Intra-pancreatic release of bradykinin during the course experimental pancreatitis in rat

Kallikrein/Kininogn activation is an important pathophysiological event in acute pancreatitis, leading to microcirculatory changes within the gland. Hitherto, only indirect measurements of pancreatic bradykinin formation have been performed, monitoring the peptide in the circulation and in the peritoneal exudate. In the present study, intra-pancreatic bradykinin release was assessed using microdialysis during experimental acute pancreatitis in rat. In mild, oedematous pancreatitis, induced by caerulein hyperstimulation, the levels of bradykinin within the gland were not elevated compared with those of control rats. However, in necrotic pancreatitis, induced by retrograde injection of taurocholate into the pancreatic duct, significantly elevated levels of intraglandular bradykinin were seen. Several rats in this group died whilst in a state of circulatory shock.


Introduction
The purpose of this work was to elucidate whether the release of bradykinin in the pancre-Release of bradykinin has been demonstrated atic gland is more profuse during taurocholateduring acute pancreatitis and is proposed to con-induced pancreatitis in rat, often leading to tribute to the development of hypotension and shock, than during caerulein-induced pancreatitis.
shock. '2 Bradykinin is one of the most potent vasodilators in man, causing loss of intravascular Materials and Methods fluid by increasing the capillary permeability. It may also generate severe pain. Bradykinin is Animal experiments: Adult Wistar rats of both formed in blood through limited proteolysis of sexes, weighing 205-260g were obtained from the kininogens by activated plasma kallikrein. M611egaard Avelslab A/S (Skensved, Denmark). Kallidin (lysyl-bradykinin) is released from The animals were deprived of food but not of kininogens by glandular kallikrein and may be water for 18h before the experiments. Anaesconverted to bradykinin by amino peptidases in thesia was maintained by repeated injections of plasma. 4 In pancreatitis, trypsinogen is activated mebumal (barbital) into the peritoneal cavity. to trypsin in the pancreatic gland. 5'6 Trypsin may The body temperature was continuously monithen activate the pro-kallikreins and also directly tored and adjusted if necessary. A polyethylene release bradykinin from kininogen. 2 catheter (PE 90) was introduced into a femoral Direct measurement of activated bradykinin in vein. All animals received saline intravenously at the circulation is difficult due to the great capa-4 ml/h. The pancreas was exposed through a city of blood to both generate and degrade this mildline incision. Rats were randomly segregated peptide. 7'8 The kinins are rapidly inactivated by into three groups; caerulein-induced pancreatitis carboxypeptidases (kinase I and II) and have an (n 6), taurocholate-induced pancreatitis estimated half-life of about 20 s in the human cir-(n 8) and controls (n 6).
culation. 9 The plasma levels of bradykinin repor-Caerulein pancreatitis was induced by intraveted for man vary considerably when measured by nous infiasion of caerulein (Sigma, St Louis, MO, 10 11 bio-assay and radioimmunoassay. USA), 10 I,tg/kg/h for 6 h.
The microdialysis technique 12-15 offers the Taurocholate pancreatitis was induced by retpossibility of studying the bradykinin release rograde injection of sodium taurocholate (Sigma, directly in the interstitial fluid of the pancreas, by St Louis, MO, USA) into the pancreatic duct selecting probe membranes with a permeability system. The common bile and pancreatic duct allowing the passage of bradykinin (MW , 1 was cannulated trans-duodenally and the duct kDa) but not of the kinases (MW 45 kDa). was temporarily occluded proximal of the con-necting pancreatic ductuli by application of a vas-Statistics: Results are expressed as means cular clamp. Taurocholate diluted in saline, 0.8 (_S.E.M.). The Mann-Whitney U-test was used ml 4% (w/v), was injected into the common duct to determine possible differences between for 90 s and the clamp was removed, groups of rats with pancreatitis and the control All three groups were subjected to micro-group. A p value less than 0.05 was considered dialysis of the extracellular fluid of the pancreatic sigfiificant.
gland. A 10 mm long probe, with a diameter of 0.5 mm, was introduced into the central part of Results the pancreatic gland. The polycarbonate membrane had a molecular weight cut-off of less than All animals in the control group and in the 20 kDa. A micro-injection pump (CMA/100) and caerulein group sueeived the observation period a micro-sampler (CMA/200) were used (CMA/ of 6 h, while three of the animals in the tauro-Microdialysis AB, Stockholm, Sweden). Ringer's cholate group died. At autopsy, a severe inflamsolution (Baxter, Norway) was used as perfusion matory reaction was seen in the pancreatic gland fluid at a flow rate of 5 l.tl/min and 30 min fracin the taurocholate group with haemorrhage, tions were collected at + 4C in the fraction col-oedema and necrotic areas, while in the caerulein lector and stored at -70C until analysed, group, only modest oedema was seen, without Microdialysis was started when pancreatitis was bleeding or obvious necrosis. The microdialysis induced, procedure did not cause any significant inflam-Samples of venous blood were drawn at 0 h, matory reaction. The microscopic appearance of 2h and 6 h into EDTA-containing tubes. After the pancreas was essentially normal after 6 h of centrifugation, the plasma samples were stored at microdialysis (Fig. 1).
The concentration of bradykinin in the micro-All animals surviving the observation period dialysate from taurocholate rats showed a rapid were killed with an overdose of mebumal. The study was approved by the ethics committee of tund University.
Assays: Bradykinin was measured using a radioimmunoassay technique described previously. was incubated at + 4C for 18h. For the separation, 1 ml of the buffer containing 0.15% charcoal and 0.015% dextran P70 was added. After centrifugation at 2 500 x g for 15 min, 1 ml of the supernatant was transferred to a disposable plastic tube and counted for 20 min in a gamma scintillation counter. The sensitivity of the RIA, defined as the lowest standard producing a significan displacement of labelled antigen, was 10 lg/1. With the bradykinin cross-reaction set at 100%, the antiserum cross-reacted with kallidin to 97%, with metionyl-lysyl bradykinin to 42% and with des-arginine-bradykinin to 82%. Amylase was measured using the Phadebas Amylase Test increase, with a mean maximum value of 3.5 l.tg/1 1 hour after induction of the pancreatitis (Fig. 2).
The level was significantly increased for about 4 h compared with the control rats. The increase in bradykinin concentration in the dialysates from the animals with caerulein-induced pancreatitis was modest and did not differ significantly from that of the control group.
The caerulein group showed the most pronounced increase in amylase levels with a mean value of 450 t.tcat/1 after 6 h. In comparison, the mean value of the taurocholate group was 180 I.tcat/1 and the control group 70 t.tcat/1 after 6 h (Fig. 3).
The plasma concentration of t2-AP increased in all three groups, but the most pronounced increase was seen in the taurocholate group (Fig. 4).  C)). *p < 0.05 and **p < 0.01 indicate differences between control rats and rats with pancreatitis.

