Urinary Trypsin Inhibitor an Fxpenment Al Arid Clinical Study

Home , Acrosin, Pancreatic elastase

Organization Document name LUND UNIVERSITY DOCTORAL DISSERTATION Departments of Surgical Pathophysiology Date of issue and Anaesthesiology 1991-10-30 Malmo General Hospital S-214 01 MAL MO, SWEDEN CODEN. LUM£DW/(HtKP-1019) 1-104(1991) Author* s) Sponsoring organinc ion ^^K^^^ Britt-Marie Berling

Title and subtitle

Urinary Trypsin Inhibitor an Fxpenment al arid Clinical Study

Abstract The urinary trypsin inhibitor (UTI) is an acid stable proleinase inhibitor present in blood and urine. It was purified from urine using affinity chromatography, ion exchange chromatography and gel nitration. Two forms of UTI were present in urine, A and B. Inhibitor A liad a molecular weight of 44 kDa and inhibitor B 20 kDa as judged from sodium dodecyl sulphate polyacrylamid gel electrophoresis. After reduction of inhibitor A its molecular weight was 20 kDa. This finding together with the ammo acid analysis of the two inhibitors suggest that inhibitor A is a dimer of inhibitor B. Specific anusera against the inhibitors were raised in sheep and rabbits. There was an immunological cross reaction between the plasma inter-a-trypsin inhibitor and inhibitor A and B. Both inhibitor A and inhibitor B inhibited trypsin and human leukocyte elastase. Complexes between the inhibitor A and human trypsin or human leukocyte elastase were produced and were found to be stable. When human serum was added elastase and trypsin dissociated from the inhibitor and associated with ctj-proteinase inhibitor and c^-macroglobulin. A radioimmunoassay for measurement of UTI ir. urine and plasma was performed. The plasma and serum samples were treated with perchloric acid to precipitate the acid labile, immunological cross reacting inter-a-trypsin inhibitor before analysis. The normal level of UTI in plasma and serum was about 2 mg/1. The normal excretion in urine was about 8mg per 24 hours. The plasma and urine levels of UTI were studied in patients with acute pancreatitis and in patients undergoing cholecystectomy. The plasma levels showed an increase during these conditions but never to a level higher than twice the normal value. In contrast there was a marked elevation of plasma C-reacti ve protein, indicating that UTI probably is not an acute phase protein. The urine levels of UTI were elevated postoperative!y and during the course of pancreatitis. This elevation correlates well with the excretion of aj-microglobulin suggesting a tubular proteinuria. Uremic patients had a marked increase of UTI in plasma compatible with decreased glomerular filtration. In samples from healthy persons as well OQ as from patients only inhibitor A was found. Inhibitor B has recently been renamed bikunin because of its two Kumtz-type inhibiting domains. Inhibitor A might be called tetrakunin. Radioactively labeled UTI (inhibitor A) was injected intravenously in three male volunteers. The plasma half-life of 125j UTJ was 2 hours. Free biologically active inhibitor was found in the urine during the first four hours after injection. The organ distribution of intravenously injected ^5j UJJwa s studied in ratS-Fifieen minutes after injection the major part of the radioactivity was found in the kidneys, suggesting that the kidneys are the primary site of UTI metabolism. Using immunohisiochemica! techniques UTI was found in the proximal tubules o~ the normal human kidney further indicating the tubular reabsorption and metabolism of UTI.

Urinary trypsin inhibitor, bikunin, tetrakunin, human leukocyte elastase, plasma proteinase inhibitors, cholecystectomy, acute pancreatitis, uremia, radioimmunoassay, turn-over

Classification system and/or index tenni (if any)

Supplementary bibliofraphical information Language Lnqljnh

ISSN and key title ISBN

Recipient's notes Number of pages ] Q4 Price

Security classification

Distribution by (name and address)

I, tbe undersigned, bcinf tbe copyright owner of the abstract of the above-mentioned dissertation, hereby (rant to all reference sources permission to publish and disseminate the abstract of the above-mentioned dissertation.

V Signature Date 1991-08-28 r From the Departments of Surgical Pathophysiology and Anaesthesiology, Lund University Malmö General Hospital, Malmö, Sweden

Urinary Trypsin Inhibitor an Experimental and Clinical Study

Britt-Marie Berling

Malmö 1991 © 1991 Britt-Marie Berling Printed in Sweden Studentlitteratur Lund 1991 praestat sero quam nunquam This thesis is based on the following publications which will be referred to by their Roman numerals.

I Human granulocyte elastase is inhibited by the urinary trypsin inhibitor. Jönsson B-M, Löffler C, Ohlsson K. Hoppe-Seyler's Z. Physiol. Chem. 363; 1167-1175, 1982.

II Interactions of the complex between human urinary trypsin inhibitor and human leukocyte elastase with aj-proteinase inhibitor and cx2-macroglobulin. Jönsson B-M, Ohlsson K. Hoppe-Seyler's Z. Physiol. Chem. 365; 1403-1408, 1984.

III Radioimmunological quantitation of the urinary trypsin inhibitor in normal blood and urine. Jönsson-Berling B-M, Ohlsson K, Rosengren M. Biol. Chem. Hoppe-Seyler 370; 1157-1161, 1989.

IV Distribution and elimination of intravenously injected urinary trypsin inhibitor. Jönsson-Berling B-M, Ohlsson K. Scand. J. Clin. Lab. Invest. 51; 000-000, 1991.

V Turnover of urinary trypsin inhibitor in uremia, pancreatitis and after surgery Berling B-M. Submitted.

Reprints were made with permission from the publishers. Contents

Abbreviations 7

Purpose of the investigation 8

Introduction and background 9 Proteinases and proteinase inhibitors 10 Acute phase response 12 Methodological aspects 13 Radioimmunoassay 13 Radioactive labelling 14 Chromatographic techniques 14 Electrophoretic procedures 15 Antisera 16 Enzyme activity, inhibition assays 16 Determination of amino acid composition and carbohydrate content 17 Immunohistoehemical techniques 17 Healthy volunteers and patients 18

Results and comments 19 Purification of UTI 19 Inhibition of human leukocyte elastase and trypsin 20 Interaction between UTI-enzyme complexes and the major plasma proteinase inhibitors, aj-proteinase inhibitor and (X2-macroglobulin 21 UTI in normal blood and urine 21 UTI and the acute phase response 22 Turnover and organ distribution of 125j labelled UTI 24 Immunohistochemical findings 24

Summary and concluding remarks 26

Acknowledgements 29 References 31

Paper 1 43

Paper II 53

Paper III 59

Paper IV 65

Paper V 81 Abbreviations

UTI = urinary trypsin inhibitor ACTH = adrenocorticotropic hormone (X2M = alpha-2-macroglobulin kDa = kiloDalton cqPI = alpha-1-proteinase inhibitor ITI = inter-alpha-trypsin inhibitor PSTI = pancreatic secretory trypsin inhibitor SLPI = secretory leukocyte proteinase inhibitor MPI = mucus proteinase inhibitor SDS-PAGE = sodium dodecylsulphate polyacrylamid gel electrophoresis RIA = radioimmunoassay CRP = C-reactive protein PAP = peroxidase antiperoxidase cDNA = complementary deoxyribonucleic acid mRNA = messenger ribonucleic acid

BAPNA = Na-benzoyl-DL-arginine p-nitroanilide HCl Suc[Ala]3-Nan = N-succinyl-ala-ala-ala p-nitroanilide ELISA = enzyme linked immunosorbent assay Purpose of the investigation

The purposes of this investigation were:

• to purify and characterize the urinary trypsin inhibitor

• to investigate the function of the inhibitor and the interaction between urinary trypsin inhibitor and the major plasma proteinase inhibitors

