proteases and protease inhibitors in cerulein-induced acute pancreatitis in rats

iki

m7kam

eclttl4dgimcpwfa7

tc

u

3d

JA

0CA

Proteases and Protease Inhibitors in Cerulein-InducedAcute Pancreatitis in Rats

Peter Kruse, M.D.,*,1 Ake Lasson, M.D.,† and Esther Hage, M.D.‡

*Department of Surgical Gastroenterology D, Copenhagen University Hospital, Glostrup, and Department of Pharmacology,The Panum Institute, University of Copenhagen, Denmark; †Department of Surgery and Department of Experimental Research,

ournal of Surgical Research 85, 294–300 (1999)rticle ID jsre.1999.5609, available online at http://www.idealibrary.com on

Malmo University Hospital, Malmo, Sweden; and ‡Department of Pathology, Rigshospitalet, Copenhagen, Denmark

atio

tkCtouelmurnisousgeaidm

mtian

Submitted for public

Background. Proteases and protease inhibitors aremportant in acute pancreatitis (AP), although little isnown about the time course in cerulein-induced AP

n the rat.Materials and methods. AP was induced by supra-aximal stimulation of cerulein, 10 mg/kg/h, and during

2 h we measured lipase, amylase, albumin, prekalli-rein, factor X, a1-protease inhibitor, a1-macroglobulin,2-antiplasmin, antithrombin III (all in plasma) andacroscopic and histologic variables.Results. Within 12 h an edematous pancreatitis was

vident with peak values of peritoneal exudate, pan-reatic wet weight ratio, and plasma amylase andipase activities. Histologically, edema and vacuoliza-ion were prominent already after 3 and 6 h, respec-ively, while inflammation, necrosis, and total histo-ogical score gradually increase to reach peak levels at8 h. Proenzymes and most plasma protease inhibitorsecreased to low levels after 6–12 h followed by aradual increase. The sequential changes over timendicate that kallikrein - kinin activation, and plas-

inogen activation are probably early events inerulein-induced AP in rats. a1-Macroglobulin and a1-rotease inhibitor gradually decreased during thehole study period, probably being “second line” de-

ense inhibitors. Levels above normal were seen for2-antiplasmin and factor X at 48 h, normalizing at2 h.Conclusions. These results suggest that protease ac-

ivation and protease inhibitor consumption occur inerulein-induced AP in the rat. © 1999 Academic Press

Key Words: acute pancreatitis; animal models; cer-lein; enzyme precursors; protease inhibitors.

depp

1 To whom correspondence should be addressed at Stavangergade, 3. th, DK-2100 Copenhagen, Denmark. E-mail: [email protected].

294022-4804/99 $30.00opyright © 1999 by Academic Pressll rights of reproduction in any form reserved.

n October 22, 1998

INTRODUCTION

Pancreatic injury due to cholinergic hyperstimula-ion and to cholecystokinin (CCK) analogues is wellnown in various experimental models [1]. The use of aCK analogue, cerulein, a decapeptide isolated from

he skin of the frog, Hyla caerula, creates a stimulationf the pancreatic secretion. The administration of cer-lein in supramaximal doses creates pancreaticdema, elevation of pancreatic enzymes, and morpho-ogic changes of acute pancreatitis (AP) [2, 3]. All ani-

als survive, even when using very high doses of cer-lein, and the structural changes in pancreas are fullyeversible [1]. Cerulein is mostly given as intraperito-eal or subcutaneous injections, but being a peptide,

ts stability depends on many environmental factors,uch as pH and enzymatic activity. Therefore, the usef extravascular administration may not give a contin-ous hyperstimulation. Furthermore, environmentaltress, such as water immersion, can increase the de-ree of the pancreatic lesions, thus converting andematous AP to a hemorrhagic AP [4]. Therefore, tollow a steady intravenous administration and to min-mize potential stress caused by fixation of the animaluring infusion, the drug can be administered viaini-osmotic pumps.The pathophysiology of human and various experi-ental models of AP is still not fully understood. Ac-

ivation of pancreatic proteases is most probably onemportant key step in the pathophysiology of humannd various experimental animal models of AP, buteither the possible trigger(s) nor the key factor(s)
etermining the actual course and severity of the dis-ase are known for sure. There are, however, manylausible candidates, as recently reviewed [5]. The im-ortance of proteases and protease inhibitors in acute

pediihtsc[tttticwdisd[uwp

wiDwrwNnGpcaE1

Cpfpibv

rc(opplvra

c7pdopawwBpa

wa(mC

mpwDAfitTSiG(aaSfcctip1ro

pmnfs1ntOnghruswaas

CE

ancreatitis has earlier been shown in several species,.g., in mice [6], dogs [7], pigs [8], and man [9]. Suchata are, however, very limited in rat, although the rats probably the most commonly used animal in exper-mental studies of AP. Furthermore, the ceruleinormone-induced model as well as the sodium-aurocholate infusion model of AP, both used exten-ively in rats, are both considered to be two of the mostlinically relevant experimental models for human AP10, 11]. Thus, the aim of this study is to characterizehe time course of changes seen in proteases and pro-ease inhibitors in cerulein-induced AP in rats. Addi-ionally, sequential changes of the proteases and pro-ease inhibitors over time might indicate the order ofmportance of protease activation or protease inhibitoronsumption in this model. Such a characterizationould facilitate proper interpretation of various futureata on changes of individual proteases or proteasenhibitors in experimental models or interventionaltudies of AP in rats. As standard monitoring of theisease, we follow the recently published guidelines12]: pancreatitis indicator enzymes, macroscopic eval-ation, histologic changes, and mortality. Specifically,e also characterize the sequential changes of variouslasma proteins and protease inhibitors in plasma.

