholangiopancreatography-induced acute pancreatitis in mice

Abstract

It has previously been shown that current smoking is protective against endoscopic retrograde cholangiopancreatography (ERCP)-induced acute pancreatitis, but the mechanism of this effect was not identified. We tested the hypothesis that nicotine is the active factor in this protection in a mouse model of ERCP. Pretreatment with nicotine dose dependently inhibited acute pancreatitis caused by infusion of ERCP contrast solution into the main pancreatic duct in mice. 3-2,4-Dimethoxybenzylidene anabaseine (GTS-21), a specific partial agonist of the α7 nicotinic cholinergic receptor (α7nAChR), also protected the pancreas against ERCP-induced acute pancreatitis. The effects of GTS-21 were abolished by pretreatment with the nicotinic receptor antagonist mecamylamine. Surgical splenectomy performed 7 days before ERCP-induced pancreatitis blocked the protective effects of GTS-21. Intravenous injection of a crude preparation of total splenocytes prepared from mice pretreated with GTS-21 inhibited ERCP-induced pancreatitis; splenocytes from mice treated with vehicle had no effect. When T cells were removed from the crude GTS-21-treated splenocyte preparation by immunomagnetic separation, the remaining non-T-cell splenocytes did not protect against ERCP-induced acute pancreatitis. We conclude that nicotine protects against ERCP-induced acute pancreatitis and that splenic T cells are required for this effect. Stimulation of α7 nicotinic cholinergic receptors may protect against ERCP-induced acute pancreatitis and may also be a novel approach to therapeutic reversal of ongoing acute pancreatitis.

NEW & NOTEWORTHY Epidemiological evidence indicated that acute smoking reduced the risk of endoscopic retrograde cholangiopancreatography (ERCP)-induced pancreatitis, but the mechanism has remained elusive. The current findings indicate the nicotine reduces the severity of ERCP-induced pancreatitis by stimulating a population of splenic T cells that exert a protective effect on the pancreas. These findings raise the possibility that nicotinic agonists might be useful in treating pancreatitis.

INTRODUCTION

Smoking has been shown to be a risk factor for both acute and chronic pancreatitis (1, 2). Tobacco smoke contains nicotine plus ∼4,000 other chemicals (2), but there is scant evidence concerning the component of tobacco smoke that is responsible for increasing the risk of developing pancreatitis. For example, there is no evidence that nicotine causes pancreatic inflammation in people or in animal models. Neither nicotine nor any of its metabolites is able to reproduce the spectrum of damaging effects (such as edema, necrosis, increased serum amylase levels, and elevated pancreatic myeloperoxidase concentrations) typically seen in pancreatic inflammation in the commonly used animal models of acute pancreatitis. Studies of the effects of nicotine on the pancreas have shown that pancreatic acinar cells express nicotinic cholinergic receptors, but nicotine does not provoke inflammatory responses in these cells similar to those observed in response to bona fide inflammatory agents such as high-dose cholecystokinin (or its analog, caerulein) or ethanol plus free fatty acids (3, 4). In addition, it has been shown that cigarette smoke extract, but not nicotine, enhances secretagogue-induced pancreatic acinar cell necrosis (5).

In contrast, several studies indicate that nicotine may have a protective effect against acute pancreatitis both in people and in experimental animal models. For example, DiMagno et al. (6, 7) showed that current smoking is protective against the pancreatitis caused by endoscopic retrograde cholangiopancreatography (ERCP). It is not known if this protective effect of current smoking is due to nicotine or some other component of tobacco smoke. However, it is possible that the pancreatic protective effect of current smoking on ERCP-induced pancreatitis is due to the effects of nicotine via the cholinergic anti-inflammatory reflex first described by Tracey and coworkers (8). In this reflex, inflammation of an organ, caused in this case by ERCP-induced damage to the pancreas, is detected by vagal sensory nerves, resulting in increased vagal motor neuron stimulation releasing acetylcholine in or near the spleen to activate cholinergic nicotinic receptors, which subsequently inhibit further inflammation (8). In this reflex, nicotine has been shown to activate the α7 subtype of cholinergic nicotinic receptors (α7nAChRs) (9). There is also strong evidence that the cholinergic anti-inflammatory reflex plays a major role in such diverse inflammatory conditions as sepsis, rheumatoid arthritis, inflammatory bowel disease, obesity, type II diabetes, and kidney inflammation (8).