• to study the metabolism of the urinary trypsin inhibitor

• to develop a method for measuring of the inhibitor in serum and urine in healthy subjects and during pathological conditions

vitam regit fortuna non sapientia Introduction and background

The urinary trypsin inhibitor (UTI) is an acid-stable proteinase inhibitor present in human serum and urine inhibiting trypsin, chymotrypsin, leukocyte elastase and cathepsin G, but its physiological function still remains unknown. That human urine was able to inhibit trypsin was discovered in the beginning of this century (Miiller 1908). Bauer and Reich reported in 1909 that urine from patients with nephritis, tuberculosis and amyloidosis in the kidney and acute infectious diseases had a high trypsin inhibiting capacity (Bauer 1909). The antitryptic capacity rose indepentently of the total protein content in the urine. In 1950 Dillard found a marked elevated trypsin inhibiting capacity in urine from patients with pneumococcal and streptococcal infections and with rheumatic fever. He also found a marked elevation during the first postoperative days following major surgery. Faarvang observed an increased excretion of trypsin inhibitor in urine during pregnancy. He found elevated levels of trypsin inhibitor in urine after administration of ACTH and hydrocortisone and suggested that the elevation might be caused by a stress reaction (Faarvang 1958). Proksch and Routh in 1972 isolated a trypsin inhibitor from the urine of pregnant women. They found that the majority of the antitrypsin activity in the urine was due to a glycoprotein containing approximately 12 per cent carbohydrate and with a molecular weight of about 70 kDa. In the 1970's Hochstrasser et al. described an acid stable trypsin inhibitor in serum with a molecular weight of 34 kDa and a high carbohydrate content. They concluded that this physiological serum inhibitor was identical with the acid-stable trypsin inhibitor in urine and that both inhibitors were transformed by trypsin into an inhibitor with a molecular weight of 17 kDa. They described an immunological cross reaction with these inhibitors and the inter-alpha-trypsin inhibitor (ITI). They concluded that the ITI was a precursor of the acid stable inhibitors (Hochstrasser 1973, 1976) and that leukocyte elastase cleaved the physiological inhibitor from ITI in vivo (Dietl 1979). Sumi et al. (1981) and Tanaka et al. (1982) purified and characterized two forms of the urinary trypsin inhibitor. The molecular weights were 43 kDa and 20 kDa, respectively, estimated by sodium dodecylsulphate polyacrylamid gel electrophoresis (SDS-PAGE). The inhibitor wit- the highest molecular weight was assumed to be the native form and the smaller one was produced during the purification procedure.

Proteinases and proteinase inhibitors pertinent to this study

Proteinases are enzymes that degrade proteins and play a determinant role in many biological processes. Most of the proteinases are present in inactive precursor forms in the body. Trypsinogen is the precursor of trypsin and is secreted by the pancreas. It enters the intestine v:a the pancreatic duct and is converted to trypsin by enterokinase, a proteinase present in the intestine. Trypsin can then activate trypsinogen and other proenzymes like chymotrypsinogen, precarboxypeptidase and proelastase, which are also secreted by the pancreas (Kenny 1977). Human leukocyte elastase is located in the azurophil granules in the neutrophil granulocytes. It degrades proteins that have been engulfed daring phagocytosis. The enzyme can be released after cell death and by leakage from cells during phagocytosis. It may then cause degradation of connective tissue proteins such as elastin and collagen. It is suggested that leukocyte elastase plays a major role in the destruction of the alveolus during the development of pulmonary emphysema (Travis 1988). Uropepsin is a proteolytical enzyme detected in human urine and it presumably originates in the kidneys. It degrades various proteins (Barret 1980).

To protect normal tissues from uncontrolled proteolysis and destruction by proteinases released at cell death, phagocytosis, complement activation, etc., there is a group of proteins that are able to inhibit these proteinases, the so called proteinase inhibitors. The proteinase binds to the proteinase inhibitor forming an enzyme-inhibitor complex. These complexes are then removed from the blood or lymph system by the reticuloendothelia! system. cc2-macroglobulin (0C2M) is probably the most important proteinase inhibitor. It is synthesized in hepatocytes and in macrophages. Its molecular weight is about 725 kDa. 0C2M inhibits all types of proteinases not by active-

10 silt* directed inhibition but by steric shielding and rapid clearance of the enzyme-inhibitor complex. coM appears to play a key role in the defence against trypsin and is a secondary inhibitor of thrombin and plasmin, which are primarily inhibited by antithrombin III and antiplasmin, respectively. o^M may function as a backup inhibitor under conditions where the primary inhibitors become depleted. A total abscence of (X2M is probably inconsistent with life. If its proteinase inhibiting capacity is exceeded in in vivo experiments the animal will develop shock (Ohlsson 1971, Sottrup-Jensen 1989). aj-protc;:iase inhibitor (oq-antitrypsin, ajPI) has a molecular weight of 52 kDa and is synthesized in the liver. It has a broad target enzyme specificity including granulocyte proteases, especially leukocyte elastase. The mean concentration of ctjPI in serum is 1.3 g/1 making it the quantitatively dominating proteinase inhibitor in plasma. Because of its relatively small size and high serum concentration it can move extravasculady and inhibit intercellulary released proteinases. The reactive centre of a j PI contains a methionine residue which is easily oxidized during smoking, resulting in decreased antiproteolytic activity. Deficiency of CC1PI will lead to a loss of elasticity in the lung resulting in pulmonary emphysema (Laurell 1963, Eriksson 1965, Beatty 1980). Inter-a-trypsin inhibitor (ITI) is a proteinase inhibitor with a molecular weight of 180 kDa and a mean concentration in serum of 0.5 g/1. The molecule consists of two heavy protein chains and one light chain encoded on three different mRNA:s. The light chain contains the inhibitory domains and its amino acid sequence is identical to that of UTI (Schreitmuller 1987). The polypeptide chains are probably linked tr».ether by a glycosaminoglycan (Jessen 1988). The physiological functL.. jf ITI is unknown. It inhibits trypsin, chymotrypsin, human leukocyte elastase, acrosin, cathepsin G and plasmin in vitro (Gebhard 1986), but in vivo there are other proteina^e inhibitors with much stronger affinities for these enzymes. Pancreatic secretory trypsin inhibitor (PSTI) is a low molecular weight trypsin inhibitor (MW 6.5 kDa). It is produced in the pancreas and exists in its free form in both normal serum and pancreatic juice. In pancreatitis PSTI forms complexes with trypsin, and the plasma level of the inhibitor increases (Lasson 1984). In serum PSTI-trypsin complexes are cleaved and trypsin is transfered to the major serum proteinase inhibitors, «2^ ar>d ex j PI (Eddeland

11 1978). Secretory leukocyte proteinase inhibitor (SLPI) is an acid stable proteinase inhibitor isolated from human parotid saliva (Thompson 1986). It is also present in other human mucous secretions. It is identical to the mucus proteinase inhibitor, MPI (Fritz 1988). SLPI inhibits trypsin, chymotrypsin, human leukocyte elastase, cathepsin G, chymase and tryptase.

Acute phase response

The acute phase response is the reaction of the organism to tissue injury, infection or tumour growth. It consists of a local reaction such as clot formation, activation of granulocytes and dilatation of blood vessels. There is also a systemic reaction with fever, leukocytosis, increase in erythrocyte sedimentation rate, increase in secretion of ACTH and glucocorticoids, negative nitrogen balance and changes in the concentration of some plasma proteins. The acute phase reaction is elicited via the production of cytokines like interleukin-6, interleukin-1 and tumour necrosis factor a by monocytes/macrophages and to some extent by fibroblasts and endothelial cells. The acute phase proteins are synthesized in the liver and include all proteins whose rates of synthesis are more than doubled during an acute phase reaction. In man serum C-reactive protein and serum amyloid A show the highest known increase during an acute phase response (Heinrich 1990).

