Gastroenterologia Japonica Copyright 9 1992 by The Japanese SocieO, of Gastroenterology
VoL 27, No. 5 Printed in Japan
Influence of plasma proteinase inhibitors and the secretory leucocyte proteinase inhibitor on pancreatic elastase-induced degradation of some plasma proteins Hans-01of HAKANSSON and Kjell OHLSSON Department of Surgical Pathophysiology, University of Lund, Malm6 General Hospital, Malm6, Sweden Summary: Pancreatic elastase-induced degradation of some plasma proteins was studied in an in vitro model. The digestion was correlated with the degree of saturation of the a 1-proteinase inhibitor (alPI) and also with varying amounts of secretory leucocyte proteinase inhibitor (SLPI). SLPI was found to inhibit pancreatic elastase showing a Ki of about 10-7 M for the complex. On the addition of human pancreatic elastase to plasma cleavage of C3, kininogen, fibrinogen and fibronectin was observed when the a lPI approached saturation. In the present in vitro model it was possible to block the cleavage of the four plasma proteins, mentioned above completely with SLPI. Addition of the inhibitor also decreased the consumption of alPI. Gastroenterol Jpn 1992;27:652-656. Key words: arproteinase inhibitor," complement factor C3; fibronectin; kininogen; pancreatic elastase; proteinase inhibitors.
Introduction Pancreatic cationic elastase is considered to play a major role in the pathophysiology of acute pancreatitis. It has been held responsible for the vascular c o m p o n e n t of autodigestion 1. In h u m a n plasma alpha 1 proteinase inhibitor (alPI) and alpha 2 macroglobulin (a2M) represent the two major inhibitors of h u m a n pancreatic cationic elastase 2,3. Alpha 1-antichymotrypsin (ACHY) also contributes to the pancreatic elastase binding in h u m a n plasma 3,4. No local specific inhibitor of elastase is to our knowledge found in the pancreatic gland. Commercially available inhibitors such as aprotinin (Trasylol| 5 and gabexate mesilate (FOY) 6 are known to inhibit trypsin, but not pancreatic cationic elastase. The recently characterised secretory leucocyte proteinase inhibitor (SLPI) may offer new possibilities. SLPI is an acid-stable, low molecular weight proteinase inhibitot 7. This inhibitor was earlier studied under different names depending on from what source it was isolated; bronchial
m u c u s inhibitor, cervical mucus inhibitor, h u m a n seminal inhibitor I, antileucoproteinas, and m u c u s proteinase inhibitor, but the "different" inhibitors proved to be identical 8,9. It inhibits leucocyte elastase, chymotrypsin, catepsin G, mast cell chymase, tryptase, h u m a n pancreatic cationic elastase, anionic and cationic trypsin, but hot porcine cationic pancreatic elastase and neutrophil proteinase IV 1~ It thus seemed important to try elucidate the capacity of SLPI as a remedy in acute pancreatitis. In the present investigation we studied some effects of h u m a n pancreatic cationic elastase on h u m a n plasma proteins in vitro, and the capacity of h u m a n secretory leucocyte proteinase inhibitor (SLPI) to inhibit pancreatic elastase and protect plasma proteins from elastase digestion. Materials and Methods
Chemicals A garose lot 70058 was purchased from F M C Bio Products (Rockland, ME, USA). H u m a n
Received June 3, 1991. Accepted March 27, 1992. Address for correspondence: Prof. Kjell Ohlsson, Department of Surgical Pathophysiology, Maim5 General Hospital, Universityof Lund, S-214 01 MaimS, Sweden.
