Human tissue kallikrein. I. Isolation and characterization of human pancreatic kallikrein from duodenal juice

Human pancreatic kallikrein was purified from duodenal juice by ion exchange chromatography on DEAE-Sepharose and immunoaffinity chromatography. Thus, an enzyme preparation with a specific activity (using Ac-Phe-Arg-OEt as substrate) of 1 000 U/mg protein was obtained. A specific biological activity of 1310 KE/mg protein was measured in the dog blood pressure assay and of 0.361 HMW kininogen-U/mg, corresponding to the liberation of 383 micrograms bradykinin-equivalents per mg enzyme per min from HMW kininogen in the rat uterus assay. In dodecyl sulfate gel electrophoresis one protein band corresponding to a molecular mass of 27 kDa was obtained. Using gel filtration on Ultrogel AcA-44 a molecular mass of 40 kDa was measured. The amino-acid composition was determined and isoleucine and alanine were identified as the only N-terminal amino-acid residues. On isoelectric focusing four protein bands with isoelectric points of 5.60, 5.65, 5.70 and 5.85 were separated. The bimolecular velocity constant for the inhibition by diisopropyl fluoro phosphate was determined as 10.5 l x mol-1 x min-1. The dissociation constant Ki of the human pancreatic kallikrein-aprotinin complex was calculated to be 1.5 x 10(-10)M. The kinetic constants for the kallikrein-catalysed hydrolysis of Ac-Phe-Arg-OEt and D Val-Leu-Arg-Nan were determined. Immunological studies showed a close relationship between the human pancreatic kallikrein and other human tissue kallikreins, especially with human urinary kallikrein. Detergents such as Triton X-100, Tween 20 and lysolecithin, as well as human serum albumin, activated the human pancreatic kallikrein preparation.     

Isolation and characterization of human urinary kallikrein

Human urinary kallikrein was purified by gel filtration on Sephacryl S-200 and affinity chromatography on aprotinin-Sepharose, followed by ion exchange chromatography on DEAE-Sepharose. Thus an enzyme preparation with a specific activity (using AcPheArgOEt as substrate) of 1 100 U/mg protein was obtained. A specific bioligical activity of 2 300 KE/mg was measured in the dog blood pressure assay and of 0.742 HMW kininogen-U/mg, corresponding to the liberation of 787 micrograms bradykinin per mg enzyme per min from HMW-kininogen, in the rat uterus assay. In dodecyl sulfate electrophoresis two protein bands with apparent molecular weights of 41 000 and 34 000 were separated. The amino acid composition was determined and isoleucine was identified as the only aminoterminal amino acid. On isoelectric focusing six protein bands with isoelectric points of 3.75, 3.80, 3.90, 4.00, 4.05 and 4.25 were separated. The kinetic constants for the kallikrein-catalyzed hyrdolysis of AcPheArgOEt and D ValLeuArgNHNp were determined. The bimolecular velocity constant for the inhibition by diisopropyl fluorophosphate was determined as 9 +/- 2l x mol-1 x min-1. Immunological studies showed that a close relationship exists between the urinary enzyme and other human glandular kallikreins. Deoxycholate, lysolecithin and other amphiphiles activated human urinary kallikrein.     

Similarity between a kininogenase (kallikrein) from human large intestine and human urinary kallikrein.

A human colon kininogenase (kallikrein) was isolated by gel filtration on Sephacryl S-200 and affinity chromatography on Trasylolbound Sepharose, yielding a material with a specific activity of 1.3 U/mg (substrate: AcPheArgOEt). The molecular weight of the enzyme as estimated by gel filtration is approximately 70 000. After reduction with mercaptoethanol two bands were obtained in dodecyl sulfate eletrophoresis with molecular weights of 27 000 and 70 000. The bimolecular velocity constant for the inhibition by diisopropyl fluorophosphate was determined as 4 l x mol-1 x min-1. The preparation was characterized by immunological and enzymatic methods. Using the radioimmumoassay for human urinary kallikrein cross-reactivity and parallel binding curves were obtained. Kinin liberation from human high Mr-kininogen was totally inhibited by antibodies directed against human urinary kallikrein. Trasylol and diisopropyl fluorophosphate, but not by antibodies directed against human trypsin and plasma kallikrein. The effect on dog blood pressure was comparable to that obtained with human urinary kallikrein. The amino acid composition of human large intestine kallikrein is very similar to that of human urinary kallikrein.     