Evidence in favor of a role for the cholinergic anti-inflammatory reflex in acute pancreatitis came from a study of caerulein-induced acute pancreatitis in mice (9). It was shown that vagotomy or pretreatment with a nicotinic cholinergic antagonist drug caused enhanced severity of pancreatitis, whereas pretreatment with a specific α7nAChR agonist drug significantly decreased the severity of acute pancreatitis. These results were strongly supportive of the concept that the cholinergic anti-inflammatory reflex can protect the pancreas from inflammation. In this scenario, local pancreatic damage initiates a vagally mediated nicotinic protective mechanism, evidenced by the worsening of inflammation if either vagotomy is performed or a nicotinic antagonistic drug is given; thus, this reflex is strengthened by vagal stimulation or nicotinic agonism. Further evidence for this mechanism came from a study showing that treatment with either nicotine or cholinesterase inhibitory drugs (to prolong the action of endogenous acetylcholine) protected rats against acute pancreatitis (10). Also, it has been shown that nicotine administration dose dependently inhibits acute pancreatitis in a model of human gallstone pancreatitis consisting of injection of a bile acid retrogradely into the bile-pancreatic duct both in mice (11) and in rats (12). It was also demonstrated that nicotine increased the number and suppressive capacity of CD4⁺CD25⁺ regulatory T cells (Tregs) in the systemic circulation, suggesting that these cells were mediating nicotine-induced pancreatic inflammation in bile acid-induced acute pancreatitis (11).

Taken together, these experimental results in animal models of acute pancreatitis suggest the hypothesis that nicotine may be the active factor in tobacco smoke that is protective against ERCP-induced acute pancreatitis in people and may also be protective against other causes of acute pancreatitis. This is an attractive concept because there are currently no effective treatments available for pancreatitis once it is established in patients. Simply applying a transdermal nicotine patch or administering a nicotinic agonist drug would be simple and cheap therapies, if effective. With this in mind, we describe studies here designed to reveal the mechanism of the protective effect of nicotine in a mouse model of ERCP-induced acute pancreatitis. We have obtained evidence in support of the hypothesis that nicotine, via α7nAChRs, acts on T cells in the spleen to protect the pancreas against acute pancreatitis. Knowledge of the mechanism of the protective effect of nicotine in this model will allow determination of the relevance of the experimental mouse results to human acute pancreatitis.

MATERIALS AND METHODS

Materials

Nicotine hydrogen tartrate, GTS-21 (3-2,4-dimethoxybenzylidene anabaseine; DMXB-A), mecamylamine, and human myeloperoxidase (MPO) were obtained from Sigma-Aldrich (St. Louis, MO).

Animals

Male C57BL/6J mice 6–8 wk old (Jackson Labs, Bar Harbor, ME) were used. Mice were housed in a 12:12-h light-dark cycle and given water and chow ad libitum. All studies were approved by the Duke University Animal Care and Use Committee.