October 1992
Secretory leukocyte proteinase inhibitor
fibrinogen and elastin from bovine ligamentum nuchae was obtained from Kabi (Stockholm, Sweden) and Washington Biochem. Corp. (Freehold, N J, USA) respectively. Specific rabbit antisera against human a lPI, fibronectin, kininogen, C3 and C3c were available at our laboratory Human fibronectin was isolated on Gelatine-Sepharose 4B columns (Pharmacia, Uppsala, Sweden). Human pancreatic cationic elastase, 95% active, was produced in our laboratory a2. Recombinant human SLPI (rhSLPI) was obtained from Synergen Biologicals, Inc., (Boulder, CO, USA). Enzyme assay Fibronectin split products were measured by electroimmunoassay with one cm intermediate agarose gel containing 0.5% gelatin immediately anodal to the 10 ~tl wells. The top gel contained antiserum to fibronectin. Normal serum/plasma or purified fibronectin did not give any immunoprecipitates when analyzed in this system. Digestion of fibronectin with pancreatic elastase produced split products that did not bind to gelatin and they appeared as precipitate peaks in the top gel. The results were given as the peak height in mm. Fibrinolytic and elastolytic activity was determined with the agarose plate method as described12,~3. Elastase activity towards Succ-AlaAla-Pro-Leu-pNA (S-8511 Lot 66F-5840, Sigma Chemical Company, St. Louis, MO, USA) was assayed at 25~ at 405 nm a4. 5 mg Succ-Ala-AlaPro-Leu-pNa was dissolved in 10 ml 0.2M TrisHC1 buffer, 0.05CAC12 at pH 8.0 and 1 ml DMSA (dimethyl sulfoxide No. D-5879, Sigma) C3d was quantified with electroimmuno assay 15 and complexation of a lPI and the electrophoretic homogeneity of C3 and kininogen were studied using crossed immunoelectrophoresis. The precipitate for free a 1-PI in the various reaction mixtures was estimated by weighing paper copies 16. Reaction mixtures of human pancreatic elastase and human plasma. Human pancreatic cationic elastase in 0.01M HCI was added to a mixture of 25 ~1 human
653
plasma and 0.05M Tris-HCl buffer, NaC1 0.15M, pH 7.4. The final volume was 215 ~1. The final elastase concentrations used were 0-10 p.M. Following incubation at 37~ for 60 rain. the electrophoretic homogeneity of alPI, C3 and kininogen was analysed with crossed immunoelectrophoresis. The presence of C3d and fibronectin split products was determined by electroimmunoassay. Elastase activity in the reaction mixtures was assayed with elastin, fibrin and Succ-Ala-AlaPro-Leu-pNA as substrates. Reaction mixtures of human pancreatic cationic elastase and increasing concentrations of SLPI. For the titration of pancreatic elastase with rhSLPI constant amounts of enzyme (about 2 [.tM) were incubated with increasing amounts of inhibitor (0-100 l_tM). The reaction volumes were adjusted to 200 ~1 with 0.2M Tris-HC1 buffer, pH 8.0 and were incubated for 15 minutes at 25~ The residual free elastase in the reaction mixtures were determined after the addition of 200 ~1 substrate solutioia Succ-Ala-Ala-Pro-Leu-pNa. The amount of free enzyme was plotted against the concentration of the inhibitor. Reaction mixtures of human plasma, SLPI and pancreatic elastase. The effects of SLPI on pancreatic cationic elastase activity in plasma in vitro were studied by preincubating 25 ~1 plasma with varying amounts of SLPI (0-40 ~M) for 15 minutes at 37~ Thereafter pancreatic elastase was added to final concentration of 8 ~tM while stirring vigorously. The total reaction volume was adjusted to 215 ~1 using the Tris-HC1 buffer. After 60 minutes incubation at 37~ the reaction mixtures were analyzed as described above. All samples were duplicated. Four pairs of reaction mixtures were analyzed as described. Results were expressed as mean values (+ SEM). Results
In vitro effects of pancreatic elastase on plasma proteins. The addition of increasing amounts of human
654
H. Hgtkansson et aL
120
8OO
100 ,...
600
_
E
80
,~
40( E
.c 60 200
P 40
Vol. 27, No. 5
pancreatic cationic elastase to human plasma leads to the gradual saturation of plasma proteinase inhibitors. Close to saturation ofa 1PI at a concentration of elastase of about 7 ptM, fibronectin and kininogen split products and C3d appeared in the reaction mixtures and free fibrinolytic activity was seen (Figure 1). The area of free a lPI was then reduced to about 30% of the area seen in normal plasma on crossed immunoelectrophoresis. This value was not reduced on the addition of more elastase.
IJ.
20
9 0
2
4
Pancreatic
6
8
elastase
10
Titration of pancreatic elastase with rhSLPI Pancreatic elastase was titrated with SLPI and the best fit between the theoretical line and the data is provided with the Ki for the rhSLPIelastase complex set at 10-TM. The capacity of rhSLPI to protect plasma proteins from digestion by pancreatic elastase was tested using 8 ~M of active human pancreatic elastase. Fibrinolysis was inhibited and no fibronectin split products were observed at an SLPI concentration of 16 and 32 ~tM, respectively (Figure 2). No cleavage of C3 and kininogen as evidenced by crossed immunoelectrophoresis was seen at about 8.0 [~M SLPI. The area of free a 1PI showed a slow increase with the addition of increasing amounts of rhSLPI.
0 2
(gM)
2O
f~
10
i g.
Discussion 0
.2--~ Pancreatic
12 elastase
(gM)
Figure 1. (Upper)Humanpancreaticcationicelastasewasadded to human plasma. The final elastase concentrations used in the reaction mixtures were 0-10 ~M. The precipitates of free alPI in the various reaction mixtures were estimated by weighing paper copies and expressed as a percentage H ) of the weight of free al PI before the addition of elastase. The fibrinolytic activity in the reaction mixtures was determined by agarose plate method, and expressed in mm2 (O~O). (Lower) The presence of C3d ( H ) and fibronectin split products ( & - - A ) was determined by electroimmunoassay and the height of the immunoprecipitates was expressed in mm.
