- Email: [email protected]
Pancreatic exocrine secretion in acute experimental pancreatitis
GASTROENTEROLOGY
1990;99:1120-1127
Pancreatic Exocrine Secretion in Acute Experimental Pancreatitis CLAUS NIEDERAU, MARCUS NIEDERAU, REINHARD GEORG STROHMEYER, LINDA D. FERRELL, and JAMES H. GRENDELL
LijTHEN,
Medizinische Klinik und Poliklinik, Abteilung for Gastroenterologie, Heinrich-Heine-Universit%t Dttsseldorf, Federal Republic of Germany; Departments of Medicine, Physiology and Pathology, University of California, San Francisco, California; and Medical Service, San Francisco General Hospital, San Francisco, California
Little is known about exocrine pancreatic secretory function in patients with acute pancreatitis, in particular during the early phase of the disease. Therefore, this study evaluates basal and stimulated pancreatic secretion in vivo and in vitro in four different models of acute pancreatitis which reflect its clinical spectrum of severity: (a) edematous pancreatitis induced in the rat by seven IP injections of 50 pg/kg cerulein at hourly intervals; (b) edematous pancreatitis with cellular necrosis induced in the mouse by seven IP injections of 50 pg/kg cerulein at hourly intervals; (c) hemorrhagic pancreatitis induced in the mouse by feeding an ethionine-supplemented, choline-deficient diet for 66 hours; and(d) hemorrhagic pancreatitis induced in the rat by retrograde infusion of 0.6 mL 5% sodium taurocholate into the pancreatic duct. Secretory studies were performed in vivo and in vitro at various times after onset of pancreatitis. The results show that the exocrine pancreas gradually became resistant to cholecystokinin stimulation after the onset of acute pancreatitis in all four animal models. Cholecystokinin-stimulated secretion was almost abolished in vivo and in vitro at the time of maximal histological damage. In vivo basal secretion was also reduced. In vitro there was an increase in basal release of amylase from isolated acini that was not caused by an increase in luminal secretion but by enzyme release from damaged cells. The time course of improvement of secretory function after acute experimental pancreatitis depended on the severity of the pancreatitis. Recovery of secretory capacity took longer after severe necrotizing pancreatitis than after edematous pancreatitis. However, the ultimate resolution of secretory function was remarkable, in particular after severe hemorrhagic pancreatitis. In all four models, secretory capacity became indistinguishable from normal before the morphological
alterations had completely resolved. The present experimental data suggest that pancreatic secretion, and particularly pancreatic secretory response to cholecystokinin, may also be reduced in patients early after the onset of acute pancreatitis. cute pancreatitis remains a disease with significant morbidity and mortality (1,Z). All previous attempts to treat acute pancreatitis by inhibition of secretion have been unsuccessful, including treatment with atropine (3), cimetidine (4-71, glucagon (8-111, calcitonin (12). and somatostatin (13). As pointed out by Steinberg and Schlesselmann, most of these previous human studies had insufficient statistical power to provide confidence in the negative outcome, either because the studies had too few patients or the event rates were too low (14). Nevertheless, the failure of all of the studies that treated pancreatitis by inhibition of secretion raises the question of whether these inhibitory substances cause the same effects in patients with acute pancreatitis as in healthy subjects. The effects of these substances on exocrine pancreatic secretion have not been studied in patients with acute pancreatitis (15-17). Furthermore, little is known about pancreatic secretory function in patients with acute pancreatitis, in particular during the early phase of the disease. A few limited reports of experimental studies suggest that pancreatic secretion may be reduced in acute pancreatitis (18-22). This study evaluates basal and stimulated pancreatic secretion in vivo and in vitro in four different
A
Abbreviations used in this paper: CDE, choline-deficient; LDH, lactate dehydrogenese. 0 1990 by the American Gastroenterological Association 00%5085/90/$3.00
PANCREATIC
October 1996
models of acute pancreatitis spectrum of severity.
that
reflect
its clinical
Materials and Methods Materials Cerulein was obtained from Farmitalia Carlo Erba (Freiburg, F.R.G.) or from Adria Laboratories (Columbus, OH); synthetic cholecystokinin-octapeptide (CCK-8) was obtained from Squibb (Princeton, NJ): Rompun (xylazine) was obtained from Bayer (Leverkusen, F.R.G.); and Ketanest was obtained (ketamine) from Parke Davis (Freiburg, F.R.G.). Purified collagenase (type CLSPA) was obtained from Worthington Biochemicals [Freehold, NJ) and from Biomedical Cooper (Malvern, PA). The choline-deficient diet was obtained from Teklad (Madison, WI), and was then supplemented with 0.5% D,L-ethionine (Sigma Chemicals, St, Louis, MO]. Other chemicals were obtained from Sigma Chemicals.
Biochemical
Determinations
Amylase concentration was determined by the Phadebas amylase assay (Pharmacia, Piscataway, NJ]. The method has been described in detail (23). In some in vitro experiments, release of lactate dehydrogenase (LDH) from isolated acini was determined using a commercially available test kit (Boehringer, Mannheim, F.R.G.). In the in vivo experiments, chymotrypsinogen and protein concentrations were determined in addition to amylase. Chymotrypsin activity was determined after activation with trypsin using methods described previously in detail (24,251. Protein concentration was estimated using the photometric method of Lowry et al. (26) with a standard curve made to bovine serum albumin.
Histological
Evaluations
The time courses of the morphological alterations have previously been published in detail for the various pancreatitis models (17,27-29). For light microscopy, the pancreas was quickly removed and fixed overnight at room temperature in 10% formaldehyde buffered to pH 7.0. Tissue was subsequently embedded in paraffin, sectioned in 5-pm slices, and stained with H&E.
Experimental
Induction
of Acute
Pancreatitis
Cerulein-induced pancreatitis in the rat. Male Sprague-Dawley rats (Simonsen Laboratories, Gilroy. CA) weighing 200-300 g were used in these studies. Pancreatitis was induced by seven IP injections of 50 Mg/kg cerulein at hourly intervals over a 6-hour period. Hyperstimulation by supramaximal doses of cerulein in the rat causes intracellular vacuolization, interstitial inflammation, and edema as well as some cellular necrosis (18.30) (Figure 1; Table 1). Cerulein-induced pancreatitis in the mouse. Male NMRI mice (Zentralinstitut fiir Tierversuchszucht, Hannover, F.R.G.) or male Swiss Webster mice (Simonsen
SECRETION
IN ACUTE PANCREATITIS
1121
Laboratories), 4-6 weeks old and weighing 14-18 grams, were used in these studies. All animals were fasted for 16-18 hours before the experiment, but were permitted water ad libitum. Pancreatitis was induced by seven IP injections of 50 pg/kg cerulein at hourly intervals over a 6-hour period. Cerulein hyperstimulation in the mouse causes similar morphological alterations compared with the corresponding rat model. However, there is more cellular necrosis in the mouse model than in the rat (27) [Figure 1; Table 1). The cerulein models in both the rat and the mouse are not associated with significant mortality and resemble the biochemical and histological abnormalities of human acute edematous-interstitial pancreatitis. Choline-deficient diet-induced pancreatitis. Female NMRI mice (Zentralinstitut fiir Tierversuchszucht) or female Swiss Webster mice (Simonsen Laboratories], 4-6 weeks old and weighing 10-14 g, were used in these studies. All mice were fed regular laboratory chow ad libitum before the experiments. The ethionine-supplemented (0.5% ethionine), choline-deficient (CDE) diet (test diet; Teklad, Madison, WI] was then substituted for a period of 66 hours, after which it was replaced by regular chow. This period of feeding the CDE diet causes a necrotizing pancreatitis with major hemorrhage and fat necrosis [17,30) (Figure 1; Table 1) that is associated with a high mortality ranging from 50% to 70% (17). Many of the biochemical and histological features of this pancreatitis model resemble the alterations seen in severe human necrotizing-hemorrhagic pancreatitis (17,20,30,31). Taurocholate-induced pancreatitis in the rat. Male Sprague-Dawley rats (Simonsen Laboratories) or male Wistar rats (Zentralinstitut ffir Tierversuchszucht), weighing LOO300 g, were used in these studies. After anesthesia had been induced by IM injection of a mixture of 87 mg/kg ketamine and 13 mg/kg xylazine, a laparotomy was performed. Thereafter, the pancreaticobiliary duct was cannulated through the duodenal papilla with a polyethylene catheter (PE 10; Clay Adams, Parsippany, NJ] that had been introduced by means of puncture of the duodenum. A precision pump was used to infuse 0.6 mL 5% sodium taurocholate into the pancreaticobiliary duct during a lo-minute period at an infusion rate of 6 mL/h. The catheter was then withdrawn, and the abdominal cavity was surgically closed. The retrograde infusion of the bile salt causes a severe necrotizing pancreatitis in a matter of a few hours (31-34) (Figure 1; Table 11. The biochemical and histological features of this pancreatitis model resemble the alterations seen in fulminant necrosis of the pancreas in humans. Similar to severe human acute necrotizing pancreatitis, mortality from taurocholate-induced pancreatitis ranges from 70% to 80% (1516.31-34).