Discussion
The activation of kininogen is an important pathophysiological event in acute pancreatitis. Bradykinin antagonists may improve the outcome in experimental pancreatitis by reducing pancreatic oedema and hypotension. 7'8 Measurements of activated bradykinin have been performed in blood and peritoneal exudate, but not where the inflammatory process beginsin the pancreas. We have earlier adapted the microdialysis technique for the exocrine pancreas. '4 In this study we utilized the system for the evaluation of intrapancreatic bradykinin release during experimental pancreatitis. Introduction of the microdialysis probe into the pancreatic gland caused only a minor trauma to the tissue, as judged from the low initial bradykinin level in the dialysate. Furthermore, the microscopic appearance of the pancreatic tissue was normal after 6 h of microdialysis. Microdialysis has already been used to measure bradykinin in clinical studies, not concerned with pancreas, but after oral surgery. 15'19 The membrane of the probe does not induce bradykinin formation. 5 In this study, we used two different models of experimental pancreatitis. Hyperstimulation of the pancreatic gland with caerulein, a synthetic cholecystokinin analogue, produces a mild oedematous form of pancreatitis in 1-2 h which later 20 21 restitutes. This form of pancreatitis did not cause release of bradykinin into the interstitium of the pancreatic gland. Increased levels of bradykinin have recently been demonstrated in venous blood from rats with caerulein-induced pancreatitis. 22 Differences in methodology and types of test samples prelude detailed comparisons, but our results indicate that the bradykinin release in this type of pancreatitis is rather low, at least in comparison with the taurocholate model. On the other hand, the highly elevated serum amylase levels reflect hyperstimulation of the pancreas.
Injection of taurocholate into the pancreatic duct induces severe pancreatitis, as evidenced by the autopsy findings and the mortality in this group during the observation period. This form of experimental pancreatitis is connected with grave microcirculatory disturbances in the pancreatic land and, eventually, hypotension and shock.23, 4 The severity of the disease is also illustrated by the pronounced release of bradykinin within the gland, as evidenced by the high levels in the dialysate. These results are also in agreement with earlier, more indirect, findings indicating that the kinin release in acute pancreatitis was localized to the abdominal region, as illustrated by kininogen consumption in the peritoneal exudates. 2 The very rapid acute phase protein ai-AP showed much higher plasma levels in the taurocholate group than in the caerulein group and in the controls, reflecting the more severe inflammation in the taurocholate group.
In conclusion, microdialysis appears to offer a convenient method of monitoring the intra-pancreatic release of bradykinin and should also be applicable to the study of other peptides. Severe pancreatitis appears to be accompanied by an extensive release of bradykinin within the gland, probably contributing to the oedema and tissue damage seen in this disease.