Based on experimental and clinical data pancreatic cationic elastase has been proposed to play a pathophysiological role in acute pancreatitis 1. The present study shows that a 1-PI as well as a2M have to be saturated with elastase before major degradation of other plasma proteins is seen after the addition of pancreatic elastase to fresh plasma. The cleavage of C3 and kininogen, seen in this in vitro model, can be inhibited by adding SLPI to a concentration of 5-8 ~M. This concentration also considerably reduces fibrinolysis and cleavage of fibronectin. The biologically effective elastase inhibiting capacity of plasma is considerably higher than the trypsin inhibiting capacity since trypsin-induced activation of the cascade systems already occurs
Secretory leukocyte proteinase inhibitor
October 1992
.-. 50
50
40 40 30
~
f
30
20 0
10 SLPI
2O 3O (~M)
40
10
8
0
,
0
lw
I
w
10
I
20
SLPI
| i
30
"
40
(~M)
Figure 2. (Upper) SLPI was mixed with human plasma, The final concentration of SLPI in the reaction mixtures were 0-40 p.M.After 15 minutes human pancreatic cationic elastase was added. The concentration of elastase was 8 ~M. The precipitates of free al PI in the various reaction mixtures were estimated by weighing paper copies and expressed as a percentage H ) of the weight of free al PI before the addition of elastase. The fibrinolytic activity in the reaction mixtures were determined by the agarose plate method, and expressed in mm2 (O~O). (Lower) The presence of C3d (H) and fibronectin split products ( A - - A ) was determined by electroimmunoassay and the height of the immunoprecipitates was expressed in mm.
after the saturation of a2M, that is after the addition of about 0.1-0.2 mg trypsin to one ml of
655
plasma 17'18. As for elastase-induced cleavage of plasma proteins, this occurs first after the addition of 0.8-0.9 mg of active elastase to one ml plasma. These differences are explained by the fact that the major proteinase inhibitor of plasma, a 1PI is a fast and efficient inhibitor of elastase, but only a relatively slow and biologically unimportant inhibitor of trypsin. Pancreatic elastase has not been demonstrated in a free form in plasma or peritoneal exudate, not even in severe pancreatitis. Active enzyme is, however, released in this disease and elastase complexed with a2M is present in the exudates of acute human pancreatitis 19 as well as in the exudates of rat 2~ and porcine experimental acute pancreatitis 2~. Active elastase has been demonstrated in experimental canine pancreatitis in exudates collected directly from the pancreatic gland 22 indicating a severe imbalance between proteinases and proteinase inhibitors locally in the gland. In addition elastolysis has been demonstrated in histologic sections of pancreatic tissues from experimental pancreatitis 23. Local proteolytic activity may be possible as the concentration of plasma proteinase inhibitors in the vicinity of, and within, the pancreatic gland probably is low, as a result of impaired local blood circulation in severe forms of acute pancreatitis 24. From a therapeutic point of view the broad proteinase inhibitor spectrum of SLPI could indicate a place for the inhibitor in the treatment of acute severe pancreatitis. However, while SLPI is a very strong inhibitor of human leucocyte elastase and of h u m a n chymotrypsin the activity against h u m a n trypsin and as shown here against human cationic elastase in only moderate 7'1~ This may be a problem from a possible therapeutic point of view as current data indicate that trypsin plays a key role in the pathophysiology of pancreatitis, at least in the initial phase of the diseaseY '26 On the other hand it is now clear that massive infiltration of PMN leucocytes in the late stages of pancreatitis is accompanied by a drastic release of leucocyte proteinases including elastase 27'28. SLPI may probably offer effective protection against pathologic leucocyte elastase digestion of important plasma proteins-including several factors of
656
H. H~tkansson et al.
the cascade systems as judged from the results of a recent study 29. In addition SLPI has recently been shown to be able to inhibit even leucocyte elastase bound to its elastin substrate 3~ The data presented thus indicate that rhSLPI may protect against the activity of human pancreatic elastase released during severe acute pancreatitis but to attain the concentrations needed, SLPI would probably have to be administered intraperitoneally or directly into the pancreatic duct. This study was supported by the Swedish Medical Research Council (3910), The medical Faculty of the University of Lund, the Albert Pfihlsson Foundation and the Torsten and Ragnar S6derberg Foundation.