ERCP Surgery

Surgery was performed as previously described (13–15) with few modifications. Mice were anesthetized by intraperitoneal injection of a mixture of 87.5% ketamine-12.5% xylazine, and a midline laparotomy was used to expose the first portion of the duodenum. A puncture wound was made in the antimesenteric surface of the duodenum opposite the ampulla of Vater, and a 30-gauge catheter was attached with tubing to an infusion pump that was passed through the puncture wound and then into the common bile duct via the ampulla of Vater. The infusion catheter was secured in the distal common bile duct distal to the entrance of the pancreatic duct with a ligature, and the bile duct near the liver was occluded with a bulldog clamp to prevent backflow into the gallbladder and liver. The infusate consisted of ERCP contrast medium [Omnipaque (iohexol) 300 mg/mL, GE Healthcare, Chicago, IL] pumped into the pancreatic duct using a high-performance PHD ultra-syringe I/W programmable pump with pressure transducer APT300 Hg from Harvard Apparatus (Holliston, MA) that was set to a range of 0–200 mmHg. Intraductal pressure was monitored continuously during injection using the software dedicated to the instrument. Contrast medium was injected at a rate of 80 µL/min for 5 min as previously described (16). Methylene blue (1%) was included in the infusate to allow identification of leakage from the duct lumen. After 10 min, the catheter, ligature, and bulldog clamp were removed, and the duodenotomy was closed using a purse-string suture. The laparotomy was closed in two layers, and analgesia was achieved by subcutaneous injection of buprenorphine hydrochloride at a dose of 50 mg/kg. The animals were given free access to food and water upon recovery. The mice were euthanized 24 h after surgery by CO₂ asphyxiation and then were weighed. The pancreas was removed and weighed; a portion was frozen at −80°C for later myeloperoxidase (MPO) assay, and a separate portion was fixed overnight at 4°C in 10% formalin for the histopathological analysis. Mixed arteriovenous blood was also collected by decapitation for serum amylase measurement.

Caerulein Hyperstimulation

Caerulein was injected intraperitoneally at 6-h intervals at doses of 50 µg/kg, and the mice were euthanized by CO₂ asphyxiation 1 h after the last injection for collection of blood and pancreatic tissue.

Splenectomy

Seven days before ERCP surgery, mice were anesthetized with isoflurane, a small incision was made in the left flank, and the splenic vasculature was then ligated and the spleen removed through the incision as previously described (17). Sham-operated controls were treated identically but without splenic artery ligation and spleen removal.

One group of mice was pretreated 30 min before ERCP surgery with 0.1, 0.5, or 1.0 mg/kg nicotine hydrogen tartrate intraperitoneally (equivalent to 32.5, 162.5, and 325 µg/kg of pure nicotine, respectively) (18). In some groups, mice were pretreated 30 min before ERCP surgery with 4 mg/kg GTS-21 intraperitoneally. In another group, mice were administered 1 mg/kg mecamylamine intraperitoneally (30 min before GTS-21 and again 12 h later). Some mice were injected intraperitoneally with 4 mg/kg GTS-21 or saline (controls), as described, 24 h before they were euthanized and their spleens removed for in vitro splenocyte preparation. Crude splenocytes or T-cell-depleted splenocytes were injected intravenously into intact recipient mice via the retro-orbital sinus under isoflurane anesthesia 24 h before ERCP surgery.

Splenocytes

After removal, spleens from mice pretreated 24 h previously with either 4 mg/kg GTS-21 or vehicle were mashed through a 70-µm filter using a 3-mL syringe plunger with a rubber tip, incubated with DNase-I for 10 min at 37°C, treated to lyse red blood cells, and washed by centrifugation in PBS three times as previously described (17). These crude splenocytes were either injected intravenously into untreated recipient mice or used to prepare a suspension of splenocytes lacking T cells. In this preparation, crude splenocytes were incubated with a biotinylated anti-CD3ε antibody (eBioscience; 2.5 µL/mL) for 30 min at 4°C. After the cells were washed by centrifugation, they were then incubated with 25 µL/mL of MACS streptavidin microbeads (Miltenyi Biotec; Somerville, MA) for 15 min at 4°C. The cells were washed again by centrifugation and then passed through negative selection LD columns attached to a MACS multistand magnetic separator (Miltenyi Biotec, Somerville, MA) at unit gravity. After washing was completed, the purified mouse splenocytes lacking T cells, or crude splenocytes from mice treated with the GTS-21 vehicle, were resuspended in sterile PBS, and then, 100 µL of the cell suspensions containing 20 million cells was injected intravenously into the retro-orbital sinus in intact mice under isoflurane anesthesia. An aliquot of the cells was separately analyzed by flow cytometry to assess the extent of the removal of T cells.

