International Journal of Pancreatology, 6 (1990) 071-079 $02.00 Copyright �9 1990 by The Humana Press Inc. All rights of any nature whatsoever reserved.

The Role of Ascites and Phospholipase on Peritoneal Permeability Changes in Acute Experimental Pancreatitis

A2

C o n n y Svens son , * R u n e SjOdahl , Inge la Li l ja , a n d I n g e m a r l h s e Department of Surgery, University Hospital, S-581 85, LinkOping, Sweden

Received February 2t, 1989; Revised July 3, 1989; Accepted July 13, 1989

Summary Peritoneal permeability to fluorescein-isothiocyanate-conjugated (FITC) dextran, tool wt 10,000 was studied in acute experimental pancreatitis (AEP) in rats. The aim of the study was to elucidate the role of the pan- creatitis ascites and its phospholipase A2 activity on the observed peritoneal permeability increase during AEP. Phospholipase A2 activity of ascites was 40 U/I~L 1 h after the induction of AEP and decreased during the next 3 h gradually to a plateau of about 20 U/t~L, where it remained to the end of the experiment at 24 h. A similar pattern was seen for protein, amylase, and lipase although the initial peak was most pronounced for lipase. Pan- creatitis ascites did not--irrespective of its age (1 or 20 h) or incubation time (3-20 h)--affect the peritoneal transport of FITC dextran 10,000 in healthy rats. Similarly, intravenously-administered ascites and intraperito- neal instillation of pancreatic phospholipase A2 dissolved in saline were without effects. On the other hand, in another group of healthy animals, phospholilpase dissolved in fresh pancreatitis ascites caused a statistically significant increase of peritoneal transport, as defined.

In rats with pancreatitis, the addition of phospholipase Az to the peri- toneal fluid increased peritoneal transport of FITC dextran 10,000 as well as phospholipase A2 itself. We conclude that phospholipase A2 when in- stilled into the peritoneal cavity in the presence of pancreatitis ascites, has the ability to increase peritoneal permeability to FITC dextran 10,000 in healthy, as well as in pancreatitis rats. However, the phospholipase A2 activity of rat pancreatitis ascites is not sufficient for this mechanism to work. This, however, does not exclude its existence in other species, in- cluding humans.

Key Words: Pancreatitis; peri toneal fluid; peritoneal permeabili ty; phospholipase A2.

*Author to whom all correspondence and reprint requests should be addressed.

International Journal of Pancreatology 71 Volume 6, 1990

72 Svensson et aL

INTRODUCTION

The systemic complications of, e.g., lungs and kidneys occurring in severe acute pancreatitis have been suggested to be related to the peritoneal fluid that is t)qgically present in this condition. The fluid is known to contain biologi- caUy-active agents, such as pancreatic amylase, lipase, phospholipase Az and proteases, complement factors, kinins, and lysolecithin, and its intravenous (1) or intraperitoneal (2) administration to healthy animals causes significant hemodynamic changes. Furthermore, pancreatic phospholipase A2 has ato tracted interest as a possible cause of the respiratory distress syndrome occur- ring in severe acute pancreatitis (3). In a previous study, we demonstrated an increased peritoneal permeability to FITC dextrans up to 20,000 dalton in rats with acute experimental pancreatitis (AEP) (4). In piglets with hemorragic pancreatitis, Lehtola and coworkers (5) recently reported early chemical peritonitis characterized by severe inflammation of the peritoneum with de- struction of the mesothelial cell layer, leading to denudation of the under- lying connective tissue. Identical findings were made after intraperitoneal administration of ascites from these piglets to healthy animals (5). The pres- ent study was done in order to evatuate the importance of the pancreatitis ascites, with special reference to its content of phospholipase A2, on perito- neal permeabiIity.

MATERIAL AND METHODS

Chemicals

Fluorescein-isothiocyanate-conjugated (FITC) dextran mol wt 10,000, was purchased from Sigma Chemical Company, St. Louis, MO. Porcine pan- creatic phospholipase A2 (tool wt 14,000) was also obtained from Sigma.