Secretory
Studies
In vitro studies. RAT AND MOUSE STUDIES. Isolated acini were prepared by collagenase digestion of rat or mouse pancreas as described previously in detail. Basal and CCK8-stimulated release of amylase from isolated acini was measured over a 30-minute period (35,361. In vitro secretory studies were performed in animals without pancreatitis and at various times after onset of pancreatitis.
1122
NIEDERAU
GASTROENTEROLOGY
ET AL.
Vol. 99, No. 4
Figure I. Pancreatic histology at the time of maximal morphological damage. A and
.
.
.
B. Cerulein-induced pancreatitis in the rat 12 hours after the beginning of the cerulein injections. Note marked interstitial inflammation and edema (A]. Higher magnification also shows intracellular vacuolization and some acinar cell necrosis (B). C and D. Cerulein-induced pancreatitis in the mouse 10 hours after the beginning of the cerulein injections. Similar to the rat model, there is interstitial edema and intracellular vacuolization as well as cell necrosis (arrows] (Cl. In contrast to the rat model, some areas show marked necrosis of acinar cells (D). E and F. Choline-deficient diet-induced pancreatitis 60 hours after the beginning of the CDE diet. In addition to
October 1990
PANCREATIC SECRETION IN ACUTE PANCREATITIS
Table 1. Time Course, Severity, and Mortality
1123
of Four Models of Acute Pancreatitis
Light microscopic findings Experimental models of pancreatitis
Early injury
Maximal injury
Resolution of injury
Cerulein model in the rat
2-3 h
12-16 h
96-144 h
Cerulein model in the mouse
2-3 h
lo-12 h
96-144 h
Diet model in the mouse
24-33 h
60-70 h
4-5 wk
Taurocholate model in the rat
0.5-1h
12-16 h
5-6 wk
In vivo studies. RAT EXPERIMENTS. In vivo secretory studies were performed in animals without pancreatitis and at various times after onset of pancreatitis. After anesthesia had been induced by IM injection of a mixture of 87 mg/kg ketamine and 13 mg/kg xylazine, the left iugular vein was cannulated with polyethylene tubing (PE 10). A laparotomy was then performed, and the pancreaticobiliary duct was cannulated through the duodenal papilla with polyethylene tubing (PE 10). Pancreaticobiliary secretion was allowed to equilibrate for 30 minutes before the experiments. After nonstimulated secretion was collected for two lo-minute periods, cerulein was administered as an IV bolus at various doses. Cerulein-stimulated secretion was collected as six consecutive lo-minute fractions from which the mean integrated l-hour response was calculated. In vivo studies. MOUSE EXPERIMENTS. Because of the small size of mice (10-12 g body wt), cannulation of the pancreatic duct was not possible for technical reasons. Therefore, the amount of enzyme in the duodenal fluid was determined with and without prior stimulation by CCK-8 in the mouse experiments in vivo. Cholecystokinin-octapeptide was injected IP at various doses and at various times after induction of pancreatitis. Mice were killed 30 minutes after injection of CCK-8, and a 5.cm duodenal loop was quickly removed. The duodenal content was then washed out with 1 mL ice-cold saline and collected for biochemical analyses.
Severity at time of maximal injury
Mortality -
Edematous pancreatitis with minor cell necrosis Edematous pancreatitis with major cell necrosis Necrotizing pancreatitis with major hemorrhage Fulminant necrosis with major hemorrhage
50%-70% 70%-60%
of cerulein injections had been administered (Figure 2); data for cerulein-induced pancreatitis in the mouse were similar to those observed in the rat and therefore are not shown. In the diet model and in the taurocholate model, both of which are not associated with increased concentrations of circulating CCK at any
20 -
15 -
10 -
5-
I
*O 1, d
,b
0
do 6’,lb0 3bo ,cloo CCK-8
(PM)
Statistical Analysis Secretory data were compared with corresponding data from experiments with other doses of agonists or to data from experiments without an agonist by analysis of variance using Duncan’s method (37). Significance was assumed for P < 0.05.
Results The secretory response to CCK-8 and cerulein was significantly altered early in the time course of experimental pancreatitis in all animal models compared with the secretory response of animals without pancreatitis. In the two models involving hyperstimulation by cerulein, the maximal response to cerulein was markedly inhibited shortly after the complete set
0
0.0 1
0,l CAERULEIN
1.0
10
100
@M/kg)
Figure 2. In vitro secretory studies (A) and in viva secretory studies (B)in cerulein-induced pancreatitis in the rat. Data represent mean * SE from four to eight independent experiments for each dose of agonist and each time point. Secretory studies were performed at various time intervals after the beginning of the cerulein injections as indicated in the figure.
1124
NIEDERAU
GASTROENTEROLOGY
ET AL.
time points at which pancreatic secretion was decreased (17, the dose-response curve was shifted to the right early after the onset of pancreatitis (Figures 3 and 4). The maximal secretory response gradually decreased during the development of diet- and taurocholate-induced pancreatitis [Figures 3 and 4). In all four models of acute pancreatitis, CCK-stimulated secretion was almost abolished in vivo and in vitro at the time of maximal histological damage (Figures l-5; Table 1). The secretory response to cerulein in vivo was inhibited for amylase, chymotrypsinogen, and protein secretion to a similar degree (data for chymotrypsinogen and protein not shown]. Secretory impairments gradually improved during resolution of pancreatitis in all models. Pancreatic secretory response to CCK became virtually normal 10
CCK-8
Vol. 99, No. 4
(PM)
CAERULEIN @M/kg) I
I
I
I
I
I
1
0
3
10
30
60
100
3ca
CCK-8
I
loo0
(PM)
CCK-8 @M/kg)
Figure 3. In vitm secretory studies (A) and in vivo secretory studies (B) in diet-induced pancreatitis in the mouse. Data represent mean k SE from four to eight independent experiments for each dose of agonist and each time point. Because of the small size of mice, cannulation of the pancreatic duct was not possible for technical reasons. Therefore, the amount of enzyme in the duodenal fluid wan determined with and without prior stimulation by CCK-8 in the mouse experiments In viva (for details, see Materials and Methods). Secretory studies were performed at various time intervals after the beginning of the CDE diet as indicated in the figure.
Figure 4. In vitro secretory studies (A) and in vivo secretory studies (B) in taurocholate-induced paucreatitis in the rat. Data represent mean f SE from four to eight independent experiments for each dose of agonist and each time point. Secretory atudiee were done at various intervals after infbsion of taurocholate as indicated in the &ure.
days after the onset of cerulein-induced pancreatitis both in the rat (Figure 2) and in the mouse (data not shown). The improvement in secretion took longer in both models of necrotizing-hemorrhagic pancreatitis (Figures 3 and 4). The secretory response became normal only 2 weeks after the onset of diet-induced pancreatitis but was still markedly reduced after 10 days at a time point at which pancreatic secretion had become normal in the models of edematous pancreatitis (compare Figures 2 and 3). In taurocholate-induced hemorrhagic pancreatitis, the secretory response was still significantly reduced 2 weeks after the onset of pancreatitis and only returned to normal after 3 weeks (Figure 4). In all models, normalization of the secretory response to CCK or cerulein occurred when there were still significant morphological abnormalities (Table 1). Although basal secretion was significantly reduced in vivo in all four models, it remained up to 40% of secretion in animals without pancreatitis (Figure 2-5). Thus, impairment of basal secretion was not as marked
PANCREATIC
October 1990
SECRETION
IN ACUTE PANCREATITIS
BASAL SECRETION
STIMULATED
1125
SECRETION
IN VITRO
Figure 5. Basal and stimulated pancreatic secretion of amylase in vitro and in vivo. Data are calculated from secretion at the time point of maximal histological damage for each model and are expressed as percentage of secretion of animals without pancreatitis. Data for stimulated secretion were calculated for the dose of CC&8 or cerulein that was maxlmally effective in animals without pancreatitie. Data represent mean values from four to eight independent experiments.