Assays

The serum amylase concentration was measured as previously described elsewhere (14), except that Phadebas amylase test tablets (Magle Life Sciences, Cambridge, MA) were used as substrate instead of Procion Yellow starch.

Pancreatic tissue activity of MPO, an enzyme produced by neutrophils and used as a marker of inflammation associated with neutrophil infiltration, was measured as previously described elsewhere using the substrate tetramethylbenzidine (14). The pancreatic total protein was measured using micro-BCA kits (Thermo Scientific, Rockford, IL).

Portions of pancreas were fixed overnight in phosphate-buffered 10% formalin. The tissue was then embedded in paraffin, sectioned at 5 µm, stained with hematoxylin-eosin, and coded for examination by two investigators blinded to the experimental design. The severity of pancreatitis was graded using a modified scoring criterion previously described elsewhere (14).

Statistics

Results are expressed as means ± SE. To control for day-to-day variations, all results were normalized as percentage of the mean value obtained in mice treated with elevated intraductal pressure alone in individual experiments. Mean differences among groups were examined by one-way analysis of variance followed by the Tukey–Kramer posttest (GraphPad Prism, ver. 8.03; GraphPad Software, San Diego, CA). P < 0.05 was considered statistically significant.

RESULTS

Our initial goal was to evaluate the potential role of the cholinergic anti-inflammatory reflex in the prevention of the acute pancreatitis that develops in 3.5% of all patients after ERCP and up to 50% of high-risk ERCPs (7). We previously studied the role of intrapancreatic duct pressure on acute pancreatitis in the mouse by infusing buffered isotonic saline into the main pancreatic duct after tying off the common bile duct to prevent backflow into the gallbladder and liver (16). We found that increasing intrapancreatic duct pressure from 7–11 mmHg to 25–33 mmHg induced acute pancreatitis via activation of the mechanoreceptor Piezo1 and that selective knockout of Piezo1 in pancreatic acinar cells prevented acute pancreatitis caused by elevated pressure (16). In the current study, we wanted more closely to mimic human ERCP conditions, so we infused the radiocontrast solution used in ERCP at the same volume and rate of infusion used previously. We found that this procedure caused reproducible increases in several indices of acute pancreatitis: increased pancreatic edema, elevated plasma amylase levels, increased pancreatic myeloperoxidase (MPO) concentrations (a marker of neutrophil influx into the pancreas), and histopathology (Fig. 1).

Figure 1.

Nicotine dose dependently inhibits endoscopic retrograde cholangiopancreatography (ERCP)-induced pancreatitis [pancreatic edema, serum amylase, pancreatic myeloperoxidase (MPO) levels, and histopathology]. Results are normalized to percentage of the mean responses to pressure alone. A–D: effects of nicotine on pressure-induced pancreatic edema, plasma amylase, pancreatic MPO, and pancreatic histology score. E: effects of nicotine on pressure-induced pancreatic histology. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.000. Scale bar = 1 mm.

When the mice were pretreated 30 min before ERCP surgery with intraperitoneal injections of nicotine hydrogen tartrate, there was a significant nicotine dose-dependent inhibition of ERCP-induced pancreatic edema, plasma amylase elevation, increased pancreatic MPO concentrations, and histopathology (Fig. 1). This is consistent with the mechanism of the cholinergic anti-inflammatory reflex seen previously in other systems (8). The effects of nicotine in the cholinergic anti-inflammatory reflex are mediated by α7nAChRs, and we also found that the inflammatory effects of ERCP in our model were inhibited by a specific α7nAChR agonist drug, GTS-21 (Fig. 2). To confirm that the effects of GTS-21 were specific to nicotinic receptors and not some other nonspecific effect, we also showed that the nicotinic receptor antagonist drug mecamylamine blocked the protective effects of GTS-21 (Fig. 2).