Laboratory Assays

Protein was determined according to the method described by Lowry (6) and amylase by use of the Phadebas test (Pharmacia Fine Chemicals, Uppsala, Sweden). Lipase was assayed titrimetrically using tributyrine as substrate (7) and phosphotipase A2 with the method described by Tagesson and Sj/Sdahl (8). Measurement of FITC dextran was done by emission spectrofluorometry (9).

Experimental Procedure

Male Wistar rats (200-250 g body wt) were anesthetized by an intraperito- heal injection of Ketalar | (Parke Davis and Co Inc., USA 80 mg/kg body wt) and Rompun | (Bayer, AG, Leverkusen, West German~y 8 mg/kg body wt). After opening the abdomen, the bile pancreatic duct was cannulated transduodenally by a thin polyethylene catheter. AEP was induced by intra- ductal injection of 0.2 mL 0.025M glycylglycin-NaOH buffer, pH 8.0, con- taining 0.02214 CaClz, 4/zmol sodiumtaurodeoxycholate (NaTDC), and 0.2 nmol trypsin. After removal of the catheter, the abdominal wall incision was closed and the animals were allowed to wake up.

International Journal o f Pancreatology Volume 6, 1990

Experimental Pancreatitis 73

Measurement of Peritoneal Permeability

In anesthetized animals, intraperitoneal deposition of 1 mL 3.33M FITC dextran i0,000 or 1 mL of a solution containing 1000 U phospholipase Az was done. Repeated blood samples for measurement of FITC dextran or the activity of PLA2, respectively, were taken from a jugular vein cannula.

Experimental Design

Experiment 1 AEP was induced in five rats, as described above, and a plastic catheter

(inner diameter 2 mm) was left through the upper midline incision for later collection of peritoneal fluid. After I, 2, 3, 4, 5, 7, and 24 h 0.5 mL samples were drawn from the catheter for determination of protein concentration and amylase, lipase, and phospholipase A2 activity.

Experiment 2 Three milliliters of pancreatitis ascites was administered intraperitonealty

to healthy rats and the passage of FITC dextran 10,000 to the blood stream measured as described above. This molecule was chosen since a previous study showed its transfer over the peritoneal membrane to be significantly increased in AEP (4). Ascites of different ages (obtained one or 20 h after induction of pancreatitis) and different incubation periods, i.e., the duration of the expo- sition of the ascites to the peritoneal surface before the instillation of the dextran (three or 20 h), were used. Controls were treated similarly with the exception that saline was used instead of ascites.

Experiment 3 One milliliter of a saline solution, with or without the addition of 1000 U

of phospholipase Az (600 U/rag protein at pH 8.0 and 37 ~ was adminis- tered intraperitoneally to healthy rats and the peritoneal passage of FITC 10,000 measured as described above. The choice of the phospholipase A2 dose was based on previous findings (4) and gives an intraperitoneal enzyme content about three times that which can be calculated from experiment 1 anticipating a mean peritoneal fluid vol of 5 mL. The incubation time was 10 min. A separate, but identical experiment was done with the exception that 1 mL of fresh pancreatitis ascites was used instead of saline for dissolution of the phospholipase A2.

Experiment 4 Two milliliters of ascites, obtained 3 h after induction of pancreatitis, were

injected intravenously to healthy rats. Peritoneal permeability to FITC dex- tran 10,000 was measured from 10 rain after the injection in the way described above. Controls had an intravenous injection of 2 mL of saline.

Experiment 5 One milliliter of a saline solution with or without the addition of 1000 U of

phospholipase A~ was administered intraperitoneally to pancreatitis rats 4 h after induction of the disease. Peritoneal passage of FITC dextran 10,000 was determined as described above.

International Journal o f Pancreat ology Volume 6, 1990

74 Svensson et al.

ck AMYLASE (!,~kat/Ixl0 3 )

X PROTEIN

PLA2 Q (U/~I)

I LIPASE O {gmoUmintmt}

5 -]-100

2-

1"

J I 2 3 4 5 7 2 4

T i m e ( h o u r s )

Fig. 1. Protein concentration and activities of amylase, lipase, and phospholi- pase A2 (PLA2) in pancreatitis ascites at different time intervals following induction of the disease. Mean+_ SEM; n = 5.