caerulein model in mice caerulein model in rats
diet model in mice taurocholate model in rats
IN WV0 caerulein node1 in mice caerulein
model
taurocholate
in rats
model
in rats 0
100 % OF AMYIASE
as impairment of secretory capacity in response to CCK. There was an in vitro increase in basal amylase release from isolated acini (Figures 2-5) that is not caused by an increase in luminal secretion but by a release from damaged cells. This assumption was substantiated by measurements of basal release of LDH from isolated acini. Release of LDH from acini that had been prepared from animals with experimental pancreatitis was 50%-100% higher than LDH release from acini prepared from animals without pancreatitis (P < 0.01; data not shown in detail]. Correspondingly, only 75%-90% of acini prepared from animals with pancreatitis excluded trypan blue compared with more than 95% of acini from animals without pancreatitis. Discussion Several experimental studies and clinical experience indicate that stimulation of pancreatic secretion via CCK or a meal may worsen the course of acute pancreatitis (15-17). Therefore, experimental stimulation of pancreatic secretion during clinical acute pancreatitis is difficult to justify for ethical reasons. Correspondingly, there is virtually no clinical information about pancreatic secretion during the early phase of acute pancreatitis. The present studies in the experimental animal show that CCK-stimulated secretion is almost abolished in vivo and in vitro during acute experimental pancreatitis and that maximal inhibition is seen at the time of maximal histological damage. The secretory blockade is seen in various models of acute pancreatitis that reflect the clinical spectrum of severity of this disease. There are some clinical studies about pancreatic secretory function following an acute attack of pancreatitis. Studies using the PABA (para-
200
0
10
20
SECRETION OF ANIMALS WITHOUT PANCREATITIS
aminobenzoic acid) test, a tubeless indirect evaluation of pancreatic secretory function, indicate that pancreatic secretion is reduced in most patients 3-10 days after the onset of pancreatitis (38). In the study of Mitchell et al., pancreatic secretory capacity then gradually improved in most patients after a few months (381. Studies using direct function tests involving duodenal intubation and stimulation by cerulein or CCK have also shown that pancreatic secretory capacity is often reduced for considerable time intervals after the acute attack (39,40). In some patients with severe necrotizing pancreatitis, complete functional recovery took >l year (39). These clinical data agree with the present experimental results in showing that the time period for recovery of pancreatic secretory function after acute pancreatitis depends mainly on the severity of the attack. The good recovery of secretory alterations after experimental pancreatitis also reflects the clinical experience which suggests a great functional reserve of the exocrine pancreas. Some clinical studies have reported that pancreatic secretory capacity may permanently remain reduced in a considerable percentage of patients after severe necrosis of the pancreas (41,421. However, in the clinical situation it is often unknown whether pancreatic function was normal before the acute attack and whether the acute attack was not merely a severe episode of chronic relapsing pancreatitis. The present experiments evaluated the pancreatic secretory response to CCK and to the CCK analogue cerulein. The results show that the acinar cells gradually became resistant to CCK stimulation after the onset of acute pancreatitis in several different animal models. This situation is an interesting state of hormone resistance that needs to be characterized by further experimental and clinical studies. At least in
GASTROENTEROLOGY
1126 NIEDERALJ ET AL.
the models of hemorrhagic pancreatitis, the CCK resistance is not caused by increased levels of circulating CCK, which remain normal in both the diet and taurocholate models (17). In cerulein-induced pancreatitis, one must consider a possible component of desensitization by cerulein to account for the decrease in responsiveness because both in vitro (43) and in vivo (44) attenuation of the secretory response have been documented with CCK pretreatment. However, the attenuation of the secretory response after pretreatment by CCK did not reach the level of an almost complete abolishment of secretory response as it was observed in the present models of pancreatitis. Previous studies have shown that CCK-receptor antagonists may ameliorate pancreatitis induced by a CDE diet (17) and by injections of taurocholate (45). According to the present results, it is unlikely that CCK agonists or CCK antagonists exert their deleterious or beneficial effects by an action on pancreatic secretion. Preliminary data from studies in dietinduced acute pancreatitis in mice suggest that CCK still binds to its receptor and stimulates intracellular biochemical events that result in an increase in oxygen consumption (46).Such binding and action are not followed by an adequate response in secretion but may, via the decrease in intracellular PO,, cause additional damage to the cell. The main defect that causes the decrease in secretory responsiveness appears to be located beyond the receptor at an intracellular level of stimulus-secretion coupling. Previous studies using acini isolated from patients with acute pancreatitis or from rats that had been treated with supramaximal doses of cerulein indicate that in addition to the secretory blockade, synthesis of proteins may also be reduced in acute pancreatitis (30,471. The impairment of protein synthesis may at least partly explain the secretory blockade. However, in other models of pancreatitis, protein synthesis of the pancreatic acinar cell has been shown to be almost normal (19). Thus, further experiments are needed to identify the exact intracellular location of the defect that results in the secretory blockade. There is only some information about whether pancreatic secretory function during acute pancreatitis is also resistant to stimuli other than CCK, such as cholinergic agents and secretin. Experimental data from cerulein-induced pancreatitis in the rat indicate that stimulation by secretin may not be reduced to the same extent as stimulation by CCK (48).This assumption is substantiated by a clinical report which showed that the secretory response to secretin was normal only 2 weeks after the beginning of acute pancreatitis in most patients ‘f49). However, in taurocholateinduced pancreatitis secretin failed to cause any increase in pancreatic secretion (23).Sensitivity to carba-
Vol. 99, No. 4
chol is reduced in cerulein-induced pancreatitis (301, but apparently not to the same extent as sensitivity to CCK. The pathophysiology of pancreatitis, both in the experimental and in the clinical situation, is still incompletely understood. Therefore, it is difficult to transfer observations made in a particular animal model to the human situation. Because of these difficulties, we have studied several models of pancreatitis that are quite different in terms of severity, time course, and pathogenesis. Despite these marked differences, the results, which show a secretory blockade during pancreatitis, are strikingly similar in all models. Therefore, the experimental data presented and the clinical evidence discussed suggest that pancreatic secretion, and in particular the pancreatic secretory response to CCK, is reduced during acute pancreatitis, certainly in the experimental setting and possibly also in the clinical situation. This secretory blockade might at least partly explain the failure to treat acute pancreatitis effectively by inhibition of secretion.