MATERIAL AND METHODS

Animals and materials. Adult male Wistar rats (Pan:WIST),eighing 222–261 g, were kept on aspen bedding (Tapvei) and single

n cages (Makrololen, Type III), fed pelleted rat diet (Altromin 1314,enmark), and subjected to regular 12-h light-dark cycles. The airas changed 12–14 times/h, the room temperature 21–23°C, and the

elative humidity 45–70%. The rats fasted 12 h before operation,ith free access to water. They were anesthetized with halothane/2O/O2 (1.5/50/50%). At the end of each operation, and repeated ifecessary, they received buprenorphine subcutaneously (Anorfin,EA, Copenhagen) 0.2 mg/kg body weight to relieve postoperativeain. The experimental procedures employed conform with the prin-iples and practice of the Danish law regulating experiments onnimals: “Lov om dyreforsøg” dated June 30, 1993. As a member ofU Denmark is bound by Directive 86/609/EEC dated November 24,986.Cerulein was purchased from Sigma Chemicals (St. Louis, MO;-9026), weighed, and dissolved in physiological saline adjusted toH 7.4 using ammonium hydroxide 0.01 M. Mini-osmotic pumpsrom Alzet were used (Model 2001D). Each pump was attached to aolyethylene tube, inner diameter 0.015 inch, outer diameter 0.030nch (Tygon, Norton Plastics, Akron, OH), and was quality-controlledy weighing the initially injected volume and the remaining endolume.

Experimental design and monitoring. Twenty-nine rats wereandomly allocated to seven groups: Control group (n 5 4) anderulein groups, T 5 3, 6, 12, 24, 48 (n 5 4 in each group), and 72-hn 5 5). One rat from the 72-h group was replaced before calculationf the results due to misplacement of the catheter. The rat waslaced supinely on a heating pad (body temperature 38.0 6 0.5°C), a

KRUSE, LASSON, AND HAGE:

refilled osmotic pump was placed subcutaneously in the midscapu-ar region, and the catheter was inserted into the external jugularein with the tip placed just above the heart. The control groupeceived osmotic pumps filled with physiologic pH-adjusted salinend was sacrificed after 6 h of infusion. The cerulein group received

a

sw

erulein 10 mg/kg/h dissolved in pH-adjusted physiologic saline. The2-h group was reanesthetized after 36 h to remove the osmoticumps in order to prevent pump leakage due to osmotic pressureamage. The 48- and 72-h groups received cerulein for only 24 h inrder to examine a possible spontaneous recovery. At the given timeoints the animals had a laparotomy and were macroscopically ex-mined. Peritoneal exudate was collected by absorbent cotton andeighed, pancreas wet weight was recorded, and pancreatic edemaas expressed as pancreatic wet weight gram/100 g body weight.lood was drawn from heart, the animals were sacrificed, and theancreas was quickly removed, freed of fat and major lymph nodes,nd weighed.

Chemical methods. Blood was kept at 0°C for a maximum of 1 h,hereafter the plasma was separated and stored at 270°C. Lipasend a-activities were determined in plasma by two commercial kitsLipase MPR1 and a-amylase EPS, Boehringer/Mannheim, Ger-

any), using a spectrophotometer (UV-2100, UV-VIS, Shimadzuorp., Kyoto, Japan).Albumin and the plasma protease inhibitor a1-macroglobulin wereeasured with electroimmunoassay [13], using specific antisera pre-

ared in the laboratory. The proenzymes prekallikrein and factor Xere assayed using the chromogenic peptide substrates S-2302 (H--Pro-Phe-Arg-pNA 2HCl) and S-2337 (Bz-Ile-Glu(Piperidyl)-Gly-rg-pNA), respectively (Chromogenix, Molndal, Sweden [14]). The

unctional activities of the plasma protease inhibitors C1-esterasenhibitor, a2-antiplasmin, antithrombin III, and a1-protease inhibi-or were also measured using chromogenic peptide substrates [14].he assay methods given by the manufacturer (Chromogenix, Molndal,weden) for humans could be used unchanged for rat C1-esterase

nhibitor (S-2302: see above), antithrombin III (S-2765: N-a-Cbo-Arg-ly-Arg-pNA 2HCl), and a1-protease inhibitor, (S-2677: Boc-Glu-