Experiment 6 One milliliter of a saline solution, with the addition of 1000 U of phospho-

lipase A2, was administered intraperitoneally to pancreatitis rats 4 h after induction of the disease and to healthy sham-operated rats, respectively. Control pancreatitis rats had intraperitoneal saline injections. Serum levels of phospholipase A2 were assayed immediately before and 20 min after the injection.

STATISTICAL CALCULATIONS

Statistical significances were determined with the use of the unpaired Stu- dent's t-test.

RESULTS

At microscopic examination, the rats had clear-cut signs of pancreatitis, such as necrosis, hemorrhage, thrombi, and leucocyte infiltration; there were no apparent differences between the groups.

The concentration of protein was slightly increased in pancreatitis ascites during the first 3 h after induction of AEP, compared to the later time inter- vals. The initial increase in pancreatic enzyme activities was even more pro- nounced, especially that of lipase (Fig. 1). Phospholipase A~ activity was elevated by 100~ after 1 h by comparison with subsequent samples at 3 h. There was no difference between samples taken at 3 or 24 h.

Interncttional Journal o f Pancreatology Volume 6, 1990

Experimental Pancreatitis

E O_ v

0 0 0

0

x rm

0 1-- ,7

200

150

100

5 0 -

0 0

A

t i i

10 2 0 3 0 m i n u t e s

?5

-5 E r v

0 0 0

0 ~T x s

0 F-- ,-F

200

1 O0 -

5 0 -

t i i

0 10 2 0 3 0 m i n u t e s

Fig. 2. Concentration of FITC dextran 10,000 in plasma following intraperito- neal deposition in healthy animals, which also had concomitant administration of pancreatitic ascites intraperitoneally. (A) Ascites, age 1 h. Exposition period 3 h (D []) n = 11 and 20 h ( ~ ~ ) n =4, respectively. Saline controls had a 3 h exposition period (X X) n = 11 (Mean _+ SEM). (B) Ascites, age 20 h. Expo- sition period 3 h ( i II). Saline controls (X X) n =8. (Mean +_ SEM).

In healthy rats, no change in peritoneal transfer of FITC dextran 10,000 was found after intraperitoneal administration of pancreatitis ascites, irre- spective of age of the ascites or incubation time (Fig. 2 A and B). Neither did intraperitoneal deposition of phospholipase A2 dissolved in saline influence the passage of FITC dextran 10,000 (Fig. 3A). On the other hand, a statisti- cally significant increase in peritoneal permeability, as defined, was found after intraperitoneal administration of phospholipase A2 dissolved in fresh pancreatitis ascites (Fig. 3B). Intravenous injection of ascites into healthy animals showed no effect (Table 1).

In pancreatitis rats, intraperitoneal phospholipase A2 administration caused a significant increase of peritoneal permeability to FITC dextran 10,000 at 10 and 20 min (Fig. 4), and much higher serum levels of phospholipase A2 were found in pancreatitis, compared to healthy animals after intraperitoneal ad- ministration of the enzyme (Fig. 5).

International Journal of Pancreatology Volume 6, 1990

76

5O

E o 40 8 CL

o 30 0 0

o 20 '7, x

10

0 F- Z. 0

A

1 0 2 0

minutes

Svensson et aI.

B 150 *

o

E 100 *

0

0

0 50

x s

0 V-- ~ . 0 ,

10 20 minutes

Fig. 3. (A) Concentration of FITC dextran 10,000 in plasma following intraperi- toneal deposition in healthy animals, which, in addition, had intraperitoneal admin- istration of phospholipase A2 dissolved in saline (E2 - - [2) n=6. Controls (X X) had saline only, n =4 (Mean + SEM). (B) An identical experiment to that described in Fig. 3A except that pancreatitis ascites was used instead of saline. Phospholipase group (U - - O ) , n = 5; control group (X ..... X), n ~ 6.