12 Methodological aspects

Radioimmunoassay

The technique used is based on competition between radiolabelled antigen (tracer) and unlabelled antigen for a limited number of antibody binding sites. The tracer, sample and antibody aie mixed and incubated. The amounts of antigen and antibody are too small to form a precipitate, therefore a second antibody against the first antibody is added. After incubation the mixture is centrifuged, the supernatant is decanted, and the radioactivity of the precipitate is measured. The amount of tracer measured in the precipitate is in inverse relationship to the amount of antigen in the sample. By using known amounts of antigen, fixed amounts of tracer and antibodies a standard curve can be constructed. This curve is used to determine the unknown amount of antigen in samples. In the radioimmunoassay used for measurements of UTI wc used purified UTI as standard (paper III). 125j labelled UTI was used as tracer. The antibodies used were sheep antiUTI antibodies and, to improve the precipitation, rabbit antisheepIgG and normal sheep serum were added. In earlier studies of the urinary trypsin inhibitor the acid stable trypsin inhibiting activity in urine or plasma was generally used as a measure of the amount of this inhibitor, but the inhibiting activity may originate from other acid stable trypsin inhibitors and is not specific for UTI (Shulman 1955, Nowak 1968). In later studies enzyme linked immunosorbent assay (ELISA) and crossed immunoelectrophoresis were used (Byrjalsen 1986, Yoshida 1987, Sakakibara 1989). Usui et al. developed a RIA for the measurement of UTI in the urine (1984). We prefered to use a RIA method because of its high sensitivity, high specificity, good precision and accuracy. By introducing an acid precipitation step we could measure UTI also in plasma as we then eliminated the cross-reacting inter-a-trypsin inhibitor while the acid stable UTI remained in solution.

13 Radioactive labelling

Radioactive labelling of human leukocyte elastase and trypsin used in the in vitro studies was performed by the lactoperoxidase method (Thorell 1971). The Chloramine-T method (Greenwood 1963) was used for radioactive labelling of UTI. The molecular weight of lactoperoxidase is close to that of UTI so we therefore prefered to use Chloiamine-T which can be seperated from UTI by gel filtration. 125j UTI to be used in human volunteers was produced by the Iodo-gen® method (Fraker 1978). Iodo-gen® is attached to the inside of a test tube and a mixture of UTI and iodide is added. The reaction takes place only on the inside surface of the test tube and when the mixture is decanted there is no Iodo-gen® contamination of the mixture. Labelling procedures may cause changes in proteins. The trypsin inhibiting capacity of UTI was however unchanged after labelling.

Chromatographic techniques

Gel filtration chromatography

This technique is used to separate molecules of varying molecular size using a gel of cross linked macromolecules e.g. dextrans. By varying the degrees of cross linking the gels differ in porosity and can therefore be used over different molecular size ranges. The gel is packed in a column and if a sample containing molecules of varying sizes is applied on top of the column large molecules excluded from the pores in the gel will pass through the column first. Smaller molecules will pass into the pores and pass through the column at a slower rate. In this investigation we used gel filtration chromatography to separate UTI from other proteins of different molecular sizes.

Affinity chromatography

A specific ligand which will bind to the compound to be purified is attached to an insoluble gel matrix. When a mixture containing the compound is applied

14 to the gel there will be a binding between the compound and ligand. The gel matrix is washed with a buffer solution and substances with relatively low affinity for the ligand are removed. The compound of interest is then separated from the gel by altering physical conditions, e.g. pH. In the present investigation we used bovine cx-chymotrypsin, which will bind to UTI, coupled to cyanogen bromide activated Sepharose 4-B gel (Pharmacia, Sweden) in the first step of purification of the inhibitor from human urine. We used the same technique to show that the radioactively labelled UTI excreted in the urine after intravenous injection retained its chymotrypsin binding ability.

Ion exchange chromatography

In ion exchange chromatography molecules are separated depending on their charges. Since UTI has an isoelectric point at 2.8 its net charge above this pH is negative. Therefore we used an anionic exchanger at a pH of 7.0. The inhibitor was bound to the positive charges in the column and was eluted using a NaCl concentration gradient. This technique was part of the purification procedure of UTI.

Electrophoretic procedures

Agarose gel electrophoresis (Johansson 1972) was used to screen the protein content during the UTI purification procedure. Immunoelectrophoresis (Scheidegger 1955) and crossed immunoelectrophoresis (Ganrot 1972) were performed to study the cross reactivity between UTI and ITI. They were also used to identify proteinase/inhibitor complexes. Electroimmunoassay (Laurell 1972) was used to analyse the content of the urinary trypsin inhibitor in gel filtration fractions, to determine C-reactive protein (CRP) in plasma and aj-microglobulin in plasma and urine samples. Sodium dodecylsulphate polyacrylamid gel electrophoresis (SDS- PAGEjwas used to determine the molecular weight of UTI (Weber 1969). SDS is an anionic detergent that denatures the protein. It also binds strongly to the protein giving it a constant negative charge per unit mass. The complex between SDS and the protein will therefore move towards the anode during

15 eiectrophoresis. Its mobility in the polyacrylamide gel depends on its molecular mass and run together with standard proteins the apparent molecular weight of the sample protein is determintd

Antisera

Antisera against UTI were raised by immunization of rabbits and sheep (paper I). Equal volumes of UTI (I mg/ml) and Freund's complete adjuvans were thoroughly emulsified. The latter consists of mineral oil, an emulsifying agent and heat-killed acid fast bacteria and will increase the immunizing stimulus. This mixture was then injected subcutaneously at multiple sites of the animals (about 0.5 mg/animal). The animals were boostered after four weeks and, if needed, additional boosterdoses were given. The resulting antibodies show a cross reaction with the ITI, which is comprehensible since the low molecular form of UTI represents the inhibitor part of the ITI molecule (Enghild 1989). Antisera against human leukocyte elastase, oq-proteinase inhibitor and 0:2- macroglobulin were prepared as earlier described (Ohlsson 1974, Ohlsson 1976a, Ohlsson 1976b).

Enzyme activity, inhibition assays

The activity and inhibition of human leukocyte elastase and porcine pancreas elastase were measured spectrophotometrically using Suc-[Ala]3-Nan as substrate (Bieth 1974). The activity and inhibition of trypsin were also estimated spectrophotometrically using Bz-Arg-Nan (BAPNA) as substrate (Erlanger 1961). The bovine trypsin was active site titrated using 4- nitrophenyl 4-guanidinobenzoate (Chase 1967).

16 Determination of amino acid composition and carbohydrate content

Quantitative determination of the amino acid composition of UTI was made using a JEOL 5-AH amino acid analyzer (Spackman 1958). Cysteine and methionine were estimated as cysteic acid and methionine sulfone, respectively, after performic acid oxidation (Moore 1963). Tryptophan was determined spectrophotometrically (Edelhoch 1967). Sialic acid and total hexosamine were assayed according to standard methods (Warren 1959, Spiro 1966).

Immunohistochemical techniques

The peroxidase-antiperoxidase (PAP) method according to Sternberger et al. (1970) was used for immunohistochemical identification of UTI in human tissues. This step-wise procedure involved sequential application to the tissues of a) rabbit antiserum against UTI, b) swine antirabbit IgG antiserum, c) peroxidase coupled to a rabbit antibody against horseradish peroxidase (PAP- complex), d) peroxidase substrate solution, diaminobenzidine, and hydrogen peroxide. After each step the tissue sections were rinsed with 0.05 mol/1 Tris- HC1 buffer, 0.15 mol/1 NaCl, pH 7.6. Finally the sections were dehydrated and mounted in Pertex, a synthetic mounting medium. The tissues were then examined in a light microscope. To reduce background staining of connective tissue, pepsin and non immune swine serum were added to the tissues before the start of the above procedure (Reading 1976). The substrate chromogen reaction cannot distinguish between the peroxidase immunologically localizing the antigen (UTI) and endogenous peroxidase activity present in the tissues. To irreversibly inhibit endogenous enzyme activity the tissues were incubated with hydrogen peroxide at the beginning of the procedure. To estimate the unspecific staining of the specimen, sections were incubated with normal non immune rabbit serum instead of rabbit anti UTI antiserum. These sections served as controls. Immunostaining was also performed using a streptavidin-biotin-peroxidase complex instead of PAP-complex. The chromogen substrate used in this case was 3-amino-9-ethylcarbazole (Yoshida 1991).