OBzl)Gly-Arg-pNA HCl), while they had to be partly modified for rat2-antiplasmin. Using the microtiter plate assay method for a2-ntiplasmin, the diluted samples were incubated with the substrate- 2251 (H-D-Val-Leu-Lys-pNA 2HCl) for 30 sec and with plasmin

or 7 min. The standard curve for heparinized EDTA as well as foritrated rat plasma turned out to be much more flat than the humanounterparts, almost horizontal between 50 and 125%, when usinghe proposed sample dilution of 1/40. Using a sample dilution of 1/80nstead resulted in a straight and steep standard curve also for ratlasma for 0–75%, whereafter the curve was slightly bent for 75–25%. Since these conditions gave the most precise and reproducibleesults out of several tested modifications, they were used through-ut the study.All values are given as percentage of normal in a reference plasma

ool of 10 normal rats, of the same weight and age as the experi-ental animals. By using specific antisera and electroimmunoassay,ormal plasma levels in rats are 35 g/L for albumin [15] and 3.7 g/Lor a1-macroglobulin [16, 17]. By using chromogenic peptide sub-trates, normal plasma levels for the other plasma proteins are all00 6 20% of the reference standard pool. An earlier study gives theormal level 1–2 g/L for a1-protease inhibitor and 0.4 g/L for anti-hrombin III [17] and normal a2-antiplasmin levels are 0.8 g/L [18].therwise earlier studies give the plasma proteins as percentage oformal pools. No exact levels are in fact given in the literature inrams per liter for the other proteins (e.g., the C1-esterase inhibitoras never been purified in rats). Instead we have compared severalat standard pools in parallel with several human plasma pools. Bysing exactly the same buffers and methods a comparison of theame plasma dilutions in rat and man gives parallel standard curvesith nearly the same values obtained for both species, except for2-antiplasmin where rat has 20–30% higher levels than man, inccordance with Karges et al. [19]. Furthermore, we have earlierhown that these proteins have the same molecular weights in rat

295RULEIN AND PANCREATITIS

nd in man [20].

Macroscopic and histological scoring. Extrapancreatic fat necro-es and pancreatic findings such as edema, hemorrhage, and necrosisere all graded each from 0 to 3. The possible range of a total

mcisfiscgtEttc

eamtu(

paw0ae(fmrra(

mfraighlea7taru

I

I

P(

V(

l7(

2 CH

acroscopic score is from 0 to 12. Each pancreas was pinned on aork plate, fixed 4 days in 10% neutral buffered formalin, and cutnto a series of 3-mm-thick sections numbered from 1 to 10. Theections were dehydrated in graded alcohol and embedded in paraf-n. For light microscopic examination 4-mm-thick sections weretained with hematoxylin-eosin. The histologic changes in the pan-reas were examined and graded by a blinded (i.e., no knowledge ofroups) pathologist specialized in gastroenterology. We used an es-ablished semiquantitative scoring system as seen in Table 1 [21].ach section could obtain a possible range of histologic score from 0

o 14. For each rat the total histological mean score of four sections,wo from the duodenal part and two from the splenic part, was thenalculated.Statistics. The data were expressed as means 6 SEM. Differ-

nces among groups were compared by the nonparametric one-waynalysis of variances on the ranks, Kruskal-Wallis ANOVA test forultiple groups [22]. A P value less than 0.05 was considered sta-

istically significant. Data handling and statistics were performedsing the statistical software package STATISTICA, version 5.0StatSoft Inc.).

RESULTS

There was no lethality in any group. All pumpsassed the quality control and delivered the expectedmount of drug. Plasma amylase and lipase activitiesere significantly different among the groups (P ,.001), both peaking at 12 h with a 32-fold rise formylase and a 109-fold rise for lipase. At 24 h, bothnzyme activities had returned near to control valuesFig. 1). The peritoneal exudate was significantly dif-erent among the groups (P 5 0.0024), having a maxi-

um at 12 h. At 48 h the peritoneal exudate had

TABLE 1

Histological Scoring of Cerulein Model

Histological pattern Assessment and (score)

nterstitial edema None (0)Mild (1)Moderate (2)Severe (3)

nflammatory infiltration None (0)Mild (1)Moderate (2)Severe (3)

arenchymal necrosis None (0)percentage of cells involved) , 5% (1)

5–25% (2)25–50% (3).50% (4)

acuolization None (0)percentage of cells involved) ,5% (1)

5–25% (2)25–50% (3).50% (4)

96 JOURNAL OF SURGICAL RESEAR

eturned to near control values. Pancreatic wet weightatio peaked in the 6- to 12-h groups (P 5 0.0005) andt 24 h it returned to a level near that of the controlsFig. 2).

eatW

All animals responded macroscopically with maxi-al edema in the 3- to 24-h groups and extrapancreatic

at necrosis appeared at 12–72 h. There were no mac-oscopic signs of pancreatic hemorrhage or necrosis inny of the groups (data not shown). The total histolog-cal scores were significantly different among theroups (P 5 0.0003) with a continuous rise until 24–72(Fig. 3). Analysis of the subgroups in the total histo-

ogical mean score is shown in Fig. 4: Edema was anarly phenomenon, as the highest levels were seenlready in the 3-h group followed by a decline toward2 h. The same pattern was seen regarding vacuoliza-ion. Inflammatory infiltration and necrosis showednother pattern. Scores of inflammatory infiltrationeached peak values at 48 h. Scores of necrosis contin-ed to increase until 48–72 h.