Table 1 Measurement of FITC-Dextran 10,000 in Plasma after iv Administration

of 2 mL Ascites (age 3 h) and ip Deposition of 1 mL 3.3M FITC-Dextran 10,000 (Mean + SEM) ~

Time, minutes

Group 5 10 15 20 30

A E P ( n = 4 ) 24.0_+3'."3 3'9.5+ 7.2 52.3+ 9.6 58.7+11 79.5+11.7 Control (n =2) t9.5 -+ 8.5 39.5 _+ 21.5 50.5 -+ 24.5 61.0 + 24.7 71.5 + 25.5

~FITC Dextran 10,000 (pmol/mL in plasma).

DISCUSSION

The results o f the present study show that neither ascites obtained f rom pancreatitis rats nor pancreatic phospholipase dissolved in saline caused any changes in peritoneal permeability to FITC dextran 10,000 when admin- istered intraperitoneally to healthy rats. However, when phospholipase A2 and pancreatitis ascites were combined in the same healthy animal, a perito-

International Journal o f Pancreatology Volume 6, 1990

Experimental PancreaEtis 77

8O

E 0 E 60

0 0 0 40

0

s 2O

0

t.l_ O' 0 10 20

minutes

Fig~ 4. Concentration of FITC dextran 10,000 in plasma following intraperito- neal deposition in pancreatitis animals 1000 U of phospholipase A2. AEP + PLA2 ( l - II) n=7. AEP controls (D 5 ) n=8. Mean_+SEM: **p<0.01.

c~ E 03 ~3

s c -

. ~

O4 <~ -J O_

500 -

400

300

200

100

[ ] AEP+PLA2 r - ' - - * * * [ ] AEP [ ] Normals+PLA2

0 20

" - - ' - I

"T-

minutes

Fig. 5. Activity of phospholipase A2 in plasma before and 20 min after intraperi- toneal deposition of 1000 U of phospholipase A2. AEP + PLAz, n = 7; AEP, n = 8; Normals + PLA2, n=8. Mean + SEM; ***p<0.001.

International Journal o f Pancreatology Volume 6, 1990

78 Svensson et al.

neal permeability increase, as defined, was obvious. Intravenously-injected ascites showed no effect in healthy animals. The addition of pancreatic phos- pholipase Az to the peritoneal fluid in pancreatitis rats caused an increase peritoneal permeability to the FITC dextran and seemingly also to phospho- lipase Az itself.

Our previous findings of an increased peritoneal permeability in acute pancreatitis (4) are consistent with the hypothesis that factors in the perito- neal fluid may be involved in the development of the systemic complications of the disease. However, the present findings show that the peritoneal fluid itself is unable to cause increased peritoneal permeability in healthy rats. Therefore, the permeability changes found in this species do not seem to be caused by local events only, but rather by systemic or a combination of local and systemic ones. The lack of influence on peritoneal permeability of intra- venously-administered pancreatitic ascites suggests that other systemic factors than those present in the ascites are decisive for the permeability changes. In the study of Lehtola et al. (5), marked peritoneal injury was found already 1 h after intraperitoneal injection of pancreatitis ascites to healthy rats. In accordance with the findings of the present study, these morphological changes do not appear to imply altered peritoneal passage for FITC dextran 10,000. Furthermore, Ellison et al. (2), in a study similar to one by Lehtola et al. (5), did not find any histological changes from intraperitoneal deposition of ascites.

Whereas the phospholipase activity present in pancreatitis ascites in rats does not seem sufficient to cause peritoneal permeability changes, the addi- tion of exogenous pancreatic phospholipase A2 to the ascites increased the peritoneal transport of FITC dextran 10,000. Also, in healthy animals, the peritoneal transport of the dextran molecule was increased if the phospholi- pase Az was administered together with fresh pancreatitis ascites. However, since phospholipase A2 alone in heathly animals was without influence, it seems as if still unknown factors present in the ascites are necessary for the enzyme to exert its effects on peritoneum. Future studies should focus on the critical phospholipase level for these effects, and comparison with the enzyme levels in peritoneal fluid of patients with acute pancreatitis are of interest.