17 Healthy volunteers and patients

Healthy volunteers were used for blood and urine sampling for determination of the normal values of UTI in these body fluids. Three volunteers were also

intravenously injected with 125i_iabelled UTI to investigate the turnover of UTI in man. To study the changes in the turnover of UTI during surgical trauma, inflammatory disease and renal insufficiency, patients who underwent gall stone operation, patients with acute pancreatitis and uremic pateints were investigated.

18 Results and comments

Purification of UTI (paper I)

UTI was purified from urine from pregnant women. The purification procedure started with affinity chromatography using bovine a-chymotrypsin coupled to a gel matrix. The inhibitor containing eluate was then concentrated and applied to a ion-exchange chromatography column. One protein peak, called inhibitor B, was eluted with the starting buffer and one protein peak, called inhibitor A, was eluted by the gradient. The two inhibitors were further purified using gel filtration. Inhibitor B, reduced or unreduced, showed two closely spaced protein bands with a molecular weight of about 20 kDa on SDS-PAGE. Inhibitor A, unreduced, showed one protein band of about 44 kDa. After reduction inhibitor A gave two bands with a molecular weight of about 20 kDa on SDS-PAGE suggesting that inhibitor A consists of two units of inhibitor B linked together by disulfide bridges. The amino acid composition and carbohydrate content of the two inhibitors were analysed. Both inhibitors contained a high frequency of glycine, glutamic acid and aspartic acid residues. They also contained neutral hexose and sialic acid. The amino acid composition also suggests that inhibitor A is a dimer of B and was similar to that reported by others (Hochstrasser 1973, Muramatu 1980, Fex 1981, Balduyck et al 1982 ) Fresh human urine contained only inhibitor A. If urine was stored at an acid pH (2.5) and +4 °C for 24 hours inhibitor B was also detected. If pepstatin was added to the urine before storage no inhibitor B was detectable (paper I). We suggested that inhibitor A probably was the native form and inhibitor B appeared during storage of the urine before purification. Using immunoelectrophoresis we showed a cross reactivity between ITI and the two inhibitors using antibodies against inhibitor A or inhibitor B. When ITI was precipitated from serum with specific antibodies there was no precipitate between serum and antiserum against the inhibitors. The

19 explanation is probably that the concentration of inhibitor A in serum is too low to be detectable by the immunoelectrophoresis, since we now know, using more sensitive techniques, that the inhibitor is present in human serum.

Inhibition of human leukocyte elastase and trypsin (paper I)

We found that both inhibitor A and inhibitor B inhibited human leukocyte elastase but no inhibition of porcine pancreatic elastase was detected (paper I). Identical results were simultaneously published by Bromke and Kueppers (1982) and later by other authors (Albrecht 1983, Balduyck 1985, Ogawa 1987). Our preparation of inhibitor A and inhibitor B inhibited the amidolytic activity of trypsin in an apparent inhibitor/enzyme molar ratio of 1:2 for inhibitor B and 1:4 for inhibitor A (paper I), which was also found by others (Muramatu 1980, Tominaga 1982, Salier 1983, Okumichi 1984). In 1986 Kaumeyer et al. isolated a cDNA clone from human liver library which coded for a polypeptide related to the light chain of ITI. They determined its entire DNA sequence and found that this sequence codes for both aj-microglobulin and the low molecular form of UTI, corresponding to our inhibitor B. This form consists of two Kunitz-like inhibitor domains and it was therefore renamed bikunin (Gebhard 1989). Performing sequence studies Swaim and Pizzo (1988) found that the two separate inhibitor domains contain different amino acid residues, methionine and arginine, respectively, at their reactive centres. By modifying these residues they found that elastase and cathepsin G were inhibited at either of the two reactive centres, but trypsin and chymotrypsin were inhibited only at the arginine reactive centre. It therefore now seems established that the inhibitor in its dimeric form inhibits trypsin in a molar ratio of 1:2, as had previously been suggested by some authors (Kochstrasser 1976, Barthelemy-Clavey 1979, Tanaka 1982, Balduyck 1985).

20 Interaction between UTI-enzyme complexes and the major plasma proteinase inhibitors, aj-proteinase inhibitor and oo-maeroglobulin (paper II)

We found that the complex between UTI and human leukocyte elastase is stable during incubation at room temperature for 24 h. During incubation there was some degradation of inhibitor A to inhibitor B which was still in complex with elastase. If human serum is added to the complex it rapidly dissociates and leukocyte elastase is found mainly in complex with ccj- proteinase inhibitor but also with 0C2-macroglobulin. UTI is a stronger inhibitor of trypsin than of leukocyte elastase. The UTI-trypsin complex however also dissociates rapidly in the presence of human serum. Trypsin is found mainly bound to 012-macroglobulin and a smaller amount to a.\- proteinase inhibitor. Thus it does not seem likely that UTI is a significant inhibitor of leukocyte elastase or trypsin in human serum, but it may act as a local inhibitor interstitially where low concentrations of other proteinase inhibitors may occur. A similar pattern of interaction is shown by other acid stable inhibitors as PSTI and SLPI which are both judged to be local inhibitors (Eddeland 1978, Gauthier 1982).

UTI in normal blood and urine (paper III)

A radioimmunoassay was developed for measuring UTI in plasma and urine. Because of the immunological cross reactivity between ITI and UTI the plasma and serum samples were treated with perchloric acid to precipitate the acid labile ITI before analysing the samples. The sensitivity of the assay was ljj.g/1. The intra-assay variation was 2.4% and the inter-assay variation was 13.3%. The recovery of exogenous UTI in urine and serum was 85% and 108%, respectively. Dilution curves of acid treated serum and plasma samples and urine samples were parallel to the UTI (inhibitor A) standard curve indicating that we really measured UTI. Gel fL ation of the serum and urine samples revealed that the UTI present in normal serum and urine was the dimeric form, inhibitor A. UTI was not released from ITI during acid treatment of plasma and serum as shown by gel filtration of both acid treated

21 and non acid treated serum. As inhibitor B was renamed bikunin (Gebhard 1989) because of its two Kunitz-type domains I suggest that the physiological dimeric A form present in plasma and urine will be called tetrakunin. We found a mean value of 7.14 mg/1 in plasma in 24 healthy persons and 6.38 mg/1 in serum in 30 healthy blood donors (paper III). We have later found lower levels in repeated assays. Mean values of UTI found in serum from healthy blood donors were 1.7 mg/1 (range 1.5-2.1 mg/1, n=15). In a sample of pooled serum from 20 healthy blood donors the values was 2.1 mg/1 and in a pool of plasma from 27 healthy blood donors we found 2.5 mg/1 of UTI. The normal amount of UTI in the urine per 24 hours is 8.17±1.18 mg (paper III). Other authors have found similar amounts of UTI in serum and urine. Hochstrasser et al. (1981) found about 11 mg/1 in serum and 12-36 mg/1 excreted in the urine per 24 h. Yoshida et al. (1987) used an ELISA to determine UTI and found 11.5±3.O mg/1 (mean±SEM) in acid treated plasma and 17.0±0.9 mg/1 in the urine. Toki and Sumi (1982) measured the activity of UTI in the urine using casein agar. They found 1.78-10.48 mg/1 in normal urine. Usui et al. (1984) found 4.83 ±2.46 mg/24 h (meaniSD) in the urine of healthy men and 3.86+1.35 mg/24 h in healthy women using a radioimmunoassay.