References 1. Fayne
SD, Barkin JS. Acute pancreatitis. Mt Sinai J Med 1986;53:396-403. 2. Corfield AP. Cooper MJ, Williamson RCN. Acute pancreatitis: a lethal disease of increasing incidence. Gut 1985;26:724-729. 3. Cameron JL, Mehigan IJ, Zuidema GD. Evaluation of atropine in acute pancreatitis. Surg Gynecol Obstet 1979;148:206-208. 4. Broe PJ. Zinner MJ, Cameron JL. A clinical trial of cimetidine in acute pancreatitis. Surg Gynecol Obstet 1982;154:13-16. 5. Loiudice TA, Lang J, Mehta H. Treatment of acute alcoholic pancreatitis. Am J Gastroenterol1984;79:553-558. 6. Meshkinpour H, Molinari M, Gardner L. Cimetidine in the treatment of acute alcoholic pancreatitis. Gastroenterology 1979; 77:687-690. 7. Regan PT. Malegelada JR, Go VLW, Wolf AM, DiMagno EP. A prospective study of the antisecretory effects of cimetidine and glucagon in human acute pancreatitis. Mayo Clin Proc 1981;56: 499-503. a. DUrr HK, Maroske D, Zelder 0, Bode J. Glucagon therapy in acute pancreatitis. Gut 1978;19:175-179. 9. Kronberg 0, Bulow S, Joergensen PN. A randomized doubleblind trial of glucagon treatment of first attack of severe acute pancreatitis without associated biliary disease. Am J Gastroenterol1983;73:423-425. 10.MRC multicentre trial of glucagon and aprotinin. Lancet 1977;2: 632-635. 11.Olazabal A, Fuller R. Failure of glucagon in the treatment of alcoholic pancreatitis. Gastroenterology 1978;74:489-491. 12.Goebell H, Ammann R, Herfarth C. A double-blind trial of synthetic salmon calcitonin in the treatment of acute pancreatitis. Stand J Gastroenterol1979;14:881-889. 13.Usadel KH, Uberla KK, Leuscher U. Treatment of acute pancreatitis with somatostatin (abstr]. Dig Dis Sci 1985;30:A992. 14.Steinberg WM. Schlesselmann SE. Treatment of acute pancreatitis. Gastroenterology 1987;93:1420-1427. 15.Evander A, Ihse I. Lundquist 1. Influence of hormonal stimulation by caerulein on acute experimental pancreatitis in the rat. Eur SurgRes198X13:257-268,
October 1990
PANCREATIC
16. Evander
36. Niederau C, Niederau M, Williams JA, Grendell JH. New proglumide analogue CCK-receptor antagonists. Am J Physiol 1986;251:G856-G860. 37. Duncan DB. Multiple range tests for correlated and heteroscedastic means. Biometrics 1957;13:164-176. 38. Mitchell CJ, Playforth MJ, Kelleher J, McMahon MJ. Functional recovery of the exocrine pancreas after acute pancreatitis. Stand J Gastroenterol1983:18:5-8. 39. Angelini G, Pederzoli P, Caliari S, Fratton S, Brocco G, Marzoli G, Bovo P, Cavallini G. Scuro LA. Long-term outcome of acute necrohemorrhagic pancreatitis. Digestion 1984;30:131-137. 40. Gullo L. Sarles H, Mott CB. Functional investigation of the exocrine pancreas following acute pancreatitis. Rend Gastroenterol1972;4:18-21. 41.Seligson U, Cho JW, Ihre G, Lundh G, Pyk E. Evaluation of the diagnosis of pancreatitis. Stand J Gastroenterol1982;17:905-911. 42. Bttchler M, Malfertheiner P, Block S, Maier W, Beger HG. Morphologlsche und funktionelle Ver8nderungen des Pankreas nach akuter nekrotisierender Pankreatltis. Z Gastroenterol1985; 23:79-83. 43.Abdelmoumene S, Gardener JD. Cholecystokinin-induced desensitization of enzyme secretion in dispersed acini from guinea pig pancreas. Am J Physiol1989;239:G272-G279. 44. Otsuki M, Williams JA. Amylase secretion by isolated pancreatic acini after chronic cholecystokinin treatment in vivo. Am J Physiol1983;244:G683-G688. 45. Makovec F, Bani M, Cereda R, Chiste R, Revel L, Rovati LC, Setnikar I, Rovati LA. Protective effect of CR1409 (choleystokinin antagonist] on experimental pancreatitls in rats and mice. Peptides 1986;7:1159-1164. 46. Bhattacharya I, Hootman S, Williams J, Grendell JH. Cholecystokinin-stimulated oxygen consumption by mouse pancreatic acinar cells in vitro in diet-induced acute pancreatitis (abstr). Gastroenterology 1988;94:A33. 47. Adler G, Kern HF. Fine structure and biochemical studies in human acute pancreatitis. In: Gyr KE, Singer MV, Sarles H. eds. Pancreatitis. Amsterdam: Elsevier, 1984:37-42. 48. Renner IG. Wisner JR, Lavigne BC. Partial restoration of pancreatic function by exogenous secretin in rats with ceruletideinduced acute pancreatitis. Dig Dis Sci 1986;31:305-313. 49. Tiscornia OM, Dreiling DA. Does the pancreatic gland regenerate? Gastroenterology 1966;51:267-271.
A, Lundquist I, Ihse I. Influence of gastro-intestinal hormones on the course of acute experimental pancreatitis. Hepatogastroenterology 1962;29:161-166. 17. Niederau C, Liddle RA, Ferrell LD, Grendell JH. Beneficial effects of CCK-receptor blockade and inhibition of proteolytic enzyme activity in experimental acute hemorrhagic pancreatitis in mice. J Clin Invest 1986;78:1056-1083. 18. Adler G. Hupp T, Kern HF. Course and spontaneous regression of acute pancreatitis in the rat. Virchows Arch [A] 1979:382:3247. 19. Gilliland L, Steer ML. Effects of ethionine on digestive enzyme synthesis and discharge by mouse pancreas. Am J Physiol 1980:239:G418-G426. 20. Kioke K, Steer ML, Meldolesi J. Pancreatic effects of ethionine: blockade of exocytosis and appearance of crinophagy and autophagy precede cellular necrosis. Am J Physiol 1982;242: G297-G307. 21. Saluja A. Saito I, Saluja M. Houlihan MJ, Powers RE, Meldolesi J. Steer ML. In vivo rat pancreatic cell function during supramaximal stimulation with caerulein. Am J Physiol 1985;249:G702G710. 22. Evander A, Hederstrom E. Hultberg B, Ihse I. Exocrine pancreatic secretion in acute experimental pancreatitis. Digestion 1982;24:159-167. 23. Grendell JH, Rothman SS. Digestive end products mobilize secretory proteins from subcellular stores in the pancreas. Am J Physiol1981;241:G67-G73, 24. Glazer G, and Steer ML. Requirements for activation of trypsinogen and chymotrypsinogen in rabbit pancreatic juice. Anal Biochem 1977;77:130-140. 25. Hummel BCW. A modified spectrophotometric determination of chymotrypsin, trypsin, and thrombin. Can J Biochem Physiol 1959;37:1393-1399. 26. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurements with Folin phenol reagent. J Biol Chem 195X193: 265-275. 27.Niederau C, Ferrell LD, Grendell JH. 1985. Caerulein-induced acute necrotizing pancreatitis in mice. Gastroenterology 1985;88: 1192-1204. 28.Niederau C, Ferrell LD, Liddle RA. Grendell JH. Biochemical, secretory, and morphological effects of repetitive experimental acute pancreatitis (abstr). Gastroenterology 1986;90:A1565. 29. Tsay DG, Niederau C. Schmidt H, Goldberg HJ. Higgins CB, Ferrell LD, Grendell JH. Experimental acute pancreatitis. Invest Radio1 1987;22:556-561. 30. Lampel M. Kern HF. Acute interstitial pancreatitls in the rat induced by excessive doses of a pancreatic secretagogue. Virchows Arch [A] 1987;373:97-117. 31. Niederau C, Ferrell LD, Grendell JH. Akute experimentelle Pankreatitis. Internist 1986;27:681-696. 32. Lombardi B, Estes LW, Longnecker DS. Acute hemorrhagic pancreatitis [massive necrosis) with fat necrosis induced in mice by DL-ethionine fed with a choline-deficient diet. Am J Pathol 1975:79:464-480. 33.Lankisch PG. Winckler K. Bokermann M, Schmidt H, Creutzfeldt W. The influence of glucagon on acute experimental pancreatitis in the rat. Stand J Gastroenterol1974;9:725-729. 34. Niederau C, Chang L, James TL, Grendell JH. In vivo 31P magnetic resonance (MR] spectroscopy in normal rat pancreas and in experimental acute pancreatitis (abstr]. Digestion 1986;35: A43. 35. Williams JA. Korc M, Dormer RL. Action of secretagogues on a new preparation of functionally intact pancreatic acini. Am J Physiol1978;235:E517-E524.
SECRETION
IN ACUTE
PANCREATITIS
1127
Received September 18.1989. Accepted April 9.1990. Address requests for reprints to: Claus Niederau, M.D., Medizinische Klinik und Poliklinik, Abteilung fur Gastroenterologie, Heinrich-Heine-Universitgt Dusseldorf, MoorenstraBe 5, 4000 DUsseldorf, Federal Republic of Germany. Dr. Claus Niederau was supported by grants from the Deutsche Forschungsgemeinschaft (Ni 224/1-l, 224/2-l, and 224/2-2) and from the Minister fur Wissenschaft und Forschung des Landes Nordrhein-Westfalen. Dr. James H. Grendell was supported by grants from the Research Evaluation and Allocation Committee, the Committee on Research of the Academic Senate, School of Medicine, University of California, San Francisco, and the National Institute of Health (DK 38939). This work was presented in part at the Annual Meeting of the European Pancreatic Club, Marseille, France, September 5,1987 (abstract published in Digestion 1987;38:46). and at the Annual Meeting of the American Pancreatic Association, Chicago, Illinois, November 5.1987 (abstract published in Dig Dis Sci 1987;32:1178). The authors thank Christine Genz and Monika Ebbert for expert technical assistance.