References I. Geokas MC, Rinderknecht H, Swanson V, et al. The role of elastase in acute hemorrhagic pancreatitis in man. Laboratory investigation 1968;219:235-239. 2. Gustavsson EL, Ohlsson K, Olsson AS. Interaction between human pancreatic elastase and plasma protease inhibitors. HoppeSeyler's Z Physiol Chem 1980;361:169-176. 3. H~kansson HO, Ohlsson K. Interactions in vitro and in vivo between human and porcine cationic pancreatic elastase and plasma protease inhibitors. Biol Chem Hoppe-Seyler 1988;369:309-315. 4. Laine A, Davril M, Rabaud M, et al. Human serum alpha 1-antichymotrypsin is an inhibitor of pancreatic elastase. Eur J Biochem 1985;151:327-331. 5. Fritz H, Wunderer G. Biochemistry and applications ofaprotinin, the kallikrein inhibitor of bovine organs. Artzneim~ttel Forsch 1983;33:479-494. 6. Tamura Y, Hirado M, Okamura K, et al. Synthetic inhibitors of trypsin, plasmin, kallikrein, thrombin, Clr and C1 esterase. Biochem Biophys Acta 1977;484:417-422. 7. Thompson RC, Ohlsson K. Isolation, properties, and complete amino acid sequence of human secretory leucocyte protease inhibitor, a potent inhibitor of leucocyte elastase. Proc Natl Acad Sci USA1986;83:6692-6696. 8. Heinzel R, Appelhans H, Gassen HG. The Neutrophil ElastaseCathepsin G inhibitor of human mucus tissues and secretions (Antileukoprotease, HUSI-1) In: Taylor JC, Mittman CH, eds. Complete Primary structure as revealed by protein and DNA sequencing. New York: Academic Press, 1986;297-306. 9. Ohlsson K, Rosengren M, Stetler G. Structure, genomic organization, and tissue distribution of human secretory leucocyteprotease inhibitor (SLPI) In: Taylor JC, Mittman CH, eds. a potent inhibitor of neutrophil elastase. New York: Academic Press, 1986;307-324. 10. Schiessler H, Arnhold M, Ohlsson K, et al. Inhibitors of acrosin and granulocyte proteinases from human genital tract secretions. Hoppe-Seyler's Z Physiol Chem 1976;357:1251-1260.
Vol. 27, No. 5
11. Ohlsson K, Linder C, Rosengren M. Monoclonal antibodies specific for neutrophil proteinase 4. Hoppe-Seyler's Z Physiol Chem 1990;371:549-555. 12. Ohlsson K, Olsson AS. Purification and partial characterization of human pancreatic elastase. Hoppe-Seyler's Z Physiol Chem 1976;357:1153-1161. 13. Lassen M. Heat denaturation of plasminogen in the fibrin plate method. Acta Physiologica Scandinavica 1952;27:371-376. 14. Bieth J. In vivo significance of kinetic constants of macromolecular proteinase inhibitors. Biochem Med 1984;32:387-397. 15. Brandslund I, Teisner B, Hyltofl-Petersen P, et al. Development and clinical application of electroimmunoassays for the direct quantification of complement C3 split products C3c and C3d. Scand J Clin Lab Invest 1984;44 (Suppl 168):57-73. 16. Ganrot PO. Crossed immunoelectrophoresis. Scand J Clin Lab Invest 1972;29(Suppl 124):39-41. 17. Balldin A, Gustafsson EL, Ohlsson K. Influence of plasma protease inhibitors and trasylol on trypsin-induced bradykinin-release in vitro and in vivo. Eur Surg Res 1980;12:260-269. 18. Balldin A, Eddeland A, Ohlsson K. Studies on the role of the plasma protease inhibitors on in vitro C3-activation and in acute pancreatitis. Scand J Gastroenterol 1981;16:603-609. 19. H~.kansson HO. Studies on the possible role of pancreatic cationic elastase in acute pancreatitis. Thesis, University of Lund, Sweden 1991. 20. Satake K, Chung YS, Yoshimoto I", et al. Radioimmunoreactive serum elastase levels and histologic changes during experimental pancreatitis in rats. Arch Surg Res 1982;117:777-780. 21. H~lkansson HO, Borgstr6m A, Ohlsson K. Pancreatic cationic elastase in porcine experimental pancreatitis. Eur Surg Res 1991;23:73-84. 22. Ohlsson K, Eddeland A. Release of proteolytic enzymes in bileinduced pancreatitis in dogs. Gastroenterology 1975;369:668-675. 23. Schoenemann J. Elastase des Pankreas und ihre Inhibitoren. Georg Stuttgart: Thieme Verlag 1978;14:1-50. 24. Hjelmqvist B, Ohlsson K, Aronsen KF. Protease-antiprotease imbalance, hemodynamic and regional blood flow changes in experimental pancreatitis. Scand Gastroenterol 1986;21:8-11. 25. Ohlsson K, Olsson R, Bj6rk P. Local administration of human pancreatic secretory trypsin inhibitor prevents the development of experimental acute pancreatitis in rats and dogs. Scand J Gastroenterol 1989;24:693-704. 26. Ohlsson K, Balldin G, Bohe M, et al. Pancreatic proteases and antiproteases in pancreatic disease; biochemical, pathophysiological and clinical aspects. International Journal of Pancreatology 1988;3:67-78. 27. Lasson A, BaUdin G, Ohlsson K. Leucocyte elastase al-proteinase inhibitor complexes may diagnose pancreatic abscesses early. Scand J Gastroenterol 1986;21:221-224. 28. Bergenfeldt M, Bj6rk P, Ohlsson K, et al. Release ofimmunoreactive canine leucocyte elastase, normally and in endotoxin and pancreatic shock. Scand J Clin Invest 1990;50:35-42. 29. Bj6rk P, Axelsson L, Bergenfeldt M, et al. Influence of plasma protease inhibitors and the secretory leucocyte protease inhibitor on. Scand J Clin Invest 1988;48:205-211. 30. Laurent P, Rabaud M, Bieth JB. Inhibition of free and elastinbound human pancreatic elastase by human bronchial inhibitor. Biochemical pharmacology 1987;536:765-767.