UTI and the acute phase response (paper V)

An increased concentration of UTI in urine and plasma has been found in several pathological conditions such as cancer, inflammation, renal insufficiency (Rudman 1979, Maruyama 1984, Ödum 1987, Gebhard 1989, Laroui 1988, Sakakibara 1989). Some authors suggest that UTI is an acute phase reactant (Salier 1983, Ödum 1987, Kuwajima 1990). To study UTI during an acute phase situation we measured UTI in plasma and urine as well as C-reactive protein (CRP) in serum in patients who underwent abdominal surgery (gall stone operation). We chosei CRP because this is the acute phase reactant in serum that shows the highest increase and the fastest response to trauma and inflammation (Heinrich 1990). We found a slight increase of plasma UTI postoperatively and in four out of ten patients the values on days 2-4 postoperatively were doubled. The mean excretion of UTI in urine was elevated approximately six times on the second postoperative day and the time

22 course of this elevation closely followed that of serum CRP. We also studied nine patients with acute pancreatitis. Their maximal mean plasma UTI value was reached on the third day of observation and was approximately twice the normal value. These patients also had an elevated excretion of UTI in the urine during the observation time, being 7-25 times the normal value. The slight elevations of the plasma levels may be caused by an elevated synthesis of UTI during these pathological conditions although according to Bourguignon et al. (1989) UTI is not released in its free form from human hepatoma cells but only in association with the two heavy peptide chains forming ITI. Another explanation may be a degradation of ITI but when 125i labelled ITI was injected intravenously or intraperitoneally in mouse it was not split into fragments after induction of inflammation (Pratt 1986). During inflammation Salier et al. (1983) found elevated serum levels of ITI-derivates without a decrease in serum ITI levels. A decrease in the glomerular filtration rate may of course lead to elevated serum UTI levels. UTI is thus probably not an acute phase reactant because it is not evident that there is an increased synthesis of free UTI. It is also doubtful whether the increase of the serum level is sufficient to be called acute phase response. Tubular proteinuria is common following trauma and acute pancreatitis (Lankish 1979, Meier 1986, Gosling 1986). UTI seems to be metabolized in the proximal tubules of the human kidney (paper IV) and increased excretion of UTI in the urine following surgical trauma and during acute pancreatitis may depend mainly on tubular dysfunction rather than an increase in gJomerular filtration with a simultaneously elevated plasma level. a\- microglobulin concentration in urine is elevated in the urine in patients with tubular proteinuria (Ekström 1975). To estimate the tubular proteinuria in these patients aj-microglobulin was measured in the urine samples and we found that the excretion of aj-riiicroglobulin very closely followed the excretion of UTI in the urine. Gel filtration of plasma and urine samples from the patients as well as in healthy volunteers showed that the inhibitor was present only in its dimeric form, i.e. inhibitor A.

23 Turnover and organ distribution of 125i labelled UTI (paper IV)

To study the turnover of UTI in homo 125j labelled UTI (inhibitor A) was injected intravenously in three healthy male volunteers. There was a rapid elimination of 125i UTI and the plasma half life of the labelled inhibitor was 2 h. During the first four hours after injection labelled biologically active inhibitor was detected in the urine. This finding and the regular presence of UTI in the urine indicate that the tubular reabsorption is not complete. Proximal tubular reabsorption of various proteins can be blocked by infusion of arginine in a dose-dependent manner (Mogensen 1975, 1977). Sakakibara et al. used intravenous arginine infusion in healthy subjects showing that the tubular reabsorption of UTI is about 70% (1989). 1251 UTI was also injected intravenously in rats. There was a rapid disappearance of free inhibitor from plasma. The half life of the inhibitor was 9 min. Animals were sacrificed after 15 and 60 min, and the organ distribution of the radioactivity was investigated. Forty-four per cent of the radioactivity was found in the kidneys and 9% in the liver 15 min after injection. One hour after the injection of the radioactively labelled inhibitor the values were 7% and 3%, respectively. This finding also indicates that the major part of UTI is metabolized in the kidneys. UTI in plasma and urine was measured in uremic patients. The levels were about ten times higher in uremic patients than in heathy subjects (paper V). This result together with the above findings show that UTI is filtered by the glomerulus. The high urine level in uremia might be the result of decreased tubular reabsorption.

Immunohistochemical findings (paper V)

Using two different immunohistochemical techniques including peroxidase- antiperoxidase (PAP) complexes (Sternberger 1970) or streptavidin-biotin- peroxidase complexes (Yoshida 1991) we found immunoreactivity against UTI in the proximal tubules of the human kidney, which is in agreement with reports by Yoshida et al. (1989) and ödum (1989). We also examined human liver, brain, stomach, colon as well as tumours of colon and kidney but no

to immunoreactivity corresponding to UTI was found in these tissues. This is in contrast to the findings of Yoshida et al. (1989, 1991) who demonstrated immunoreactivity against UTI in the Kuppfer cells in the liver, glial cells of the cerebrum, lamina propria of the stomach and colon and in the connective tissues surrounding tumour infiltrations. Differences in the antisera used by the various authors may explain this discrepancy.

25 Summary and concluding remarks

The proteinase inhibitor UTI was purified from urine from pregnant women. Its molecular weight and amino acid content were determined. Our purification procedure resulted in two forms of UTI, inhibitor A and inhibitor B, where inhibitor A probably is a dimer of B. When analysing plasma and urine samples from patients as well as from healthy persons only inhibitor A was found suggesting that this is the physiological inhibitor form. It has now been established that there is a DNA sequence on chromosome 9 coding for both aj-microglobulin and the inhibitor B form of UTI (Traboni 1989). This UTI form contains two Kunitz-type domains and it was called bikunin (Gebhard 1989). Bikunin is secreted from the hepatocytes linked to two heavy peptide chains forming ITI. Bikunin is secreted in its free form from rat hepatocytes (Sjöberg 1991), but it is not known if it is released in its free form in homo. Degradation of ITI may explain the presence of the inhibitor B form in plasma and urine, but it is harder to understand how simple degradation of ITI leads to the presence of the dimeric form (inhibitor A) of UTI in the organism. During storage of urine samples there seems to be a degradation of the dimeric form into the bikunin-inhibitor B-form. UTI was found to inhibit human leukocyte elastase. Complexes between trypsin and UTI and leukocyte elastase and UTI rapidly dissociated in the presence of human serum. Elastase was found mainly in complex with a.\- proteinase inhibitor and trypsin was mainly bound to 0C2-macroglobulin. It seems unlikely that the physiological function of UTI is as a trypsin or elastase inhibitor in plasma. It may act as a local inhibitor or as a shuttle transporting the proteinases to the major plasma proteinase inhibitors. UTI may be an inhibitor of some unknown proteinase released during physiological or pathological conditions. UTI has been isolated from human hepatoma cells and was found to stimulate the growth of human endothelial cells (McKeehan

1986). When 125i_iabelled UTI was injected intravenously to three healthy volunteers the half life of the inhibitor in plasma was two hours. This

26 relatively short half life suggests that its main function is not in plasma. Biologically active UTI was excreted in the urine after this intravenous injection, indicating that the inhibitor is filtered through the glomeruli. To study the organ distribution of UTI radioactively labelled inhibitor was injected intravenously in rats. Compared to other organs the kidneys contained most of the radioactivity. This was also found in mice by Sugiki et al. (1989) and in rats by Sjöberg et al. (1991). Using immunohistochemical techniques immunoreactivity corresponding to UTI was found in the proximal tubules of the kidney. These findings suggest that UTI is filtered through the glomeruli and reabsorbed by the proximal tubules. The reabsorption is, however, not complete as healthy persons excrete about 8 rr.g/day, but UTI does probably not play its major physiological role in the urine. The levels of the inhibitor in urine and plasma have shown to be elevated during several pathological conditions (paper V). UTI has also been called an acute phase reactant. We studied the UTI levels during surgical trauma, acute pancreatitis and in renal insufficiency. Postoperatively after biliary tract surgery and during the course of acute pancreatitis there was a slight elevation of the plasma UTI level and there was a marked elevation of UTI in the urine. This increase in the urine very closely followed the excretion of a\- microglobulin indicating a decreased tubular reabscrption. The excretion of UTI also closely followed the plasma level of C-reactive protein after biliary tract surgery, but the elevation in plasma is too low to be called an acute phase reaction. The elevation of UTI in plasma may be due to an increased synthesis, but there is no corresponding increase of ITI or aj-microglobulin in plasma. Several proteinases have been tested to convert ITI into active, •icid-stable derivates, but the ITI degradation in plasma is extremely slow due to the presence of other proieinase inhibitors (Potempa 1989). We did not study the levels of ITI, but according to some authors there is a decrease in the ITI-concentration during the acute phase response (Chawla 1978, Gebhard 1986, Heinrich 1990, Liony 1991). Other authors, however, found no decrease in the ITI-concentration during the acute phase (Salier 1983, Sesboiie 1983, Ödum 1987). Another explanation for the elevated plasma UTI level may be a slight decrease in the glomerular filtration rate. In uremic patients we found an almost ten-fold increase of UTI in plasma and in urine probably depending on a decreased glomerular filtration rate and a decreased reabsorption capacity of the proximal tubules. As the physiological role of UTI is still unknown it is difficult to speculate on possible therapeutic benefits of the protein. Intravenously administered