Figure 2.

The α7nAChR agonist GTS-21 inhibits endoscopic retrograde cholangiopancreatography (ERCP) pressure-induced pancreatic edema, serum amylase, pancreatic myeloperoxidase (MPO) levels, and histopathology, and this effect is blocked by the nicotinic antagonist mecamylamine. The results are normalized to percentage of the mean responses to pressure alone. A–D: effects of GTS-21 and mecamylamine on pressure-induced pancreatic edema, plasma amylase, pancreatic MPO, and pancreatic histology score. E: effects of GTS-21 and mecamylamine on pressure-induced pancreatic histology. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. Scale bar = 1 mm.

We evaluated the role of the spleen in these effects. We found that prior splenectomy alone has no effect on ERCP-induced acute pancreatitis, but prior splenectomy blocked the protective effects of GTS-21 on ERCP-induced pancreatic inflammation (Fig. 3). Based on this finding, we tested the hypothesis that the adoptive transfer of splenocytes from GTS-21-treated mice could prevent ERCP-induced acute pancreatitis in intact, normal mice. We indeed found that splenocytes prepared from mice treated 24 h previously with GTS-21 exhibited significantly less ERCP-induced inflammation than splenocytes prepared from control mice treated with the GTS-21 vehicle (Fig. 4), suggesting that a cell type (or types) that is stimulated by α7nAChRs mediates the splenic protection of the pancreas in this model. We hypothesized that T cells may be involved in this effect based on previous studies by others (11, 19), so we specifically deleted T cells from a suspension of crude splenocytes using immunomagnetic depletion of CD3ε-expressing T cells. T-cell depletion was confirmed by flow cytometric analysis (Fig. 4A). When injected intravenously into intact mice, splenocytes depleted of T cells from GTS-21-treated mice did not inhibit acute pancreatic inflammation (Fig. 4, B–F).

Figure 3.

The α7nAChR agonist GTS-21 inhibits endoscopic retrograde cholangiopancreatography (ERCP) pressure-induced pancreatic edema, serum amylase, pancreatic myeloperoxidase (MPO) levels, and histopathology, and this effect is blocked by prior splenectomy. Results are normalized to percentage of the mean responses to pressure alone. A–D: effects of GTS-21 and splenectomy on pressure-induced pancreatic edema, plasma amylase, pancreatic MPO, and pancreatic histology score. E: effects of GTS-21 and splenectomy on pressure-induced pancreatic histology.*P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. Scale bar = 1 mm.

Figure 4.

Splenocytes from GTS-21-treated mice inhibit endoscopic retrograde cholangiopancreatography (ERCP)-induced pancreatitis but splenocytes from control (vehicle-treated) mice do not. Results are normalized to percentage of the mean responses to pressure alone. A: flow cytometric analysis demonstrating the depletion of T cells from a preparation of total mouse splenocytes by the use of magnetic beads coupled to the T-cell-specific antigen CD3ε. T cells from the spleens of both vehicle-treated (control) and GTS-21-treated mice were depleted by about 94%. B–E: effects of various splenocyte preparations on pressure-induced pancreatic edema, plasma amylase, pancreatic myeloperoxidase (MPO), and pancreatic histology score. F: effects of various splenocyte preparations on pressure-induced pancreatic histology. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. Scale bar = 1 mm.

The results described previously have come from pretreating the mice with nicotine or GTS-21 30 min before initiating pancreatic damage by ERCP or before harvesting splenocytes/T cells. To determine if nicotinic stimulation can reverse pancreatic damage after it has begun, we administered GTS-21 at 1, 2, 4, or 12 h after performing the ERCP procedure. When given after initiating pancreatic damage, GTS-21 was able to inhibit inflammation significantly and time dependently (Supplemental Fig. S1; see https://doi.org/10.6084/m9.figshare.20122724.v1).