VoL 27, No. 5 Printed in Japan
Influence of plasma proteinase inhibitors and the secretory leucocyte proteinase inhibitor on pancreatic elastase-induced degradation of some plasma proteins Hans-01of HAKANSSON and Kjell OHLSSON Department of Surgical Pathophysiology, University of Lund, Malm6 General Hospital, Malm6, Sweden Summary: Pancreatic elastase-induced degradation of some plasma proteins was studied in an in vitro model. The digestion was correlated with the degree of saturation of the a 1-proteinase inhibitor (alPI) and also with varying amounts of secretory leucocyte proteinase inhibitor (SLPI). SLPI was found to inhibit pancreatic elastase showing a Ki of about 10-7 M for the complex. On the addition of human pancreatic elastase to plasma cleavage of C3, kininogen, fibrinogen and fibronectin was observed when the a lPI approached saturation. In the present in vitro model it was possible to block the cleavage of the four plasma proteins, mentioned above completely with SLPI. Addition of the inhibitor also decreased the consumption of alPI. Gastroenterol Jpn 1992;27:652-656. Key words: arproteinase inhibitor," complement factor C3; fibronectin; kininogen; pancreatic elastase; proteinase inhibitors.
Introduction Pancreatic cationic elastase is considered to play a major role in the pathophysiology of acute pancreatitis. It has been held responsible for the vascular c o m p o n e n t of autodigestion 1. In h u m a n plasma alpha 1 proteinase inhibitor (alPI) and alpha 2 macroglobulin (a2M) represent the two major inhibitors of h u m a n pancreatic cationic elastase 2,3. Alpha 1-antichymotrypsin (ACHY) also contributes to the pancreatic elastase binding in h u m a n plasma 3,4. No local specific inhibitor of elastase is to our knowledge found in the pancreatic gland. Commercially available inhibitors such as aprotinin (Trasylol| 5 and gabexate mesilate (FOY) 6 are known to inhibit trypsin, but not pancreatic cationic elastase. The recently characterised secretory leucocyte proteinase inhibitor (SLPI) may offer new possibilities. SLPI is an acid-stable, low molecular weight proteinase inhibitot 7. This inhibitor was earlier studied under different names depending on from what source it was isolated; bronchial
m u c u s inhibitor, cervical mucus inhibitor, h u m a n seminal inhibitor I, antileucoproteinas, and m u c u s proteinase inhibitor, but the "different" inhibitors proved to be identical 8,9. It inhibits leucocyte elastase, chymotrypsin, catepsin G, mast cell chymase, tryptase, h u m a n pancreatic cationic elastase, anionic and cationic trypsin, but hot porcine cationic pancreatic elastase and neutrophil proteinase IV 1~ It thus seemed important to try elucidate the capacity of SLPI as a remedy in acute pancreatitis. In the present investigation we studied some effects of h u m a n pancreatic cationic elastase on h u m a n plasma proteins in vitro, and the capacity of h u m a n secretory leucocyte proteinase inhibitor (SLPI) to inhibit pancreatic elastase and protect plasma proteins from elastase digestion. Materials and Methods
Chemicals A garose lot 70058 was purchased from F M C Bio Products (Rockland, ME, USA). H u m a n
Received June 3, 1991. Accepted March 27, 1992. Address for correspondence: Prof. Kjell Ohlsson, Department of Surgical Pathophysiology, Maim5 General Hospital, Universityof Lund, S-214 01 MaimS, Sweden.
October 1992
Secretory leukocyte proteinase inhibitor
fibrinogen and elastin from bovine ligamentum nuchae was obtained from Kabi (Stockholm, Sweden) and Washington Biochem. Corp. (Freehold, N J, USA) respectively. Specific rabbit antisera against human a lPI, fibronectin, kininogen, C3 and C3c were available at our laboratory Human fibronectin was isolated on Gelatine-Sepharose 4B columns (Pharmacia, Uppsala, Sweden). Human pancreatic cationic elastase, 95% active, was produced in our laboratory a2. Recombinant human SLPI (rhSLPI) was obtained from Synergen Biologicals, Inc., (Boulder, CO, USA). Enzyme assay Fibronectin split products were measured by electroimmunoassay with one cm intermediate agarose gel containing 0.5% gelatin immediately anodal to the 10 ~tl wells. The top gel contained antiserum to fibronectin. Normal serum/plasma or purified fibronectin did not give any immunoprecipitates when analyzed in this system. Digestion of fibronectin with pancreatic elastase produced split products that did not bind to gelatin and they appeared as precipitate peaks in the top gel. The results were given as the peak height in mm. Fibrinolytic and elastolytic activity was determined with the agarose plate method as described12,~3. Elastase activity towards Succ-AlaAla-Pro-Leu-pNA (S-8511 Lot 66F-5840, Sigma Chemical Company, St. Louis, MO, USA) was assayed at 25~ at 405 nm a4. 