1990;99:1120-1127
Pancreatic Exocrine Secretion in Acute Experimental Pancreatitis CLAUS NIEDERAU, MARCUS NIEDERAU, REINHARD GEORG STROHMEYER, LINDA D. FERRELL, and JAMES H. GRENDELL
LijTHEN,
Medizinische Klinik und Poliklinik, Abteilung for Gastroenterologie, Heinrich-Heine-Universit%t Dttsseldorf, Federal Republic of Germany; Departments of Medicine, Physiology and Pathology, University of California, San Francisco, California; and Medical Service, San Francisco General Hospital, San Francisco, California
Little is known about exocrine pancreatic secretory function in patients with acute pancreatitis, in particular during the early phase of the disease. Therefore, this study evaluates basal and stimulated pancreatic secretion in vivo and in vitro in four different models of acute pancreatitis which reflect its clinical spectrum of severity: (a) edematous pancreatitis induced in the rat by seven IP injections of 50 pg/kg cerulein at hourly intervals; (b) edematous pancreatitis with cellular necrosis induced in the mouse by seven IP injections of 50 pg/kg cerulein at hourly intervals; (c) hemorrhagic pancreatitis induced in the mouse by feeding an ethionine-supplemented, choline-deficient diet for 66 hours; and(d) hemorrhagic pancreatitis induced in the rat by retrograde infusion of 0.6 mL 5% sodium taurocholate into the pancreatic duct. Secretory studies were performed in vivo and in vitro at various times after onset of pancreatitis. The results show that the exocrine pancreas gradually became resistant to cholecystokinin stimulation after the onset of acute pancreatitis in all four animal models. Cholecystokinin-stimulated secretion was almost abolished in vivo and in vitro at the time of maximal histological damage. In vivo basal secretion was also reduced. In vitro there was an increase in basal release of amylase from isolated acini that was not caused by an increase in luminal secretion but by enzyme release from damaged cells. The time course of improvement of secretory function after acute experimental pancreatitis depended on the severity of the pancreatitis. Recovery of secretory capacity took longer after severe necrotizing pancreatitis than after edematous pancreatitis. However, the ultimate resolution of secretory function was remarkable, in particular after severe hemorrhagic pancreatitis. In all four models, secretory capacity became indistinguishable from normal before the morphological
alterations had completely resolved. The present experimental data suggest that pancreatic secretion, and particularly pancreatic secretory response to cholecystokinin, may also be reduced in patients early after the onset of acute pancreatitis. cute pancreatitis remains a disease with significant morbidity and mortality (1,Z). All previous attempts to treat acute pancreatitis by inhibition of secretion have been unsuccessful, including treatment with atropine (3), cimetidine (4-71, glucagon (8-111, calcitonin (12). and somatostatin (13). As pointed out by Steinberg and Schlesselmann, most of these previous human studies had insufficient statistical power to provide confidence in the negative outcome, either because the studies had too few patients or the event rates were too low (14). Nevertheless, the failure of all of the studies that treated pancreatitis by inhibition of secretion raises the question of whether these inhibitory substances cause the same effects in patients with acute pancreatitis as in healthy subjects. The effects of these substances on exocrine pancreatic secretion have not been studied in patients with acute pancreatitis (15-17). Furthermore, little is known about pancreatic secretory function in patients with acute pancreatitis, in particular during the early phase of the disease. A few limited reports of experimental studies suggest that pancreatic secretion may be reduced in acute pancreatitis (18-22). This study evaluates basal and stimulated pancreatic secretion in vivo and in vitro in four different
A
Abbreviations used in this paper: CDE, choline-deficient; LDH, lactate dehydrogenese. 0 1990 by the American Gastroenterological Association 00%5085/90/$3.00
PANCREATIC
October 1996
models of acute pancreatitis spectrum of severity.
that
reflect
its clinical
Materials and Methods Materials Cerulein was obtained from Farmitalia Carlo Erba (Freiburg, F.R.G.) or from Adria Laboratories (Columbus, OH); synthetic cholecystokinin-octapeptide (CCK-8) was obtained from Squibb (Princeton, NJ): Rompun (xylazine) was obtained from Bayer (Leverkusen, F.R.G.); and Ketanest was obtained (ketamine) from Parke Davis (Freiburg, F.R.G.). Purified collagenase (type CLSPA) was obtained from Worthington Biochemicals [Freehold, NJ) and from Biomedical Cooper (Malvern, PA). The choline-deficient diet was obtained from Teklad (Madison, WI), and was then supplemented with 0.5% D,L-ethionine (Sigma Chemicals, St, Louis, MO]. Other chemicals were obtained from Sigma Chemicals.
Biochemical
Determinations
Amylase concentration was determined by the Phadebas amylase assay (Pharmacia, Piscataway, NJ]. The method has been described in detail (23). In some in vitro experiments, release of lactate dehydrogenase (LDH) from isolated acini was determined using a commercially available test kit (Boehringer, Mannheim, F.R.G.). In the in vivo experiments, chymotrypsinogen and protein concentrations were determined in addition to amylase. Chymotrypsin activity was determined after activation with trypsin using methods described previously in detail (24,251. Protein concentration was estimated using the photometric method of Lowry et al. (26) with a standard curve made to bovine serum albumin.
Histological
Evaluations
The time courses of the morphological alterations have previously been published in detail for the various pancreatitis models (17,27-29). For light microscopy, the pancreas was quickly removed and fixed overnight at room temperature in 10% formaldehyde buffered to pH 7.0. Tissue was subsequently embedded in paraffin, sectioned in 5-pm slices, and stained with H&E.
Experimental
Induction
of Acute
Pancreatitis
Cerulein-induced pancreatitis in the rat. Male Sprague-Dawley rats (Simonsen Laboratories, Gilroy. CA) weighing 200-300 g were used in these studies. Pancreatitis was induced by seven IP injections of 50 Mg/kg cerulein at hourly intervals over a 6-hour period. Hyperstimulation by supramaximal doses of cerulein in the rat causes intracellular vacuolization, interstitial inflammation, and edema as well as some cellular necrosis (18.30) (Figure 1; Table 1). Cerulein-induced pancreatitis in the mouse. Male NMRI mice (Zentralinstitut fiir Tierversuchszucht, Hannover, F.R.G.) or male Swiss Webster mice (Simonsen
SECRETION
IN ACUTE PANCREATITIS
1121
Laboratories), 4-6 weeks old and weighing 14-18 grams, were used in these studies. All animals were fasted for 16-18 hours before the experiment, but were permitted water ad libitum. Pancreatitis was induced by seven IP injections of 50 pg/kg cerulein at hourly intervals over a 6-hour period. Cerulein hyperstimulation in the mouse causes similar morphological alterations compared with the corresponding rat model. However, there is more cellular necrosis in the mouse model than in the rat (27) [Figure 1; Table 1). The cerulein models in both the rat and the mouse are not associated with significant mortality and resemble the biochemical and histological abnormalities of human acute edematous-interstitial pancreatitis. Choline-deficient diet-induced pancreatitis. Female NMRI mice (Zentralinstitut fiir Tierversuchszucht) or female Swiss Webster mice (Simonsen Laboratories], 4-6 weeks old and weighing 10-14 g, were used in these studies. All mice were fed regular laboratory chow ad libitum before the experiments. The ethionine-supplemented (0.5% ethionine), choline-deficient (CDE) diet (test diet; Teklad, Madison, WI] was then substituted for a period of 66 hours, after which it was replaced by regular chow. This period of feeding the CDE diet causes a necrotizing pancreatitis with major hemorrhage and fat necrosis [17,30) (Figure 1; Table 1) that is associated with a high mortality ranging from 50% to 70% (17). Many of the biochemical and histological features of this pancreatitis model resemble the alterations seen in severe human necrotizing-hemorrhagic pancreatitis (17,20,30,31). Taurocholate-induced pancreatitis in the rat. Male Sprague-Dawley rats (Simonsen Laboratories) or male Wistar rats (Zentralinstitut ffir Tierversuchszucht), weighing LOO300 g, were used in these studies. After anesthesia had been induced by IM injection of a mixture of 87 mg/kg ketamine and 13 mg/kg xylazine, a laparotomy was performed. Thereafter, the pancreaticobiliary duct was cannulated through the duodenal papilla with a polyethylene catheter (PE 10; Clay Adams, Parsippany, NJ] that had been introduced by means of puncture of the duodenum. A precision pump was used to infuse 0.6 mL 5% sodium taurocholate into the pancreaticobiliary duct during a lo-minute period at an infusion rate of 6 mL/h. The catheter was then withdrawn, and the abdominal cavity was surgically closed. The retrograde infusion of the bile salt causes a severe necrotizing pancreatitis in a matter of a few hours (31-34) (Figure 1; Table 11. The biochemical and histological features of this pancreatitis model resemble the alterations seen in fulminant necrosis of the pancreas in humans. Similar to severe human acute necrotizing pancreatitis, mortality from taurocholate-induced pancreatitis ranges from 70% to 80% (1516.31-34).