FIG. 1. Supramaximal cerulein stimulation in rats: Plasma amy-ase (■) and lipase (h) activity in the controls (Co) and in the 3- to2-h groups (n 5 4 in each). Each point represents the mean 6 SEMP , 0.001 for both enzymes, Kruskal-Wallis ANOVA).

: VOL. 85, NO. 2, AUGUST 1999

FIG. 2. Supramaximal cerulein stimulation in rats: Peritonealxudate (h) and pancreatic wet weight ratio (■) in the controls (Co)nd in the 3- to 72-h groups (n 5 4 in each). Each point representshe mean 6 SEM (P 5 0.0024 and P 5 0.0005, respectively, Kruskal-

allis ANOVA).

i5tnmtwctdt7tbtec(dP

lswprtmdiAm

tisfitrci

catciiftraavwtihpA7awtaa

hg0

CE

The plasma levels of albumin did not change signif-cantly during the 72-h study period (P 5 0.33, Fig.A). The proenzyme prekallikrein decreased to 69% ofhe normal plasma pool at 6 h, and then returned toormal (P 5 0.035, Fig. 5A). The other proenzymeeasured, factor X, decreased slightly to 89 and 91% of

he normal plasma pool at 3 and 12 h, respectively,hereafter levels above normal were reached, espe-

ially at 48 h (P 5 0.034, Fig. 5A). Most plasma pro-ease inhibitors decreased (Fig. 5B). a1-Macroglobulinecreased gradually during the whole study period, tohe lowest value of 54% of the normal plasma pool at2 h (P 5 0.0024). a2-Antiplasmin decreased to 76% ofhe normal plasma pool at 12 h, whereafter levelsecame above normal at 24 and 48 h, and then re-urned toward normal at 72 h (P 5 0.001). The C1-sterase inhibitor decreased to very low levels, espe-ially at 12 h where the value was 30% of the normalP 5 0.047). a1-Protease inhibitor and antithrombin IIIid not change significantly over time (P 5 0.38 and5 0.054, respectively).

DISCUSSION

Intravenous infusion of cerulein in maximal physio-ogical doses (0.25 mg/kg/h in rats) induces increasedecretory activity and degranulation of the pancreasith a reduction of the enzyme stores, but no rise inlasma enzymes [3]. The increased protein synthesiseturns to normal after 48–72 h [3] and no degenera-ive changes occur in the exocrine pancreas. Supra-aximal stimulation, however, results in a pancreatic

FIG. 3. Supramaximal cerulein stimulation in rats. The totalistological mean score in the controls (Co) and in the 3- to 72-hroups (n 5 4 in each). Each point represents the mean 6 SEM (P 5.0003, Kruskal-Wallis ANOVA).


amage with edema, elevation of pancreatic enzymesn plasma, and morphological changes characteristic ofP [3]. The mechanism of how cerulein in supramaxi-al doses results in AP is still unclear, although many

s}

(r

heories have been proposed, such as cerulein interact-ng with low-affinity CCK receptors on the acinar cellurface mediating an inhibition of enzyme secretionrom the pancreas; increase in cathepsin B and trypsinn the zymogen granule-enriched fraction; “colocaliza-ion” of digestive zymogens with lysosomal enzymesesulting in intracellular enzyme activation; and in-reased cellular, lysososmal, and mitochondrial fragil-ty [3].

The histological scoring system we used was suffi-ient to detect the early morphological changes, edema,nd vacuolization (Fig.4) [21, 23]. Similar vacuoliza-ion has been found in ultrastructural studies on pan-reatic acinar cells in human AP [24]. Inflammatorynfiltration was noted 9 h after initiation of cerulein-nduced pancreatitis in mice [25], and we found mani-est inflammatory infiltration at 12 h. The initiationime and the extent of acinar cell necrosis have beeneported very differently, from less than 5% of thecinar cells necrotic at any time point, beginning at 4 hfter start of infusion [21], to necrosis after 12 h in-olving up to 60% of the acinar cells [26]. In our studye found 30–40% necrosis at 12 h with a drop to less

han 25% necrosis at 72 h (Fig. 4). Both inflammatorynfiltration and necrosis in the pancreas are seen inuman AP. Although all the rats had “tubular” com-lexes, i.e., duct tubular structures also seen in humanP [27], we found no fibrosis and no complications after2 h. Cerulein-induced AP in rats is fully reversiblend no lethality has been reported [1]. After 3 daysith high connective tissue turnover, an increased mi-

otic activity in the pancreas is detected [28]. Pancre-tic tissue regenerates completely, and within 14 daysfter induction of AP, both morphological and biochem-cal changes have disappeared [21]. As we only fol-owed the animals 48 h after cessation of cerulein in-


il

FIG. 4. Supramaximal cerulein stimulation in rats: Histologicalcores divided in subgroups: h, edema (0–3); ‚, inflammation (0–3);, parenchymal necrosis (0–4); F, vacuolization (0–4) in the controls