27 UTI has been shown to increase the survival rate of animals with experimental induced acute pancreatitis (Ohnishi 1984) and in patients with pneumocystis carinii infection (Akashi 1991). Very recently Ishigami et al. (1991) found that UTI has a protective effect on renal failure induced in rats by gentamycin and mercuric chlorid.

tempus omnia revelat

28 Acknowledgements

I wish to express my sincere gratitude to all friends who have helped me to make this thesis possible, especially:

Professor Kjell Ohlsson, my tutor, for his enthusiasm, patience and support.

Docent Anders Borgström for fruitful discussions and criticism.

Docent Sven Genell for introducing me to Kjell and for his friendship and care.

Professor Hans Renck for his criticism and encouragement.

Margareta Rosengren, Anita Holmgren, Lena Axelsson, Åsa Polling and Irena Ljungcrantz for their very skillful assistance and friendship throughout the years.

Magnus Bergenfeldt, Peter Björk and Kirsten Dan-Olsen for many fruitful discussions and their great support during the last year.

Joyce Carlson for her friendship and for always revising the english text.

Eddie Fremer, Lennart Johansson, Gun-Britt Lindqvist and Birgitta Larsson at the Medical Library for their excellent aid in finding the litterature.

Friskis och Svettis for helping me to relax and stimulating the endorphin- secretion necessary for this work.

Rikard, my husband, friend and colleague, for his excellent house-keeping, patience and encouragement.

Fredrik, Anders and Marie, our children, who have kept asking about the hat.

My parents for everything.

29 This investigation was supported by grants from the Swedish Medical Research Council (3910), the Medical Faculty, University of Lund, the Foundation for Cancer Research administered by Malmö General Hospital, the Albert Påhlsson Foundation, the Crafoord Foundation and the Torsten and Ragnar Söderberg Foundations.

30 References

Akashi K, Ishimaru T, Shibuya T, Harada M, Niho Y. Human urinary proteinase inhibitor in the treatment of P carinii pneumonia. Chest 99; 1055- 1056 (1991).

Akazawa K, Sumi H, Maruyama M, Mihara H. Acid stable protease inhibitor in ascites of ovarian carcinoma. Clin. Chim. Ada 131; 87-99 (1983).

Albrecht GJ, Hochstrasser K, Salier J-P. Elastase inhibition by the inter-a- trypsin inhibtor and derived inhibitors of man and cattle. Hoppe-Seyler's Z. Physiol. Chem. 364; 1703-1708 (1983).

Balduyck M, Davril M, Mizon C, Smyrlaki M, Hayem A, Mizon J. Human urinary proteinase inhibitor: inhibitory properties and interaction with bovine trypsin. Biol. Chem. Hoppe-Seyler 366; 9-14 (1985).

Balduyck M, Hayem A, Kerckaert JP, Mizon C, Mizon J. Isolation of a human urinary trypsin inhibitor. Biochem. Biophys. Res. Commun. 109; 1247-1255 (1982).

Balduyck M, Mizon C, Loutfi H, Richet C, Roussel P, Mizon J. The major human urinary trypsin inhibitor is a proteoglycan. Eur. J. Biochem. 158; 417-422 (1986).

Barret AJ, McDonald JK. In Mammalian proteases. A glossary and bibliography. Volume 1. Endopeptidases. Academic Press, London 320-321 (1980)

Barthelemy-Clavey V, Yapo EA, Vanhoutte G, Hayem A, Mizon J. Purification et caracterisation des inhibiteurs de proteases de l'urine humaine. Biochim. Biophys. Ada 580; 154-165 (1979)

31 Bauer J, Reich Z. Uber die antitr>ptische Wirkung des Harns. Med. Klin. 46; 1744-1747 (1909).

Beatty K, Bieth J, Travis J. Kinetics of association of serine proteinase with native and oxidized a-1-proteinase inhibitor and a-1-antichymotrypsin. J. Biol. Chem. 255;3931-3934 (1980).

Bieth J, Spiess B, Wermuth CG. The synthesis and analytical use of a highly sensitive and convenient substrate of elastase. Biochem. Med. 11; 350-357 (1974).

Bourguignon J, Sesboiié R, Diarra-Mehrpour M, Daveau M, Martin J-P. Human inter-a-trypsin inhibitor. Synthesis and maturation in hepatoma HepG2 cells. Biochem. J. 261; 305-308 (1989).

Bromke BJ, Kueppers F. The major urinary protease inhibitor: simplified purification and characterization. Biochem. Med. 27; 56-67 (1982).

Byrjalsen I, Hansen-Nord G, ödum L. Human inter-a-trypsin inhibitor and immunologically related inhibitors investigated by quantitative immunoelectrophoresis. I. Method and reference material. Clin. Chim. Acta 161; 59-68 (1986).

Chase T, Shaw E. p-nitrophenyl-p'-guanidinobenzoate HC1: a new active site titrant for trypsin. Biochem. Biophys. Res. Commun. 29; 508-514 (1967).

Chawla RK, Wadsworth AD, Rudman D. Relation of the urinary cancer- related glycoprotein EDC 1 to plasma inter-a-trypsin inhibitor. J Immun. 121; 1636-1639(1978).

Dietl T, Dobrinski W, Hochstrasser K. Human inter-a-trypsin inhibitor. Limited proteolysis by trypsin, plasmin, kallekrein and granulocytic elastase and inhibitory properties of the cleavage products. Hoppe-Seyler's Z. Physiol. Chem. 360; 1313-1318 (1979).

Dillard GHL. The trypsin inhibitor of the urine in health and disease. J.Lab.

32 Clin. Med. 36;266-271 (1950).

Eddeland A, Ohlsson K. Studies on the pancreatic secretory trypsin inhibitor in plasma and its complex with trypsin in vivo and in vitro. Scand. J. Clin. Lab. Invest 38; 507-515 (1978).

Edelhoch H. Spectroscopic determination of tryptophan and tyrosine in proteins. Biochemistry 6; 1948-1954 (1967).

Ekström B, Peterson PA, Berggård I. A urinary and plasma aj-glycoprotein of low molecular weight: isolation and some properties. Biochem. Biophys. Res. Commun. 65; 1427-1433 (1975).

Enghild JJ, Thogersen IB, Pizzo SV, Salvesen G. Analysis of inter-oc-trypsin inhibitor and a novel trypsin inhibitor, pre-a-trypsin inhibitor, from human plasma. J. Biol. Chem. 264; 15975-15981 (1989).

Eriksson S. Studies in aj-antitrypsin defiency. Acta Med. Scand. 177,suppl 432; 1-85 (1965).

Erlanger BF, Kokowsky N, Cohen W. The preparation and properties of two new chromogenic substrates of trypsin. Arch. Biochem. Biophys. 95; 271- 278 (1961).

Faarvang HJ. Relation between pituitary-adrenal system and excretion of trypsin inhibitor in urine. P.S.E.B.M. 98;89-91 (1958).