To determine if this posttreatment nicotinic protection was mediated by splenocytes, we repeated the same experiment using splenectomized mice. Just as in the pretreatment study, splenectomy abolished the protective effects of GTS-21 administered after ERCP surgery (Supplemental Fig. S2). This suggests that nicotinic α7nAChR stimulation can treat as well as prevent acute pancreatitis in this model via a spleen-dependent mechanism.

Finally, since it has previously been demonstrated that nicotinic stimulation is protective in the caerulein hyperstimulation model of acute pancreatitis in mice (9), we tested the effects of prior splenectomy in this model to determine if nicotinic stimulation of splenocytes may be protective in general. We found that pretreatment with GTS-21 is protective in the caerulein model and that this effect is blocked by prior splenectomy (Supplemental Fig. S3).

DISCUSSION

The current study demonstrates that nicotine is likely to be the active factor in tobacco smoke that protects against ERCP-induced acute pancreatitis in a mouse model and that the spleen is involved in this effect. It has previously been shown that nicotinic stimulation can ameliorate acute pancreatitis caused by caerulein hyperstimulation in mice (9), but the present results are the first to demonstrate, to our knowledge, that nicotinic stimulation can protect against experimental ERCP-induced acute pancreatitis. The previous study demonstrated that vagotomy or pretreatment with a cholinergic antagonist drug caused enhanced severity of pancreatitis, whereas pretreatment with GTS-21, a specific α7nAChR partial agonist, or vagal stimulation significantly decreased the severity of acute pancreatitis. These findings were consistent with the concept that local pancreatic damage caused by caerulein hyperstimulation initiates a vagally mediated cholinergic nicotinic protective mechanism, evidenced by the worsening of inflammation if either vagotomy is performed or a cholinergic antagonist drug is given; this reflex is strengthened by vagal stimulation or nicotinic cholinergic agonism. Further evidence for this protective effect of nicotine came from a study showing that intravenous treatment with nicotine protected rats against acute pancreatitis caused by injection of glycodeoxycholic acid into the pancreatic duct plus caerulein hyperstimulation (10). Also, it has been shown that intraperitoneal nicotine pretreatment dose dependently inhibits acute pancreatitis in models in which a bile acid is injected retrogradely into the pancreatic duct in both mice (11) and rats (12). Notably, the effective doses of nicotine in the previous mouse study (10) mirror closely the doses found to be effective in the current study.

In agreement with the earlier study of nicotinic protection against acute pancreatitis caused by caerulein hyperstimulation (9), we found that GTS-21, a specific partial agonist of α7nAChRs, also elicits nicotinic protection against ERCP-induced acute pancreatitis. In addition, we found for the first time, to our knowledge, that splenectomy performed 7 days before ERCP abolished the protective effect of GTS-21, suggesting that a splenocyte mediates the nicotinic protective effect. It has been shown previously that α7nAChRs mediate the immunosuppressive functions of CD4⁺CD25⁺ regulatory T cells (Tregs; 19) and that nicotine pretreatment in mice with acute pancreatitis caused by intrapancreatic duct bile acid injection resulted in upregulation of the number and suppressive effects of CD4⁺CD25⁺ T cells (Tregs) circulating in the blood (11). This study did not report the effects of T cells on pancreatitis. These findings led us to hypothesize that T cells may be the splenic cell type mediating nicotinic protection in ERCP-induced pancreatitis.

To determine if T cells mediate the protective nicotinic effects in this model, we devised a strategy of specifically deleting T cells from preparations of crude splenocytes. We found that magnetically deleting T cells labeled with CD3ε antibodies was 94% effective (Fig. 4). T-cell-depleted splenocytes prepared from GTS-21-treated mice did not protect against ERCP-induced pancreatitis, whereas crude splenocytes from GTS-21-treated mice did protect the pancreas (Fig. 4). These results are supported by the finding that splenic CD4⁺CD25⁺ regulatory T cells (Tregs) are increased in the caerulein hyperstimulation model of acute pancreatitis in mice (20).