Co) and in the 3- to 72-h groups (n 5 4 in each). Each pointepresents the mean 6 SEM.

fp

pplliptttlktaictkvpkilk[tad[b[mp

rTiwi[[otair

gnaapmapardspiafiap

( h).P al ci asmc nt r0

2

usion, the time span was not long enough to see a totalancreatic regeneration.During the early phase of the AP the measured

roenzymes prekallikrein and factor X, and most of thelasma protease inhibitors, decreased, to reach theowest levels after 6–12 h (Figs. 5A and 5B). The lowevels found for prekallikrein and C1-esterase inhibitorndicate proteolytic activation of prekallikrein androtease-antiprotease complexation or destruction ofhe C1-esterase inhibitor. The time lag of 6 h betweenhe lowest levels found for prekallikrein (at 6 h) andhe C1-esterase inhibitor (at 12 h) also indicate proteo-ytic activity and protease inhibitor consumption. Pre-allikrein is the proenzyme in the kallikrein-kinin sys-em and also an important proenzyme in contact-phasectivation [29]. The C1-esterase inhibitor is the mainnhibitor in both the kallikrein-kinin system and theomplement system as well as of contact-phase activa-ion [30]. Thus, we cannot from our study definitelynow if one, two, or all these three systems are acti-ated, although the earlier and quite extensive drop inrekallikrein levels indicates that activation of theallikrein-kinin system is an early event in cerulein-nduced AP in the rat. Shimizu et al. also found highevels of bradykinin, the active end product of theallikrein-kinin system, in cerulein-induced AP in rats31] indicating activation of the kallikrein-kinin sys-em. The complement system is probably also activatedccording to earlier studies, since complement-eprived rabbits did not develop severe pancreatitis32]. A more recent study, however, could not show any

FIG. 5. (A) Supramaximal cerulein stimulation in rats. Plasma leF) in the controls (Co) and in the 3- to 72-h groups (n 5 4–5 in eac

5 0.034, respectively, Kruskal-Wallis ANOVA). (B) Supramaximnhibitors a1-macroglobulin (■), a1-protease inhibitor (h), a2-antiplontrols (Co) and in the 3- to 72-h groups (n 5 4–5 in each). Each poi.054, P 5 0.047, respectively, Kruskal Wallis-ANOVA).


enefit at all of complement receptor type 1 (sCR1)33]. a2-Antiplasmin, the important inhibitor of plas-

in, also decreased quite extensively in this earlyhase, indicating early plasminogen activation, with

ocaa

esultant plasmin activity, in cerulein-induced AP.hese two findings of an early decrease of the C1-

nhibitor and of a2-antiplasmin are also in agreementith our own recent findings in rats with taurocholate-

nduced AP [34], intestinal ischemia and reperfusion35], and zymosan-induced multiple organ failure20]. Thus, kallikrein-kinin activation, and plasmin-gen activation are probably early events in severerauma in rats. In fact various (inflammatory) medi-tors have been implicated to play early and/or latemportant pathophysiological roles in AP, as recentlyeviewed [5].During the later phase of the AP most of the proteins

radually increased (Figs. 5A and 5B). Levels aboveormal were seen for both a2-antiplasmin and factor Xt 48 h, whereafter these high levels dropped to bebout normal at 72 h. This is probably due to an acutehase production of these proteins during the inflam-atory trauma caused by the pancreatitis, and a2-

ntiplasmin has earlier been shown to be an acutehase protein in the rat [36]. Our findings indicate thatlso factor X behaves as an acute phase protein in theat. a1-Macroglobulin, instead, gradually decreaseduring the whole study period. a1-Protease inhibitorhowed a late, but not significant, decrease. Since a1-rotease inhibitor is said to be an acute phase proteinn the rat [17], levels above normal would be expectedt 48–72 h, as e.g., seen for a2-antiplasmin. Thus, ourndings indicate protease-antiprotease complexationnd elimination of both a1-macroglobulin and a1-rotease inhibitor during AP [37]. This later decrease

ls of albumin (■) and the proenzymes prekallikrein (h) and factor XEach point represents the mean 6 SEM (P 5 0.33, P 5 0.035, anderulein stimulation in rats: Plasma levels of the plasma proteasein (‚), antithrombin III (E), and C1-esterase-inhibitor (F) in theepresents the mean 6 SEM (P 5 0.0084, P 5 0.38, P 5 0.0010, P 5

: VOL. 85, NO. 2, AUGUST 1999

ve

CH

f both a1-protease inhibitor and a1-macroglobulin, inomparison with the early decrease of the C1-inhibitornd a2-antiplasmin, indicates a1-protease inhibitornd a1-macroglobulin to be “second line” defense inhib-

ithii

((rttbtpegtptclsmhaA

asNrpTeAwtswttcaska

u1wHatiel

ttMm

tFM

1

1

1

1

1

CE

tors, although their much higher plasma levels makehis conclusion somewhat more uncertain. The ratherigh and unchanged levels found for antithrombin III

ndicate that thrombin activation is not important dur-ng cerulein-induced AP in rats.