Fex G, Grubb A, Loeffler C, Larsson J. Isolation and partial characterization of a low molecular weight trypsin inhibitor from human urine. Biochim. Biophys. Acta 667; 303-308 (1981).

Fraker PJ, Speck JC Jr.. Protein and cell membrane iodinations with a sparingly soluble chloroamine, l,3,4,6-tetrdchloro-3a,6a-diphenylglycoluril. Biochem. Biophys. Res. Commun. 80; 849-857 (1978).

Fritz H. Human mucus proteinase inhibitor (human MPI). Human seminal inhibitor I (HUSI-I), antileukoprotease (ALP), secretory leukocyte protease inhibitor (SLPI). Biol. Chem. Hoppe-Seyler 369,suppl;79-82 (1988).

33 Ganrot PO. Crossed immunoelectrophoresis. Scand. J. Clin. Lab. Invest. 29 suppl. 39-41 (1972).

Gauthier F, Fryksmark U, Ohlsson K, Bieth JG. Kinetics of the inhibition of leukocyte elastase by the bronchial inhibitor. Biochim. Biophys. Acta 700; 178-183 (1982).

Gebhard W, Hochstrasser K. Inter-a-trypsin inhibitor and its close relatives. In Proteinase inhibitors (Barett AJ, Salvesen G. eds.) Elsevier, Amsterdam (1986).

Gebhard W, Schreitmuller T, Hochstrasser K, Wachter E. Two out of three kinds of subunits of inter-a-trypsin inhibitor are structurally related. Eur. J. Biochem. 181; 571-576 (1989).

Gosling P, Sutcliffe A. Proteinuria following trauma. Ann. Clin. Biochem. 23; 681-685 (1986).

Greenwood FC, Hunter WM, Glover JS. The preparation of 1311-labelled human growth hormone of high specific radioactivity. Biochem. J. 89; 114- 123 (1963).

Heinrich PC, Castell JV, Andus T. Review article. Interleukin-6 and the acute phase response. Biochem J. 265; 621-636 (1990).

Hochstrasser K, Bretzel G, Feuth H, Hilla W, Lempart K. The inter-a-trypsin inhibitor as precursor of the acid-stable proteinase inhibitors in human serum and urine. Hoppe-Seyler's Z. Physiol. Chem. 357; 153-162 (1976).

Hochstrasser K, Feuth H, Steiner O. Zur Charakterisierung der säurestabilen Proteaseinhibitoren aus Humanplasma. Hoppe-Seyler's Z. Physiol. Chem. 354; 927-932 (1973).

Hochstrasser K, Schönberger ÖL, Rossmanith I, Wachter E. Kunitz-type proteinase inhibitors derived by limited proteolysis of the inter-a-trypsin inhibitor^. Hoppe-Seyler's Z. Physiol. Chem. 362; 1357-1362 (1981).

34 Ishigami M, Eguchi M, Yabuki So. Beneficial effects of the urinary trypsin inhibitor urinastatin on renal insults induced by gentamycin and mercuric chloride (HgCl?) poisoning. Nephron 58; 300-305 (1991).

Jessen TE, Faarvang KL, Ploug M. Carbohydrate as covalent crosslink in human inter-a-trypsin inhibitor: a novel plasma protein structure. FEBS leu.. 230; 195-200 (1988).

Johansson BG. Agarose gel electrophoresis. Scand. J. Clin. Lab. Invest. 29 suppl. 7-19 (1972).

Kaumeyer JF, Polazzi JO, Kotick MP. The mRNA for a proteinase inhibitor related to the HI-30 domain of inter-a-trypsin inhibitor also encodes a-1- microglobulin (protein HC). Nucleic Acids Res. 14; 7839-7850 (1986).

Kenny AJ. Proteinases associated with ceil membranes. In Proteinases in mammalian cells and tissues (Barett AJ. ed.) Elsevier, Amsterdam 393-444 (1977)

Kuwajima S, Matsui T, Kitahashi S, Kishida T, Noda T, Izumi Y, Naka K, Okuda K. Automated measurement of trypsin inhibitor in urine with a centrifugal analyzer: comparison with other acute phase reactants. Clin. Biochem. 23; 167-171 (1990).

Lankisch PG, Wolfrum DI, Koop H, Winckler K. Amylase/creatinine clearance ratio and tubular proteinuria in acute pancreatitis. Digestion 19; 375-379 (1979).

Laroui S, Balduyck M, Mizon C, Selloum L, Mizon J. Plasma proteins immunologically related to inter-a-trypsin inhibitor. Biochim. Biophys. Ada 953; 263-268(1988).

Lasson Å, Ohlsson K. Protease inhibitors in acute human pancreatitis. Correlation between biochemical changes and clinical course. Scand. J. Gastroent. 19; 779-786 (1984).

Laurell C-B. Electroimmuno assay. Scand. J. Clin. Lab. Invest. 29 suppl. 21-

t 35 37(1972).

Laurell C-B, Eriksson S. The electrophoretic aj-globulin pattern of serum in ajantitrypsin deficiency. Scand. J. Clin. Lab. Invest. 15; 132-140 (1963).

Liony Co, Sesboiié R, Manchon ND, Bercoff E, Delzant G, Martin JP. Inter- alpha-trypsine inhibiteur et ses derives dans les syndromes inflammatoires. Presse Méd. 20; 203-206 (1991).

Niaruyama M, Sumi H, Akazawa K, Mihara H. Marked increase of plasma acid stable/soluble trypsin inhibitor in hemodialysis patients. Clin. Chim. Acta 138; 205-213 (1984).

McKeehan WL, Sakagami Y, Hoshi H, McKeehan KA. Two apparent human endothelial cell growth factors from human hepatoma cells are tumor- associated proteinase inhibitors. J. Biol. Chem. 261; 5378-5383 (1986).

Meier PB, Levitt MD. Urine protein excretion in acute pancreatitis. /. Lab. Clin. Med. 108;628-634 (1986).

Mogensen CE, Soiling K. Studies on renal tubular protein reabsorption: Partial and near complete inhibition by certain amino acids. Scand. J. Clin. Lab. Invest. 37; 477-486 (1977).

Mogensen CE, Vittinghus E, Soiling K. Increased urinary excretion of albumin light chains, and p2-microglobulin after intravenous arginine administration in normal man. Lancet, sept 27, 581-583 (1975).

Moore S. On the determination of cystine as cysteic acid. J. Biol. Chem. 238; 235-237 (1963).

Muramatu M, Mori S, Matsuzawa Y, Horiguchi Y, Nakanishi Y, Tanaka M. Purification and characterization of urinary trypsin inhibitor, UTI 68- from normal human urine, and its cleavage by human uropepsin. /. Biochem. 88; 1317-1329(1980).

Miiller E. Uber das Verhalten des proteolytischen Leukocytenfermentes und seines 'Antifermentes' in den normalen und krankhaften Ausscheidungen des

36 menschlichen Körpers. Dt. Arch. klin. Med. 92; 199-216 (1908).

Nishino N, Aoki K, Tokura Y, Sakaguchi S, Fujie M, Sugawara Y, Takada Y, Takada A. Measurement of urinary trypsin inhibitor in urine, plasma and cancer tissues of patients with stomach cancer. Haemostasis 19; 112-119 (1989).

Nowak S. The urinary trypsin inhibitors in some diseases. Enzym. Biol. Clin. 9; 41-52 (1968).

Ogawa M, Nishibe S, Mori T, Neumann S. Effect of human urinary trypsin inhibitor on granulocyte elastase activity. Res. Commun. Chem. Pathol. Pharmacol. 55; 271-274 (1987).

Ohlsson K. Collagenase and elastase released during peritonitis are complexed by plasma protease inhibitors. Surgery 79; 652-657 (1976b).

Ohlsson K, Ganrot PO, Laurell C-B. In vivo interaction between trypsin and some plasma proteins in relation to tolerance to intravenous infusion of trypsin in dog. Ada Chir. Scand. 137; 113-121 (1971).