Secretory
Studies
In vitro studies. RAT AND MOUSE STUDIES. Isolated acini were prepared by collagenase digestion of rat or mouse pancreas as described previously in detail. Basal and CCK8-stimulated release of amylase from isolated acini was measured over a 30-minute period (35,361. In vitro secretory studies were performed in animals without pancreatitis and at various times after onset of pancreatitis.
1122
NIEDERAU
GASTROENTEROLOGY
ET AL.
Vol. 99, No. 4
Figure I. Pancreatic histology at the time of maximal morphological damage. A and
.
.
.
B. Cerulein-induced pancreatitis in the rat 12 hours after the beginning of the cerulein injections. Note marked interstitial inflammation and edema (A]. Higher magnification also shows intracellular vacuolization and some acinar cell necrosis (B). C and D. Cerulein-induced pancreatitis in the mouse 10 hours after the beginning of the cerulein injections. Similar to the rat model, there is interstitial edema and intracellular vacuolization as well as cell necrosis (arrows] (Cl. In contrast to the rat model, some areas show marked necrosis of acinar cells (D). E and F. Choline-deficient diet-induced pancreatitis 60 hours after the beginning of the CDE diet. In addition to
October 1990
PANCREATIC SECRETION IN ACUTE PANCREATITIS
Table 1. Time Course, Severity, and Mortality
1123
of Four Models of Acute Pancreatitis
Light microscopic findings Experimental models of pancreatitis
Early injury
Maximal injury
Resolution of injury
Cerulein model in the rat
2-3 h
12-16 h
96-144 h
Cerulein model in the mouse
2-3 h
lo-12 h
96-144 h
Diet model in the mouse
24-33 h
60-70 h
4-5 wk
Taurocholate model in the rat
0.5-1h
12-16 h
5-6 wk
In vivo studies. RAT EXPERIMENTS. In vivo secretory studies were performed in animals without pancreatitis and at various times after onset of pancreatitis. After anesthesia had been induced by IM injection of a mixture of 87 mg/kg ketamine and 13 mg/kg xylazine, the left iugular vein was cannulated with polyethylene tubing (PE 10). A laparotomy was then performed, and the pancreaticobiliary duct was cannulated through the duodenal papilla with polyethylene tubing (PE 10). Pancreaticobiliary secretion was allowed to equilibrate for 30 minutes before the experiments. After nonstimulated secretion was collected for two lo-minute periods, cerulein was administered as an IV bolus at various doses. Cerulein-stimulated secretion was collected as six consecutive lo-minute fractions from which the mean integrated l-hour response was calculated. In vivo studies. MOUSE EXPERIMENTS. Because of the small size of mice (10-12 g body wt), cannulation of the pancreatic duct was not possible for technical reasons. Therefore, the amount of enzyme in the duodenal fluid was determined with and without prior stimulation by CCK-8 in the mouse experiments in vivo. Cholecystokinin-octapeptide was injected IP at various doses and at various times after induction of pancreatitis. Mice were killed 30 minutes after injection of CCK-8, and a 5.cm duodenal loop was quickly removed. The duodenal content was then washed out with 1 mL ice-cold saline and collected for biochemical analyses.
Severity at time of maximal injury
Mortality -
Edematous pancreatitis with minor cell necrosis Edematous pancreatitis with major cell necrosis Necrotizing pancreatitis with major hemorrhage Fulminant necrosis with major hemorrhage
50%-70% 70%-60%
of cerulein injections had been administered (Figure 2); data for cerulein-induced pancreatitis in the mouse were similar to those observed in the rat and therefore are not shown. In the diet model and in the taurocholate model, both of which are not associated with increased concentrations of circulating CCK at any
20 -
15 -
10 -
5-
I
*O 1, d
,b
0
do 6’,lb0 3bo ,cloo CCK-8
(PM)
Statistical Analysis Secretory data were compared with corresponding data from experiments with other doses of agonists or to data from experiments without an agonist by analysis of variance using Duncan’s method (37). Significance was assumed for P < 0.05.
Results The secretory response to CCK-8 and cerulein was significantly altered early in the time course of experimental pancreatitis in all animal models compared with the secretory response of animals without pancreatitis. In the two models involving hyperstimulation by cerulein, the maximal response to cerulein was markedly inhibited shortly after the complete set
0
0.0 1
0,l CAERULEIN
1.0
10
100
@M/kg)
Figure 2. In vitro secretory studies (A) and in viva secretory studies (B)in cerulein-induced pancreatitis in the rat. Data represent mean * SE from four to eight independent experiments for each dose of agonist and each time point. Secretory studies were performed at various time intervals after the beginning of the cerulein injections as indicated in the figure.
1124
NIEDERAU
GASTROENTEROLOGY
ET AL.
time points at which pancreatic secretion was decreased (17, the dose-response curve was shifted to the right early after the onset of pancreatitis (Figures 3 and 4). The maximal secretory response gradually decreased during the development of diet- and taurocholate-induced pancreatitis [Figures 3 and 4). In all four models of acute pancreatitis, CCK-stimulated secretion was almost abolished in vivo and in vitro at the time of maximal histological damage (Figures l-5; Table 1). The secretory response to cerulein in vivo was inhibited for amylase, chymotrypsinogen, and protein secretion to a similar degree (data for chymotrypsinogen and protein not shown]. Secretory impairments gradually improved during resolution of pancreatitis in all models. Pancreatic secretory response to CCK became virtually normal 10
CCK-8
Vol. 99, No. 4
(PM)
CAERULEIN @M/kg) I
I
I
I
I
I
1
0
3
10
30
60
100
3ca
CCK-8
I
loo0
(PM)
CCK-8 @M/kg)
Figure 3. In vitm secretory studies (A) and in vivo secretory studies (B) in diet-induced pancreatitis in the mouse. Data represent mean k SE from four to eight independent experiments for each dose of agonist and each time point. Because of the small size of mice, cannulation of the pancreatic duct was not possible for technical reasons. Therefore, the amount of enzyme in the duodenal fluid wan determined with and without prior stimulation by CCK-8 in the mouse experiments In viva (for details, see Materials and Methods). Secretory studies were performed at various time intervals after the beginning of the CDE diet as indicated in the figure.
Figure 4. In vitro secretory studies (A) and in vivo secretory studies (B) in taurocholate-induced paucreatitis in the rat. Data represent mean f SE from four to eight independent experiments for each dose of agonist and each time point. Secretory atudiee were done at various intervals after infbsion of taurocholate as indicated in the &ure.
days after the onset of cerulein-induced pancreatitis both in the rat (Figure 2) and in the mouse (data not shown). The improvement in secretion took longer in both models of necrotizing-hemorrhagic pancreatitis (Figures 3 and 4). The secretory response became normal only 2 weeks after the onset of diet-induced pancreatitis but was still markedly reduced after 10 days at a time point at which pancreatic secretion had become normal in the models of edematous pancreatitis (compare Figures 2 and 3). In taurocholate-induced hemorrhagic pancreatitis, the secretory response was still significantly reduced 2 weeks after the onset of pancreatitis and only returned to normal after 3 weeks (Figure 4). In all models, normalization of the secretory response to CCK or cerulein occurred when there were still significant morphological abnormalities (Table 1). Although basal secretion was significantly reduced in vivo in all four models, it remained up to 40% of secretion in animals without pancreatitis (Figure 2-5). Thus, impairment of basal secretion was not as marked
PANCREATIC
October 1990
SECRETION
IN ACUTE PANCREATITIS
BASAL SECRETION
STIMULATED
1125
SECRETION
IN VITRO
Figure 5. Basal and stimulated pancreatic secretion of amylase in vitro and in vivo. Data are calculated from secretion at the time point of maximal histological damage for each model and are expressed as percentage of secretion of animals without pancreatitis. Data for stimulated secretion were calculated for the dose of CC&8 or cerulein that was maxlmally effective in animals without pancreatitie. Data represent mean values from four to eight independent experiments.