Recent studies have shown a very early increasewithin a few h) in trypsinogen activation peptideTAP) levels in various models of acute pancreatitis inat and mice [38], indicating that activation ofrypsinogen occurs. Activation of rat pancreatic elas-ase is probably also of main importance, since it coulde a primary candidate for the pulmonary complica-ions seen during experimental AP in rats [39]. Theseancreatic proteases as well as chymotrypsin are pref-rably bound to rat a1-macroglobulin, a2-macro-lobulin, and a1-protease inhibitor, and to a lesser ex-ent also to a1-inhibitor I3 [16, 40]. Activation of theseancreas specific proteases, together with probable ac-ivation of other proteolytic enzymes of various plasmaascade systems, could well explain the low inhibitoryevels found during AP, by way of consumption or de-truction [41] of the protease inhibitors. Especially a1-acroglobulin, which has a wide inhibitory capacity,as been shown to inhibit mast cell tryptase, probablylso of importance for the lung complications seen inP [42].Our results of proteolytic activity in rats are in

greement with earlier studies on pancreatitis in otherpecies, e.g., in mice [6], dogs [7], pigs [8], and man [9].evertheless, data on various plasma proteins in the

at are very limited [19] and few studies have assessedroteases and protease inhibitors in rats [38, 43–47].he few studies done found TAP levels as a proof ofarly activation of proteolytic pancreatic proteases inP [48], and TAP has been shown to correlate ratherell with severity and mortality [6]. Protease inhibi-

ors have been studied twice earlier in the more severeodium taurocholate-induced pancreatitis in rats, butith somewhat contradictory data using two different

ime periods [46, 47]. Our study, however, indicateshat both protease activation and protease inhibitoronsumption occur in cerulein-induced AP in the rat,nd that most of these changes are reversible. Theequential changes over time indicate that kallikrein-inin activation, and plasminogen activation are prob-bly early events in cerulein-induced AP in rats.In conclusion, supramaximal stimulation with cer-

lein creates an edematous pancreatitis in rats with2-h peak values of peritoneal exudate, pancreatic weteight ratio, and plasma amylase and lipase activities.istologically, edema and vacuolization are prominentlready after 3 and 6 h, respectively, while inflamma-


ion, necrosis, and total histological score graduallyncrease to reach peak levels at 48 h. The findings of anarly decrease in proenzyme levels, together with aater decrease in the levels of various protease inhibi-

1

ors, indicate protease activation and protease inhibi-or consumption in cerulein-induced AP in the rat.ost of the measured variables returned toward nor-al baseline levels at 48–72 h.

ACKNOWLEDGMENTS

The authors thank Anders Fuglsang and Dorte Hansen for theirechnical assistance. This work is supported in part by The Påhlssonoundation, The Medical Faculty of Lund University, and Thealmo General Hospital Foundation.

REFERENCES

1. Gorelick, F. S., Adler, G., and Kern, H. F. Cerulein-inducedpancreatitis. In W. Go (Ed.), The Pancreas: Biology, Pathobiol-ogy and Diseases. New York: Raven, 1993. Pp. 501–526.

2. Bieger, W., Martin-Achard, A., Bassler, M., and Kern, H. F.Studies on intracellular transport of secretory proteins in therat exocrine pancreas. IV. Stimulation by in vivo infusion ofcaerulein. Cell Tissue Res. 165: 435, 1976.

3. Willemer, S., Elsasser, H. P., and Adler, G. Hormone-inducedpancreatitis. [Review]. Eur. Surg. Res. 24(Suppl. 1): 29, 1992.

4. Takano, S., Kimura, T., Kawabuchi, M., Yamaguchi, H., Kinjo,M., and Nawata, H. Ultrastructural study of the effects of stresson the pancreas in rats. Pancreas 9: 249, 1994.

5. Kingsnorth, A. Mediators in acute pancreatitis. Scand. J. Gas-troenterol. Suppl. 31: 1, 1996.

6. Rao, K. N., Tuma, J., and Lombardi, B. Acute hemorrhagicnecrosis in mice. Intraparenchymal activation of zymogens, andother enzyme changes in pancreas and serum. Gastroenterology70: 720, 1976.

7. Ohlsson, K., Olsson, R., Bjork, P., Balldin, G., Borgstrom, A.,and Lasson, A., et al. Local administration of human pancreaticsecretory trypsin inhibitor prevents the development of exper-imental acute pancreatitis in rats and dogs. Scand. J. Gastro-enterol. 24: 693, 1989.