5 mg Succ-Ala-AlaPro-Leu-pNa was dissolved in 10 ml 0.2M TrisHC1 buffer, 0.05CAC12 at pH 8.0 and 1 ml DMSA (dimethyl sulfoxide No. D-5879, Sigma) C3d was quantified with electroimmuno assay 15 and complexation of a lPI and the electrophoretic homogeneity of C3 and kininogen were studied using crossed immunoelectrophoresis. The precipitate for free a 1-PI in the various reaction mixtures was estimated by weighing paper copies 16. Reaction mixtures of human pancreatic elastase and human plasma. Human pancreatic cationic elastase in 0.01M HCI was added to a mixture of 25 ~1 human
653
plasma and 0.05M Tris-HCl buffer, NaC1 0.15M, pH 7.4. The final volume was 215 ~1. The final elastase concentrations used were 0-10 p.M. Following incubation at 37~ for 60 rain. the electrophoretic homogeneity of alPI, C3 and kininogen was analysed with crossed immunoelectrophoresis. The presence of C3d and fibronectin split products was determined by electroimmunoassay. Elastase activity in the reaction mixtures was assayed with elastin, fibrin and Succ-Ala-AlaPro-Leu-pNA as substrates. Reaction mixtures of human pancreatic cationic elastase and increasing concentrations of SLPI. For the titration of pancreatic elastase with rhSLPI constant amounts of enzyme (about 2 [.tM) were incubated with increasing amounts of inhibitor (0-100 l_tM). The reaction volumes were adjusted to 200 ~1 with 0.2M Tris-HC1 buffer, pH 8.0 and were incubated for 15 minutes at 25~ The residual free elastase in the reaction mixtures were determined after the addition of 200 ~1 substrate solutioia Succ-Ala-Ala-Pro-Leu-pNa. The amount of free enzyme was plotted against the concentration of the inhibitor. Reaction mixtures of human plasma, SLPI and pancreatic elastase. The effects of SLPI on pancreatic cationic elastase activity in plasma in vitro were studied by preincubating 25 ~1 plasma with varying amounts of SLPI (0-40 ~M) for 15 minutes at 37~ Thereafter pancreatic elastase was added to final concentration of 8 ~tM while stirring vigorously. The total reaction volume was adjusted to 215 ~1 using the Tris-HC1 buffer. After 60 minutes incubation at 37~ the reaction mixtures were analyzed as described above. All samples were duplicated. Four pairs of reaction mixtures were analyzed as described. Results were expressed as mean values (+ SEM). Results
In vitro effects of pancreatic elastase on plasma proteins. The addition of increasing amounts of human
654
H. Hgtkansson et aL
120
8OO
100 ,...
600
_
E
80
,~
40( E
.c 60 200
P 40
Vol. 27, No. 5
pancreatic cationic elastase to human plasma leads to the gradual saturation of plasma proteinase inhibitors. Close to saturation ofa 1PI at a concentration of elastase of about 7 ptM, fibronectin and kininogen split products and C3d appeared in the reaction mixtures and free fibrinolytic activity was seen (Figure 1). The area of free a lPI was then reduced to about 30% of the area seen in normal plasma on crossed immunoelectrophoresis. This value was not reduced on the addition of more elastase.
IJ.
20
9 0
2
4
Pancreatic
6
8
elastase
10
Titration of pancreatic elastase with rhSLPI Pancreatic elastase was titrated with SLPI and the best fit between the theoretical line and the data is provided with the Ki for the rhSLPIelastase complex set at 10-TM. The capacity of rhSLPI to protect plasma proteins from digestion by pancreatic elastase was tested using 8 ~M of active human pancreatic elastase. Fibrinolysis was inhibited and no fibronectin split products were observed at an SLPI concentration of 16 and 32 ~tM, respectively (Figure 2). No cleavage of C3 and kininogen as evidenced by crossed immunoelectrophoresis was seen at about 8.0 [~M SLPI. The area of free a 1PI showed a slow increase with the addition of increasing amounts of rhSLPI.
0 2
(gM)
2O
f~
10
i g.
Discussion 0
.2--~ Pancreatic
12 elastase
(gM)
Figure 1. (Upper)Humanpancreaticcationicelastasewasadded to human plasma. The final elastase concentrations used in the reaction mixtures were 0-10 ~M. The precipitates of free alPI in the various reaction mixtures were estimated by weighing paper copies and expressed as a percentage H ) of the weight of free al PI before the addition of elastase. The fibrinolytic activity in the reaction mixtures was determined by agarose plate method, and expressed in mm2 (O~O). (Lower) The presence of C3d ( H ) and fibronectin split products ( & - - A ) was determined by electroimmunoassay and the height of the immunoprecipitates was expressed in mm.