caerulein model in mice caerulein model in rats
diet model in mice taurocholate model in rats
IN WV0 caerulein node1 in mice caerulein
model
taurocholate
in rats
model
in rats 0
100 % OF AMYIASE
as impairment of secretory capacity in response to CCK. There was an in vitro increase in basal amylase release from isolated acini (Figures 2-5) that is not caused by an increase in luminal secretion but by a release from damaged cells. This assumption was substantiated by measurements of basal release of LDH from isolated acini. Release of LDH from acini that had been prepared from animals with experimental pancreatitis was 50%-100% higher than LDH release from acini prepared from animals without pancreatitis (P < 0.01; data not shown in detail]. Correspondingly, only 75%-90% of acini prepared from animals with pancreatitis excluded trypan blue compared with more than 95% of acini from animals without pancreatitis. Discussion Several experimental studies and clinical experience indicate that stimulation of pancreatic secretion via CCK or a meal may worsen the course of acute pancreatitis (15-17). Therefore, experimental stimulation of pancreatic secretion during clinical acute pancreatitis is difficult to justify for ethical reasons. Correspondingly, there is virtually no clinical information about pancreatic secretion during the early phase of acute pancreatitis. The present studies in the experimental animal show that CCK-stimulated secretion is almost abolished in vivo and in vitro during acute experimental pancreatitis and that maximal inhibition is seen at the time of maximal histological damage. The secretory blockade is seen in various models of acute pancreatitis that reflect the clinical spectrum of severity of this disease. There are some clinical studies about pancreatic secretory function following an acute attack of pancreatitis. Studies using the PABA (para-
200
0
10
20
SECRETION OF ANIMALS WITHOUT PANCREATITIS
aminobenzoic acid) test, a tubeless indirect evaluation of pancreatic secretory function, indicate that pancreatic secretion is reduced in most patients 3-10 days after the onset of pancreatitis (38). In the study of Mitchell et al., pancreatic secretory capacity then gradually improved in most patients after a few months (381. Studies using direct function tests involving duodenal intubation and stimulation by cerulein or CCK have also shown that pancreatic secretory capacity is often reduced for considerable time intervals after the acute attack (39,40). In some patients with severe necrotizing pancreatitis, complete functional recovery took >l year (39). These clinical data agree with the present experimental results in showing that the time period for recovery of pancreatic secretory function after acute pancreatitis depends mainly on the severity of the attack. The good recovery of secretory alterations after experimental pancreatitis also reflects the clinical experience which suggests a great functional reserve of the exocrine pancreas. Some clinical studies have reported that pancreatic secretory capacity may permanently remain reduced in a considerable percentage of patients after severe necrosis of the pancreas (41,421. However, in the clinical situation it is often unknown whether pancreatic function was normal before the acute attack and whether the acute attack was not merely a severe episode of chronic relapsing pancreatitis. The present experiments evaluated the pancreatic secretory response to CCK and to the CCK analogue cerulein. The results show that the acinar cells gradually became resistant to CCK stimulation after the onset of acute pancreatitis in several different animal models. This situation is an interesting state of hormone resistance that needs to be characterized by further experimental and clinical studies. At least in
GASTROENTEROLOGY
1126 NIEDERALJ ET AL.
the models of hemorrhagic pancreatitis, the CCK resistance is not caused by increased levels of circulating CCK, which remain normal in both the diet and taurocholate models (17). In cerulein-induced pancreatitis, one must consider a possible component of desensitization by cerulein to account for the decrease in responsiveness because both in vitro (43) and in vivo (44) attenuation of the secretory response have been documented with CCK pretreatment. However, the attenuation of the secretory response after pretreatment by CCK did not reach the level of an almost complete abolishment of secretory response as it was observed in the present models of pancreatitis. Previous studies have shown that CCK-receptor antagonists may ameliorate pancreatitis induced by a CDE diet (17) and by injections of taurocholate (45). According to the present results, it is unlikely that CCK agonists or CCK antagonists exert their deleterious or beneficial effects by an action on pancreatic secretion. Preliminary data from studies in dietinduced acute pancreatitis in mice suggest that CCK still binds to its receptor and stimulates intracellular biochemical events that result in an increase in oxygen consumption (46).Such binding and action are not followed by an adequate response in secretion but may, via the decrease in intracellular PO,, cause additional damage to the cell. The main defect that causes the decrease in secretory responsiveness appears to be located beyond the receptor at an intracellular level of stimulus-secretion coupling. Previous studies using acini isolated from patients with acute pancreatitis or from rats that had been treated with supramaximal doses of cerulein indicate that in addition to the secretory blockade, synthesis of proteins may also be reduced in acute pancreatitis (30,471. The impairment of protein synthesis may at least partly explain the secretory blockade. However, in other models of pancreatitis, protein synthesis of the pancreatic acinar cell has been shown to be almost normal (19). Thus, further experiments are needed to identify the exact intracellular location of the defect that results in the secretory blockade. There is only some information about whether pancreatic secretory function during acute pancreatitis is also resistant to stimuli other than CCK, such as cholinergic agents and secretin. Experimental data from cerulein-induced pancreatitis in the rat indicate that stimulation by secretin may not be reduced to the same extent as stimulation by CCK (48).This assumption is substantiated by a clinical report which showed that the secretory response to secretin was normal only 2 weeks after the beginning of acute pancreatitis in most patients ‘f49). However, in taurocholateinduced pancreatitis secretin failed to cause any increase in pancreatic secretion (23).Sensitivity to carba-
Vol. 99, No. 4
chol is reduced in cerulein-induced pancreatitis (301, but apparently not to the same extent as sensitivity to CCK. The pathophysiology of pancreatitis, both in the experimental and in the clinical situation, is still incompletely understood. Therefore, it is difficult to transfer observations made in a particular animal model to the human situation. Because of these difficulties, we have studied several models of pancreatitis that are quite different in terms of severity, time course, and pathogenesis. Despite these marked differences, the results, which show a secretory blockade during pancreatitis, are strikingly similar in all models. Therefore, the experimental data presented and the clinical evidence discussed suggest that pancreatic secretion, and in particular the pancreatic secretory response to CCK, is reduced during acute pancreatitis, certainly in the experimental setting and possibly also in the clinical situation. This secretory blockade might at least partly explain the failure to treat acute pancreatitis effectively by inhibition of secretion.