8. Lium, B., Ruud, T. E., Pillgram-Larsen, J., Stadaas, J. O., andAasen, A. O. Sodium taurocholate-induced acute pancreatitis inpigs. Pathomorphological studies of the pancreas in untreatedanimals and animals pretreated with high doses of corticoste-roids or protease inhibitors. Acta Pathol. Microbiol. Immunol.Scand. A 95: 377, 1987.

9. Lasson, A., and Ohlsson, K. Acute pancreatitis. The correlationbetween clinical course, protease inhibitors, and complementand kinin activation. Scand. J. Gastroenterol. 19: 707, 1984.

0. Buchler, M., Friess, H., Uhl, W., and Beger, H. G. Clinicalrelevance of experimental acute pancreatitis. Eur. Surg. Res.24(Suppl. 1): 85, 1992.

1. Schoenberg, M. H., Buchler, M., and Beger, H. G. Experimentalmodels in acute pancreatitis. In B. Jeppsson (Ed.), AnimalModelling in Surgical Research. Amsterdam: Harwood Aca-demic 1998. Pp. 135–142.

2. Friess, H., Weber, A., and Buchler, M. Standards in monitoringacute experimental pancreatitis. Review. Eur. Surg. Res.24(Suppl. 1): 1, 1992.

3. Laurell, C. B. Electroimmunoassay. Scand. J. Clin. Lab. Invest.29(Suppl. 124): 21, 1972.

4. Gallimore, M. J., and Friberger, P. Chromogenic peptide sub-


strate assay and their clinical applications. Blood Rev. 5: 117,1991.

5. Matsuzawa, T., Nomura, M., and Unno, T. Clinical pathologyreference ranges of laboratory animals. J. Vet. Med. Sci. 55 (3):351, 1993.

1

1

1

1

2

2

2

2

2

2

2

2

2

2

3

3

3

3

3

3

3

3

3

3

4

4

4

4

4

4

4

4

3 CH

6. Gauthier, F., Genell, S., Mouray, H., and Ohlsson, K. The invitro interactions of rat pancreatic elastase and normal inflam-matory rat serum. Biochim. Biophys. Acta 526: 218, 1978.

7. Koj, A., Magielska-Zero, D., Kurdowska, A., and Bereta, J.Proteinase inhibitors as acute phase reactants: Regulation ofsynthesis and turnover. Adv. Exp. Med. Biol. 240: 171, 1988.

8. Kumada, T., and Abiko, Y. Physiological role of alpha 2-plasmininhibitor in rats. Thromb. Res. 36(2): 153, 1984.

9. Karges, H. E., Funk, K. A., and Ronneberger, H. Activity ofcoagulation and fibrinolysis parameters in animals. Arzneim.Forsch. Drug Res. 44(1): 793, 1994.

0. Deng, X., Sun, Z., Lasson, Å., Wang, X., and Andersson, R.Alterations in the functions of the reticuloendothelial andprotease-antiprotease systems after intraperitoneal injectionsof Zymosan in rats. Eur. J. Surg. 164: 605, 1998.

1. Tani, S., Otsuki, M., Itoh, H., Fujii, M., Nakamura, T., and Oka,T., et al. Histologic and biochemical alterations in experimentalacute pancreatitis induced by supramaximal caerulein stimu-lation. Int. J. Pancreatol. 2: 337, 1987.

2. Siegel, S., and Castellan, N. J. Nonparametric Statistics for theBehavioral Sciences. New York: McGraw-Hill, 1988.

3. Adler, G., Rohr, G., and Kern, H. F. Alteration of membranefusion as a cause of acute pancreatitis in the rat. Dig. Dis. Sci.27: 993, 1982.

4. Adler, G., and Kern, H. F. Fine structural and biochemicalstudies in human acute pancreatitis. In K. E. Gyr, M. V. Singer,and H. Sarles (Eds.), Pancreatitis—Concepts and classification.Amsterdam: Elsevier Science, 1984. Pp. 37–42.

5. Silverman, M., Ilardi, C., Bank, S., Kranz, V., and Lendvai, S.Effects of the cholecystokinin receptor antagonist L-364,718 onexperimental pancreatitis in mice (see comments). Gastroenter-ology 96: 186, 1989.

6. Schoenberg, M. H., Buchler, M., Gaspar, M., Stinner, A.,Younes, M., and Melzner, I., et al. Oxygen free radicals in acutepancreatitis of the rat. Gut 31: 1138, 1990.

7. Willemer, S., and Adler, G. Histochemical and ultrastructuralcharacteristics of tubular complexes in human acute pancreati-tis. Dig. Dis. Sci. 34: 46, 1989.

8. Elsasser, H. P., Adler, G., and Kern, H. F. Fibroblast structureand function during regeneration from hormone-induced acutepancreatitis in the rat. Pancreas 4: 169, 1989.

9. Cochrane, C. G., and Griffin, J. H. The biochemistry and patho-physiology of the contact system of plasma. Adv. Immunol. 33:241, 1982.

0. Wachtfogel, Y. T., DeLa Cadena, R. A., and Colman, R. W.Structual biology, cellular interactions and pathophysiology ofthe contact system. Thromb. Res. 72: 1, 1993.