Based on experimental and clinical data pancreatic cationic elastase has been proposed to play a pathophysiological role in acute pancreatitis 1. The present study shows that a 1-PI as well as a2M have to be saturated with elastase before major degradation of other plasma proteins is seen after the addition of pancreatic elastase to fresh plasma. The cleavage of C3 and kininogen, seen in this in vitro model, can be inhibited by adding SLPI to a concentration of 5-8 ~M. This concentration also considerably reduces fibrinolysis and cleavage of fibronectin. The biologically effective elastase inhibiting capacity of plasma is considerably higher than the trypsin inhibiting capacity since trypsin-induced activation of the cascade systems already occurs
Secretory leukocyte proteinase inhibitor
October 1992
.-. 50
50
40 40 30
~
f
30
20 0
10 SLPI
2O 3O (~M)
40
10
8
0
,
0
lw
I
w
10
I
20
SLPI
| i
30
"
40
(~M)
Figure 2. (Upper) SLPI was mixed with human plasma, The final concentration of SLPI in the reaction mixtures were 0-40 p.M.After 15 minutes human pancreatic cationic elastase was added. The concentration of elastase was 8 ~M. The precipitates of free al PI in the various reaction mixtures were estimated by weighing paper copies and expressed as a percentage H ) of the weight of free al PI before the addition of elastase. The fibrinolytic activity in the reaction mixtures were determined by the agarose plate method, and expressed in mm2 (O~O). (Lower) The presence of C3d (H) and fibronectin split products ( A - - A ) was determined by electroimmunoassay and the height of the immunoprecipitates was expressed in mm.
after the saturation of a2M, that is after the addition of about 0.1-0.2 mg trypsin to one ml of
655
plasma 17'18. As for elastase-induced cleavage of plasma proteins, this occurs first after the addition of 0.8-0.9 mg of active elastase to one ml plasma. These differences are explained by the fact that the major proteinase inhibitor of plasma, a 1PI is a fast and efficient inhibitor of elastase, but only a relatively slow and biologically unimportant inhibitor of trypsin. Pancreatic elastase has not been demonstrated in a free form in plasma or peritoneal exudate, not even in severe pancreatitis. Active enzyme is, however, released in this disease and elastase complexed with a2M is present in the exudates of acute human pancreatitis 19 as well as in the exudates of rat 2~ and porcine experimental acute pancreatitis 2~. Active elastase has been demonstrated in experimental canine pancreatitis in exudates collected directly from the pancreatic gland 22 indicating a severe imbalance between proteinases and proteinase inhibitors locally in the gland. In addition elastolysis has been demonstrated in histologic sections of pancreatic tissues from experimental pancreatitis 23. Local proteolytic activity may be possible as the concentration of plasma proteinase inhibitors in the vicinity of, and within, the pancreatic gland probably is low, as a result of impaired local blood circulation in severe forms of acute pancreatitis 24. From a therapeutic point of view the broad proteinase inhibitor spectrum of SLPI could indicate a place for the inhibitor in the treatment of acute severe pancreatitis. However, while SLPI is a very strong inhibitor of human leucocyte elastase and of h u m a n chymotrypsin the activity against h u m a n trypsin and as shown here against human cationic elastase in only moderate 7'1~ This may be a problem from a possible therapeutic point of view as current data indicate that trypsin plays a key role in the pathophysiology of pancreatitis, at least in the initial phase of the diseaseY '26 On the other hand it is now clear that massive infiltration of PMN leucocytes in the late stages of pancreatitis is accompanied by a drastic release of leucocyte proteinases including elastase 27'28. SLPI may probably offer effective protection against pathologic leucocyte elastase digestion of important plasma proteins-including several factors of
656
H. H~tkansson et al.
the cascade systems as judged from the results of a recent study 29. In addition SLPI has recently been shown to be able to inhibit even leucocyte elastase bound to its elastin substrate 3~ The data presented thus indicate that rhSLPI may protect against the activity of human pancreatic elastase released during severe acute pancreatitis but to attain the concentrations needed, SLPI would probably have to be administered intraperitoneally or directly into the pancreatic duct. This study was supported by the Swedish Medical Research Council (3910), The medical Faculty of the University of Lund, the Albert Pfihlsson Foundation and the Torsten and Ragnar S6derberg Foundation.
References I. Geokas MC, Rinderknecht H, Swanson V, et al. The role of elastase in acute hemorrhagic pancreatitis in man. Laboratory investigation 1968;219:235-239. 2. Gustavsson EL, Ohlsson K, Olsson AS. Interaction between human pancreatic elastase and plasma protease inhibitors. HoppeSeyler's Z Physiol Chem 1980;361:169-176. 3. H~kansson HO, Ohlsson K. Interactions in vitro and in vivo between human and porcine cationic pancreatic elastase and plasma protease inhibitors. Biol Chem Hoppe-Seyler 1988;369:309-315. 4. Laine A, Davril M, Rabaud M, et al. Human serum alpha 1-antichymotrypsin is an inhibitor of pancreatic elastase. Eur J Biochem 1985;151:327-331. 5. Fritz H, Wunderer G. Biochemistry and applications ofaprotinin, the kallikrein inhibitor of bovine organs. Artzneim~ttel Forsch 1983;33:479-494. 6. Tamura Y, Hirado M, Okamura K, et al. Synthetic inhibitors of trypsin, plasmin, kallikrein, thrombin, Clr and C1 esterase. Biochem Biophys Acta 1977;484:417-422. 7. Thompson RC, Ohlsson K. Isolation, properties, and complete amino acid sequence of human secretory leucocyte protease inhibitor, a potent inhibitor of leucocyte elastase. Proc Natl Acad Sci USA1986;83:6692-6696. 8. Heinzel R, Appelhans H, Gassen HG. The Neutrophil ElastaseCathepsin G inhibitor of human mucus tissues and secretions (Antileukoprotease, HUSI-1) In: Taylor JC, Mittman CH, eds. Complete Primary structure as revealed by protein and DNA sequencing. New York: Academic Press, 1986;297-306. 9. Ohlsson K, Rosengren M, Stetler G. Structure, genomic organization, and tissue distribution of human secretory leucocyteprotease inhibitor (SLPI) In: Taylor JC, Mittman CH, eds. a potent inhibitor of neutrophil elastase. New York: Academic Press, 1986;307-324. 10. Schiessler H, Arnhold M, Ohlsson K, et al. Inhibitors of acrosin and granulocyte proteinases from human genital tract secretions. Hoppe-Seyler's Z Physiol Chem 1976;357:1251-1260.