One possible limitation to the data presented here is the lack of demonstration of the effect of administering a purified preparation of T cells on ERCP-induced acute pancreatitis. We showed that a preparation of splenocytes lacking nicotinic receptor-stimulated T cells does not protect the pancreas, whereas a preparation of total splenocytes, including T cells, similarly stimulated, does protect the pancreas. It may be possible in future studies to corroborate these findings by treating mice undergoing ERCP with T cells selected by immunomagnetic or other methods of cell subtype selection prepared from crude splenocyte suspensions.

We were also interested in whether nicotine treatment initiated after the ERCP-induced acute pancreatitis can reverse ongoing inflammation. This is an important question because there are currently no clinically effective pharmacological treatments for human acute pancreatitis. Therefore, we administered GTS-21 intraperitoneally at 1, 2, 4, and 12 h after ERCP surgery. We observed significant inhibition of most inflammatory indices when GTS-21 was given 2–12 h after ERCP, suggesting both that nicotine treatment started after the original inflammatory insult is effective and that there is some resistance to the effects of nicotinic stimulation in the first hour after ERCP starts. These results are consistent with earlier findings in the glycodeoxycholic acid (GDOC)/caerulein model in which intravenous nicotine was infused for 9 h starting 3 h after GDOC and caerulein were administered (10). When GTS-21 was given 2 h after ERCP to splenectomized mice, the nicotinic protection was abolished (Supplemental Fig. S1), demonstrating that splenocytes are also involved in nicotinic protection against ongoing pancreatic inflammation.

These results may be relevant to ERCP-induced acute pancreatitis in people because DiMagno et al. (6, 7) showed that current smoking is protective against ERCP-induced pancreatitis in human patients. It is not known if this protective effect of current smoking is due to nicotine or some other component of tobacco smoke. However, it is possible that the pancreatic protective effect of current smoking on ERCP-induced pancreatitis is due to the effects of nicotine on splenocytes. Support for this concept comes from a case-control study of human acute pancreatitis in which it was found that a prior splenectomy is a significant risk factor for developing acute pancreatitis, suggesting that the spleen provides tonic protection against pancreatic inflammation in people (21).

The spleen is a reservoir of many types of immune cells, with both pro- and anti-inflammatory effects. A proinflammatory role for splenocytes in experimental acute pancreatitis was described in studies demonstrating that splenectomy protected against pancreatic inflammation in a mouse autoimmune pancreatitis model (22) and that splenectomy performed after induction of acute pancreatitis in rats was partially protective of pancreatic damage (23). Our findings that splenic T cells protect against acute pancreatitis when stimulated by nicotine coupled with previous observations that the vagus provides a tonic cholinergic protection of the mouse pancreas (9, 12) and that splenectomy is a risk factor for human pancreatitis (21) suggest that there is a delicate balance between splenic pro- and anti-inflammatory influences on the pancreas. Stimulation of α7nAChRs to tilt this balance toward protection may prove useful in preventing acute pancreatitis when it can be predicted, such as before ERCP procedures, or in reversing ongoing pancreatitis in other situations.

SUPPLEMENTAL DATA

Supplemental Figs. S1–S3: https://doi.org/10.6084/m9.figshare.20122724.v1.

GRANTS

This work was supported by the National Institutes of Health Grants DK120555, DK125308, and DK124474 (to R.A.L.).

DISCLOSURES

A patent application has been filed by Duke University including some data presented in this article.

AUTHOR CONTRIBUTIONS

S.R.V. and R.A.L. conceived and designed research; R.A.S. and M.-N.H. performed experiments; R.A.S., S.R.V. and M.-N.H. analyzed data; S.R.V., M.-N.H., M.D.G., and R.A.L. interpreted results of experiments; S.R.V. prepared figures; S.R.V. drafted manuscript; S.R.V., M.-N.H., M.D.G., and R.A.L. edited and revised manuscript; R.A.S., S.R.V., M.-N.H., M.D.G., and R.A.L. approved final version of manuscript.