1. Shimizu, I., Wada, S., Okahisa, T., Kamamura, M., Yano, M.,and Kodaira, T., et al. Radioimmunoreactive plasma bradykininlevels and histological changes during the course of cerulein-induced pancreatitis in rats. Pancreas 8: 220, 1993.

2. Seelig, R., and Seelig, H. P. The possible role of serum comple-ment system in the formal pathogenesis of acute pancreatisis. I.Immunopathogenetic pancreatitis—local Schwartzman-Sana-relli phenomenon—endotoxin induced pancreatitis. Acta Hepa-


togastroenterol. (Stuttg.) 22: 263, 1975.3. Weiser, M. R., Gibbs, S. A., Moore, F. D., Jr., and Hechtman,

H. B. Complement inhibition by soluble complement receptortype 1 fails to moderate cerulein-induced pancreatitis in the rat.Int. J. Pancreatol. 19(2): 129, 1996.

4

4. Kruse, P., Hage, E., and Lasson, Å. Proteases and proteaseinhibitors in taurocholate-induced acute pancreatitis in rats.Int. J. Pancreatol., 1999, in press.

5. Sun, Z. W., Wang, X., Deng, X., Lasson, Å., Wallen, R., andHallberg, E., et al. The influence of intestinal ischemia andreperfusion on bidirectional intestinal barrier permeability, cel-lular membrane integrity, protease inhibitors and cell death inrats. Shock, 1999, in press.

6. Hogstorp, H. J., Jacobsson, H., and Carlin, G. Effect of hepa-tectomy on the post-traumatic fibrinolysis inhibition and theprimary fibrinolysis inhibitor in the rat. Thromb. Res. 18: 361,1980.

7. Bergsma, J., Boelen, K. M., Duursma, A. M., Schutter, W. G.,Bouma, J. M. W., and Gruber, M. Complexes of rat alpha-1-macroglobulin and subtilisin are endocytosed by parenchymalliver cells. Biochem. J. 226: 75, 1985.

8. Foitzik, T., Lewandrowski, K. B., Fernandez-del Castillo, C.,Rattner, D. W., and Warshaw, A. L. Evidence for extraluminaltrypsinogen activation in three different models of acute pan-creatitis. Surgery 115: 698, 1994.

9. Lungarella, G., Gardi, C., de Santi, M. M., and Luzi, P. Pulmo-nary vascular injury in pancreatitis: evidence for a major roleplayed by pancreatic elastase. Exp. Mol. Pathol. 42: 44, 1985.

0. Gauthier, F., Genell, S., Mouray, H., and Ohlsson, K. Interac-tions in vitro and in vivo between rat serum protease inhibitorsand anodal and cathodal rat trypsin and chymotrypsin. Bio-chim. Biophys. Acta 566: 200, 1979.

1. Brower, H. S., and Harpel, P. C. Proteolytic cleavage and inac-tivation of a-2-plasmin inhibitor and C1 inactivator by humanpolymorphonuclear leukocyte elastase. J. Biol. Chem. 257:9849, 1982.

2. Tsuji, A., Akamatsu, T., Nagamune, H., and Matsuda, Y. Iden-tification of targeting proteinase for rat alpha-1-macroglobulin.Biochem. J. 298: 79, 1994.

3. Schmidt, J., Rattner, D. W., Lewandrowski, K., Compton, C. C.,Mandavilli, U., and Knoefel, W. T., et al. A better model of acutepancreatitis for evaluating therapy. Ann. Surg. 215: 44, 1992.

4. Largman, C., Reidelberger, R. D., and Tsukamoto, H. Correla-tion of trypsin-plasma inhibitor complexes with mortality inexperimental pancreatitis in rats. Dig. Dis. Sci. 31: 961, 1986.

5. Merriam, L. T., Wilcockson, D., Samuel, I., and Joehl, R. J.Ligation-induced acute pancreatitis increases pancreatic andcirculating trypsinogen activation peptides. J. Surg. Res. 60:417, 1996.

6. Rydzewska, G., Kosidlo, S., Gabryelewicz, A., and Rydzewski,A. Tissue plasminogen activator, plasminogen activator inhib-itor, and other parameters of fibrinolysis in the early stages oftaurocholate acute pancreatitis in rats. Int. J. Pancreatol. 11:161, 1992.

7. Gabryelewicz, A., Kosidlo, S., Prokopowicz, J., and Podkowicz,K. Does heparin modify protease-antiprotease balance in acuteexperimental pancreatitis in rats. Hepatogastroenterology 33:79, 1986.

: VOL. 85, NO. 2, AUGUST 1999

8. Schmidt, J., Fernandez-del Castillo, C., Rattner, D. W., Le-wandrowski, K., Compton, C. C., and Warshaw, A. L.Trypsinogen-activation peptides in experimental rat pancreati-tis: Prognostic implications and histopathologic correlates. Gas-troenterology 103: 1009, 1992.

proteases and protease inhibitors in cerulein-induced acute pancreatitis in rats

Documents