Vol. 27, No. 5
11. Ohlsson K, Linder C, Rosengren M. Monoclonal antibodies specific for neutrophil proteinase 4. Hoppe-Seyler's Z Physiol Chem 1990;371:549-555. 12. Ohlsson K, Olsson AS. Purification and partial characterization of human pancreatic elastase. Hoppe-Seyler's Z Physiol Chem 1976;357:1153-1161. 13. Lassen M. Heat denaturation of plasminogen in the fibrin plate method. Acta Physiologica Scandinavica 1952;27:371-376. 14. Bieth J. In vivo significance of kinetic constants of macromolecular proteinase inhibitors. Biochem Med 1984;32:387-397. 15. Brandslund I, Teisner B, Hyltofl-Petersen P, et al. Development and clinical application of electroimmunoassays for the direct quantification of complement C3 split products C3c and C3d. Scand J Clin Lab Invest 1984;44 (Suppl 168):57-73. 16. Ganrot PO. Crossed immunoelectrophoresis. Scand J Clin Lab Invest 1972;29(Suppl 124):39-41. 17. Balldin A, Gustafsson EL, Ohlsson K. Influence of plasma protease inhibitors and trasylol on trypsin-induced bradykinin-release in vitro and in vivo. Eur Surg Res 1980;12:260-269. 18. Balldin A, Eddeland A, Ohlsson K. Studies on the role of the plasma protease inhibitors on in vitro C3-activation and in acute pancreatitis. Scand J Gastroenterol 1981;16:603-609. 19. H~.kansson HO. Studies on the possible role of pancreatic cationic elastase in acute pancreatitis. Thesis, University of Lund, Sweden 1991. 20. Satake K, Chung YS, Yoshimoto I", et al. Radioimmunoreactive serum elastase levels and histologic changes during experimental pancreatitis in rats. Arch Surg Res 1982;117:777-780. 21. H~lkansson HO, Borgstr6m A, Ohlsson K. Pancreatic cationic elastase in porcine experimental pancreatitis. Eur Surg Res 1991;23:73-84. 22. Ohlsson K, Eddeland A. Release of proteolytic enzymes in bileinduced pancreatitis in dogs. Gastroenterology 1975;369:668-675. 23. Schoenemann J. Elastase des Pankreas und ihre Inhibitoren. Georg Stuttgart: Thieme Verlag 1978;14:1-50. 24. Hjelmqvist B, Ohlsson K, Aronsen KF. Protease-antiprotease imbalance, hemodynamic and regional blood flow changes in experimental pancreatitis. Scand Gastroenterol 1986;21:8-11. 25. Ohlsson K, Olsson R, Bj6rk P. Local administration of human pancreatic secretory trypsin inhibitor prevents the development of experimental acute pancreatitis in rats and dogs. Scand J Gastroenterol 1989;24:693-704. 26. Ohlsson K, Balldin G, Bohe M, et al. Pancreatic proteases and antiproteases in pancreatic disease; biochemical, pathophysiological and clinical aspects. International Journal of Pancreatology 1988;3:67-78. 27. Lasson A, BaUdin G, Ohlsson K. Leucocyte elastase al-proteinase inhibitor complexes may diagnose pancreatic abscesses early. Scand J Gastroenterol 1986;21:221-224. 28. Bergenfeldt M, Bj6rk P, Ohlsson K, et al. Release ofimmunoreactive canine leucocyte elastase, normally and in endotoxin and pancreatic shock. Scand J Clin Invest 1990;50:35-42. 29. Bj6rk P, Axelsson L, Bergenfeldt M, et al. Influence of plasma protease inhibitors and the secretory leucocyte protease inhibitor on. Scand J Clin Invest 1988;48:205-211. 30. Laurent P, Rabaud M, Bieth JB. Inhibition of free and elastinbound human pancreatic elastase by human bronchial inhibitor. Biochemical pharmacology 1987;536:765-767.