The pathways of enzymes from the inflamed gland to the circulation are controversial. In a dog study, Mayer et al. (10) found that, although the majority of the released pancreatic enzymes accumulated within the peritoneal ascitic fluid, they did not appear to have a significant influence upon lymph or plasma concentrations of amylase or lipase. This is contrary to the findings of Egdahl (11) in a series of dog experiments that, whereas the initial rise of plasma enzyme concentrations are owing to direct enzyme transfer into the portal venous blood, the dominant route later in the course of the attack is via the thoracic duct lymph, and he suggested that this was owing to trans- peritoneal absorption of enzyme-rich fluid from the peritoneal cavity. Lange et al. (12), using a similar pancreatitis model as in the present study, found that hematogenous rather that the lymphatic transport of lipase directly from the pancreas and the peritoneal surface to be the most important pathways.

International Journal o f Pancreatology Volume 6, 1990

Experimental Pancreatitis 79

Our results support the peritoneal route as--at least--one option since we found a significantly higher passage of phospholipase A2 over peritoneum in rats with pancreatitis, compared to healthy controls.

ACKNOWLEDGMENTS

This study was supported by grants from the Swedish Medical Research Council (Project No. 17X03012) and from the County Council of Ostergtit- land, Sweden.

REFERENCES

1 Innes J, Fraser I, Carey LC. The vasoactive properties of ascitic fluid in acute pancreatitis in a porcine model. Arch. Surg. 1986; 121: 665-668.

2 Etlison EC, Pappas TN, Johnsson JA, Fabri P J, Carey LC. Demonstration and charac- terization of the hemo-concentrating effect of ascitic fluid that accumulate during haemorrhagic pancreatitis. J. Surg. Res. 1981; 30: 241-248.

3 Schr6der T. Respiratory failure in acute pancreatitis. Ann. Chir. Gynaecol. 1982; 71 suppl., t96: 27-30.

4 Svensson C, Sj6dahl R, Tagesson C, Ihse I. Increased peritoneal permeability in acute experimental pancreatitis. Int. J. Pancreatol. t989; 4: 83-90.

5 Lehtola A, Talja M, Nuutinen P, Nordling S, Lempinen M, Schr~Sder T. The effect of peritoneal exudate on peritoneal .morphology in experimental acute pancreatitis. Scand. J. Gastroenterol. 1986; 21: 1173-1 t 79.

6 Lowry OH, Rosebrough N J, Farr AL, Randall R J. Protein measurement with the folin phenol reagent. J. Biol. Chem. 1951; 193: 265-275.

7 Erlanson C, Borgstr~Sm B. Tributyrine as a substrate for determination of lipase activity of pancreatic juice and small intestinal content. Scand. J. Gastroenterol. 1970; 5: 293-295.

8 Tagesson C, Sj6dahl R. Rapid and sensitive assay of human plasma phospholipase A2 activity. Ann. Chir. Gyn. 1984; 78: 1-5.

9 Tagesson C, SjiSdahl R, Thordn B. Passage of molecules through the wall of the gastro- intestinal tract. A simple experimental model. Scand. J. Gastroenterol. 1978; 13: 519-524.

10 Mayer AD, Airey M, Hodgson J, McMahon MJ. Enzyme transfer from pancreas to plasma during acute pancreatitis. The contribution of ascitic fluid and lymphatic drainage of the pancreas. Gut 1985; 26: 876-881.

11 Egdahl RH. Mechanism of blood enzyme changes following production of experimental pancreatitis. Ann. Surg. 1958; 148: 389-400.

12 Lange JF, Beyaert P J, van Vugt H, Tytgat GNJ, van Gool J. Pathways of enzyme trans- fer in sodium taurocholate-induced acute haemorrhagic pancreatitis. Digestion 1986; 35: 229-236.

International Journal of Pancreatology Volume 6, 1990