Ohlsson K, Olsson I. The neutral proteases of human granulocytes. Isolation and partial characterization of granulocyte elastase. Eur. J. Biochem. 42; 519-527 (1974).

Ohlsson K, Skude G. Demonstration and semiquantitative determination of complexes between various proteases and human 012-macroglobulin. Clin. Chim. Ada 66; 1-7 (1976a).

Ohnishi H, Kosuzume H, Ashida Y, Kato K, Honjo I. Effects of urinary trypsin inhibitor on pancreatic enzymes and experimental acute pancreatitis. Dig. Dis. Sci. 29; 26-32 (1984).

Okumichi T, Nishiki M, Takasugi S, Toki N, Ezaki H. Isolation of urinary trypsin inhibitor-like inhibitor from human lung cancer tissue. Cancer Res. 44; 2011-2015 (1984).

37 Potempa J, K.won K, Chawla R, Travis J. Inter-oc-trypsin inhibitor. Inhibition spectrum of native and derived forms J. Biol. C hem. 264; 15109-15114 (1989).

Pratt C, Pizzo SV. In vivo metabolism of inter-a-trypsin inhibitor and its proteinase complexes: evidence for proteinase transfer to c^-macroglobulin and ai-proteinase inhibitor. Arch. Biochem. Biophys. 248; 587-596 (1986).

Proksch GJ, Routh JI. The purification of the trypsin inhibitor from human pregnancy urine. J. Lab. Clin. Med. 79; 491-499 (1972).

Reading M. Technical method. A digestion technique for the reduction of background staining in the immunoperoxidase method. Technical methods..88-90 (1976).

Rudman D, Chawla RK, Nixon DW, Vogler WR, Keller JW, MacDonell RC. Proteinuria with disseminated neoplastic disease. Cancer Res. 39; 699-703 (1979).

Sakakibara K, Urano T, Takada Y, Takada A. Urinary UK, t-PA and urinary trypsin inhibitor in health and glomerular diseases. Thromb. Res. 56; 239- 249 (1989).

Salier J-P, Sesboiié R, Vercaigne D, Bourguignon J, Martin J-P. Inter-a- trypsininhibitor (ITI): use of new antisera for qualitative studies and discrete quantitation of ITI and its derivates. Anal. Biochem. 133; 336-343 (1983).

Scheidegger JJ. Une micro-méthode de 1'immuno-électrophorese. Int. Arch. Allergy l;\03-U0 (1955).

Schreitmiiller T, Hochstrasser K, Reisinger P, Wachter E, Gebhard W. cDNA cloning of human inter-a-trypsin inhibitor discloses three different proteins. Biol. Chem. Hoppe-Seyler 368;963-970 (1987).

Sesboiié R, Basuyau JP, Salier JP. Quantitative study of human blood alpha-1- antitrypsin, alpha-2-macroglobulin and inter-alpha-trypsin-inhibitor with respect to plasma renin activity. Clin. Chim. Acta 135; 23-33 (1983).

38 Shulman NR. A proteolytic inhibitor with anticoagulant activity separated from human urine and plasma. J. Biol. Chem. 213; 655-671 (1955).

Sjöberg EM, Sjöquist M, Larsson B, Fries E. After its secretion from rat hepatocytes bikunin is rapidly cleared from the plasma. Thesis, Uppsala (1991).

Sottrup-Jensen L. a-macroglobulins: structure, shape, and mechanism of proteinase complex formation. /. Biol. Chem. 264; 11539-11542 (1989).

Spackman DH, Stein WH, Moore S. Automatic recording apparatus for use in the chromatography of amino acids. Anal. Chem30; 1190-1206 (1958).

Spiro RG. [1] Analysis of sugars found in glycoproteins. [2] Characterization of carbohydrate units of glycoproteins. Methods enzymol. 8; 3-52 (1966).

Sternberger LA, Hardy PH, Cuculis JJ, Meyer HG. The unlabeled antibody enzyme method of immunohistochemistry. Preparation and properties of soluble antigen-antibody complex (horseradish peroxidase-antihorseradish peroxidase) and its use in identification of spirochetes. /. Histochem. Cytochem. 18; 315-333 (1970).

Sugiki M, Sumi H, Maruyama M, Yoshida E, Mihara H. Clearance and distribution of acid-stable trypsin inhibitor (ASTI). Enzyme 42; 31-38 (1989).

Sumi H, Maruyama M, Mihara H, Maehara S, Toki N. Human urinary trypsin inhibitors purified by affinity chromatography on trypsin-sepharose. Ada Haematol. Jpn. 44; 146-153 (1981).

Sumi H, Takada Y, Takada A. Studies on human urinary trypsin inhibitor 1. Its modification on treatment of urine with acid. Thromb. Res. 11, 747-754 (1977).

Swaim MW, Pizzo SV. Modification of the tandem reactive centres of human inter-cc-trypsin inhibitor with butanedione and cis-dichlorodiammineplatinum (II). Biochem. J. 254; 171-178 (1988).

39 Tanaka Y, Maehara S, Sumi H, Toki N, Moriyama S, Sasaki K. Purification and partial characterization of two forms of urinary trypsin inhibitor. Biochim. Biophys. Acta. 705; 192-199 (1982).

Thompson RC, Ohlsson K. Isolation, properties, and complete amino acid sequence of human secretory leukocyte protease inhibitor, a potent inhibitor of leukocyte elastase. ProcMatl. Acad. Sci. 83; 6692-6696 (1986).

Thorell JI, Johansson BG. Enzymatic iodination of polypeptides with 125 T to high specific activity. Biochim. Biophys. Acta 251; 363- 369 (1971).

Toki N, Sumi H. Urinary trypsin inhibitor and urokinase activities in renal diseases. Acta haemat. 67; 109-113 (1982).

Tominaga M, Takeda H, Muramatu M. Interaction of urinary trypsin inhibitor, UTI68, with bovine trypsin. J. Biochem. 91; 1391-1401 (1982).

Traboni C, Tosini F, Covone A, Romeo G, Rocchi M. The gene coding for proteins HC and HI-30 of inter- alpha-trypsin inhibitor maps to 9q22.3-»q33. Cytogenet. Cell Genet. 50;46-48 (1989).

Travis J. Structure, function, and control of neutrophil proteinases. Am. J. Med. 84 suppl 6A; 37-42 (1988)

Usui T, Maehara S, Kawashita E, Ishibe T, Sumi H, Toki N. Radioimmunological quantitation of urinary trypsin inhibitor. Enzyme 31; 11-16(1984).

Warren L. The thiobarbituric acid assay of sialic acids. J. Biol. Chem. 234; 1971-1975(1959).

Weber K, Osborn M. The reliability of molecular weight determinations by dodecyl sulfate-polyacrylamide gel electrophoresis. J. Biol. Chem. 244 4406- 4412(1969).

Yoshida E, Sumi H, Maruyama M, Mihara H, Sakai R. Development and application of an enzyme-linked immunosorbent assay for acid stable trypsin

40 inhibitor (ASTI) in human plasma. Clin. Chim. Ada 167; 155-164 (1987).

Yoshida E, SunA H, Maruyama M, Tsushima H, Matsuoka Y, Sugiki M, Mihara H. Distribution of acid stable trypsin inhibitor immunoreactivity in normal and malignant human tissues. Cancer 64; 860-869 (1989).

Yoshida E, Sumi H, Tsushima H, Maruyama M, Mihara H. Distribution and localization of inter-a-trypsin inhibitor and its active component acid-stable proteinase inhibitor: comparative immunohistochemical study. Inflammation 15;71-79 (1991).

ödum L. Immunohistochemical investigation of inter-a-trypsin inhibitor in the urinary tract. APMIS 97; 357-360 (1989).

ödum L, Hansen-Nord G, Byrjalsen I. Human inter-a-trypsin inhibitor and immunologically related inhibitors investigated by quantitative immunoelectrophoresis. II. Pathological conditions. Clin. Chim. Ada 162; 189-198 (1987).