References 1. Fayne
SD, Barkin JS. Acute pancreatitis. Mt Sinai J Med 1986;53:396-403. 2. Corfield AP. Cooper MJ, Williamson RCN. Acute pancreatitis: a lethal disease of increasing incidence. Gut 1985;26:724-729. 3. Cameron JL, Mehigan IJ, Zuidema GD. Evaluation of atropine in acute pancreatitis. Surg Gynecol Obstet 1979;148:206-208. 4. Broe PJ. Zinner MJ, Cameron JL. A clinical trial of cimetidine in acute pancreatitis. Surg Gynecol Obstet 1982;154:13-16. 5. Loiudice TA, Lang J, Mehta H. Treatment of acute alcoholic pancreatitis. Am J Gastroenterol1984;79:553-558. 6. Meshkinpour H, Molinari M, Gardner L. Cimetidine in the treatment of acute alcoholic pancreatitis. Gastroenterology 1979; 77:687-690. 7. Regan PT. Malegelada JR, Go VLW, Wolf AM, DiMagno EP. A prospective study of the antisecretory effects of cimetidine and glucagon in human acute pancreatitis. Mayo Clin Proc 1981;56: 499-503. a. DUrr HK, Maroske D, Zelder 0, Bode J. Glucagon therapy in acute pancreatitis. Gut 1978;19:175-179. 9. Kronberg 0, Bulow S, Joergensen PN. A randomized doubleblind trial of glucagon treatment of first attack of severe acute pancreatitis without associated biliary disease. Am J Gastroenterol1983;73:423-425. 10.MRC multicentre trial of glucagon and aprotinin. Lancet 1977;2: 632-635. 11.Olazabal A, Fuller R. Failure of glucagon in the treatment of alcoholic pancreatitis. Gastroenterology 1978;74:489-491. 12.Goebell H, Ammann R, Herfarth C. A double-blind trial of synthetic salmon calcitonin in the treatment of acute pancreatitis. Stand J Gastroenterol1979;14:881-889. 13.Usadel KH, Uberla KK, Leuscher U. Treatment of acute pancreatitis with somatostatin (abstr]. Dig Dis Sci 1985;30:A992. 14.Steinberg WM. Schlesselmann SE. Treatment of acute pancreatitis. Gastroenterology 1987;93:1420-1427. 15.Evander A, Ihse I. Lundquist 1. Influence of hormonal stimulation by caerulein on acute experimental pancreatitis in the rat. Eur SurgRes198X13:257-268,
October 1990
PANCREATIC
16. Evander
36. Niederau C, Niederau M, Williams JA, Grendell JH. New proglumide analogue CCK-receptor antagonists. Am J Physiol 1986;251:G856-G860. 37. Duncan DB. Multiple range tests for correlated and heteroscedastic means. Biometrics 1957;13:164-176. 38. Mitchell CJ, Playforth MJ, Kelleher J, McMahon MJ. Functional recovery of the exocrine pancreas after acute pancreatitis. Stand J Gastroenterol1983:18:5-8. 39. Angelini G, Pederzoli P, Caliari S, Fratton S, Brocco G, Marzoli G, Bovo P, Cavallini G. Scuro LA. Long-term outcome of acute necrohemorrhagic pancreatitis. Digestion 1984;30:131-137. 40. Gullo L. Sarles H, Mott CB. Functional investigation of the exocrine pancreas following acute pancreatitis. Rend Gastroenterol1972;4:18-21. 41.Seligson U, Cho JW, Ihre G, Lundh G, Pyk E. Evaluation of the diagnosis of pancreatitis. Stand J Gastroenterol1982;17:905-911. 42. Bttchler M, Malfertheiner P, Block S, Maier W, Beger HG. Morphologlsche und funktionelle Ver8nderungen des Pankreas nach akuter nekrotisierender Pankreatltis. Z Gastroenterol1985; 23:79-83. 43.Abdelmoumene S, Gardener JD. Cholecystokinin-induced desensitization of enzyme secretion in dispersed acini from guinea pig pancreas. Am J Physiol1989;239:G272-G279. 44. Otsuki M, Williams JA. Amylase secretion by isolated pancreatic acini after chronic cholecystokinin treatment in vivo. Am J Physiol1983;244:G683-G688. 45. Makovec F, Bani M, Cereda R, Chiste R, Revel L, Rovati LC, Setnikar I, Rovati LA. Protective effect of CR1409 (choleystokinin antagonist] on experimental pancreatitls in rats and mice. Peptides 1986;7:1159-1164. 46. Bhattacharya I, Hootman S, Williams J, Grendell JH. Cholecystokinin-stimulated oxygen consumption by mouse pancreatic acinar cells in vitro in diet-induced acute pancreatitis (abstr). Gastroenterology 1988;94:A33. 47. Adler G, Kern HF. Fine structure and biochemical studies in human acute pancreatitis. In: Gyr KE, Singer MV, Sarles H. eds. Pancreatitis. Amsterdam: Elsevier, 1984:37-42. 48. Renner IG. Wisner JR, Lavigne BC. Partial restoration of pancreatic function by exogenous secretin in rats with ceruletideinduced acute pancreatitis. Dig Dis Sci 1986;31:305-313. 49. Tiscornia OM, Dreiling DA. Does the pancreatic gland regenerate? Gastroenterology 1966;51:267-271.
A, Lundquist I, Ihse I. Influence of gastro-intestinal hormones on the course of acute experimental pancreatitis. Hepatogastroenterology 1962;29:161-166. 17. Niederau C, Liddle RA, Ferrell LD, Grendell JH. Beneficial effects of CCK-receptor blockade and inhibition of proteolytic enzyme activity in experimental acute hemorrhagic pancreatitis in mice. J Clin Invest 1986;78:1056-1083. 18. Adler G. Hupp T, Kern HF. Course and spontaneous regression of acute pancreatitis in the rat. Virchows Arch [A] 1979:382:3247. 19. Gilliland L, Steer ML. Effects of ethionine on digestive enzyme synthesis and discharge by mouse pancreas. Am J Physiol 1980:239:G418-G426. 20. Kioke K, Steer ML, Meldolesi J. Pancreatic effects of ethionine: blockade of exocytosis and appearance of crinophagy and autophagy precede cellular necrosis. Am J Physiol 1982;242: G297-G307. 21. Saluja A. Saito I, Saluja M. Houlihan MJ, Powers RE, Meldolesi J. Steer ML. In vivo rat pancreatic cell function during supramaximal stimulation with caerulein. Am J Physiol 1985;249:G702G710. 22. Evander A, Hederstrom E. Hultberg B, Ihse I. Exocrine pancreatic secretion in acute experimental pancreatitis. Digestion 1982;24:159-167. 23. Grendell JH, Rothman SS. Digestive end products mobilize secretory proteins from subcellular stores in the pancreas. Am J Physiol1981;241:G67-G73, 24. Glazer G, and Steer ML. Requirements for activation of trypsinogen and chymotrypsinogen in rabbit pancreatic juice. Anal Biochem 1977;77:130-140. 25. Hummel BCW. A modified spectrophotometric determination of chymotrypsin, trypsin, and thrombin. Can J Biochem Physiol 1959;37:1393-1399. 26. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurements with Folin phenol reagent. J Biol Chem 195X193: 265-275. 27.Niederau C, Ferrell LD, Grendell JH. 1985. Caerulein-induced acute necrotizing pancreatitis in mice. Gastroenterology 1985;88: 1192-1204. 28.Niederau C, Ferrell LD, Liddle RA. Grendell JH. Biochemical, secretory, and morphological effects of repetitive experimental acute pancreatitis (abstr). Gastroenterology 1986;90:A1565. 29. Tsay DG, Niederau C. Schmidt H, Goldberg HJ. Higgins CB, Ferrell LD, Grendell JH. Experimental acute pancreatitis. Invest Radio1 1987;22:556-561. 30. Lampel M. Kern HF. Acute interstitial pancreatitls in the rat induced by excessive doses of a pancreatic secretagogue. Virchows Arch [A] 1987;373:97-117. 31. Niederau C, Ferrell LD, Grendell JH. Akute experimentelle Pankreatitis. Internist 1986;27:681-696. 32. Lombardi B, Estes LW, Longnecker DS. Acute hemorrhagic pancreatitis [massive necrosis) with fat necrosis induced in mice by DL-ethionine fed with a choline-deficient diet. Am J Pathol 1975:79:464-480. 33.Lankisch PG. Winckler K. Bokermann M, Schmidt H, Creutzfeldt W. The influence of glucagon on acute experimental pancreatitis in the rat. Stand J Gastroenterol1974;9:725-729. 34. Niederau C, Chang L, James TL, Grendell JH. In vivo 31P magnetic resonance (MR] spectroscopy in normal rat pancreas and in experimental acute pancreatitis (abstr]. Digestion 1986;35: A43. 35. Williams JA. Korc M, Dormer RL. Action of secretagogues on a new preparation of functionally intact pancreatic acini. Am J Physiol1978;235:E517-E524.
SECRETION
IN ACUTE
PANCREATITIS
1127
Received September 18.1989. Accepted April 9.1990. Address requests for reprints to: Claus Niederau, M.D., Medizinische Klinik und Poliklinik, Abteilung fur Gastroenterologie, Heinrich-Heine-Universitgt Dusseldorf, MoorenstraBe 5, 4000 DUsseldorf, Federal Republic of Germany. Dr. Claus Niederau was supported by grants from the Deutsche Forschungsgemeinschaft (Ni 224/1-l, 224/2-l, and 224/2-2) and from the Minister fur Wissenschaft und Forschung des Landes Nordrhein-Westfalen. Dr. James H. Grendell was supported by grants from the Research Evaluation and Allocation Committee, the Committee on Research of the Academic Senate, School of Medicine, University of California, San Francisco, and the National Institute of Health (DK 38939). This work was presented in part at the Annual Meeting of the European Pancreatic Club, Marseille, France, September 5,1987 (abstract published in Digestion 1987;38:46). and at the Annual Meeting of the American Pancreatic Association, Chicago, Illinois, November 5.1987 (abstract published in Dig Dis Sci 1987;32:1178). The authors thank Christine Genz and Monika Ebbert for expert technical assistance.