Isolation of an enzymatically active tissue kallikrein from human seminal plasma by immunoaffinity chromatography

A tissue kallikrein from human seminal plasma was isolated by immunoaffinity chromatography and characterized. Its molecular mass was determined by gel filtration to be approximately 40000 Da. The enzyme preparation liberates kinin from human HMW kininogen (specific activity: 0.594 HMW kininogen-U/mg), lowers the blood pressure of dogs after intravenous injection (specific activity: 1740 biol. kallikrein unit/mg) and is strongly inhibited by aprotinin but not by soybean trypsin inhibitor. N alpha-Acetyl-L-phenylalanyl-L-arginine ethyl ester, D-valyl-L-leucyl-L-agrine ethyl ester and N-benzyloxycarbonyl-L-tyrosine p-nitrophenyl ester are cleaved with identical rates by the enzyme from human seminal plasma and human urinary kallikrein.     

Cloning and expression of human salivary-gland kallikrein in Escherichia coli.

CDNA clones for human kallikrein have been identified in a cDNA library constructed from mRNA of human salivary gland. The entire coding sequence for preprokallikrein and for the 5'- and 3'-untranslated regions were isolated by using a mixture of oligonucleotides corresponding to amino acids 51-56 of human urinary kallikrein and one oligonucleotide corresponding to amino acids 233-238 of human pancreatic kallikrein. The DNA sequence proved that, with the exception of two amino acid exchanges, kallikrein of the human salivary gland is identical with pancreatic kallikrein. Salivary gland and renal kallikrein was expressed in Escherichia coli from plasmid pKK223-3 under the control of the tac promoter. The protein was identified by Western-blot analysis and by demonstration of its specific proteolytic activity.     

Purification and some physico-chemical and enzymatic properties of tissue kallikrein from human urine

A procedure for obtaining tissue kallikrein (EC 3.4.21.35) from large specimens of human urea (100 l) has been developed. The isolation procedure included primary extraction of the protein with chitosan (a crustacean chitin deacylated by alkaline treatment), desorption from chitosan with 1 M NH3, affinity chromatography on contrical-Sepharose, ion-exchange chromatography on DEAE-Sepharose and gel filtration on Sephadex G-100. This method permits to obtain tissue kallikrein preparations purified 1080-fold (with respect to AcPheArg-OEt esterase) and 1360-fold (with respect to kininogenase) with 33 and 40% yields, respectively. Tissue kallikrein preparations were homogeneous as could be judged from the results of electrophoresis performed in 12% PAAG in the presence of 0.1% SDS as well as from the presence of one N-terminal amino acid identified as isoleucine. Purified tissue kallikrein had specific activities of 133 mumol/min/mg protein (with respect to AcPheArg-OEt hydrolysis) and 8.8 mumol/min/mg protein (with respect to D-Val-Leu-Arg-pNa hydrolysis) and liberated 462 micrograms equiv. of bradykinin/min/mg protein from heated human blood plasma used as a kininogen source. The protein exhibited the highest stability at pH 8.0-9.0; the pH optimum is at pH 8.0 with AcPheArg-OMe as substrate. The enzyme revealed a high thermostability and was fully inactivated only after 1-hour heating in a boiling water bath. The identity of the urine enzyme to tissue kallikrein could be confirmed by the resistance of the enzyme activity to SIT, high sensitivity to the inhibiting effect of aprotinin (Ki = 0.94 x 10(-10) M) and by an exceedingly low value of the second order inhibition constant for DPP (4.6 M-1 min-1). The fact that this value differs drastically from that for human blood plasma kallikrein (EC 3.4.21.34) which is equal to 360 M-1 min-1 points to marked differences in the structure of the active centers of the both kallikreins as well as to the uniqueness of the tissue kallikrein active center.     

Purification and properties of guinea-pig submandibular-gland kallikrein.

Guinea-pig submandibular kallikrein has been purified from the glands to electrophoretic homogeneity by conventional procedures. The enzyme is active as a kininogenase, releasing kallidin at a rate of 462 micrograms/min per mg of protein from bovine kininogen, and proved potently hypotensive in the guinea pig and in the dog, properties which indicate its tissue kallikrein nature. The specific activity determined on the substrate N-alpha-benzoyl-L-arginine ethyl ester (11.1 mumol/min per mg of protein) is much lower than that measured with N-acetyl-L-phenylalanyl-L-arginine ethyl ester (483 mumol/min per mg of protein). The latter value is of an order of magnitude comparable with the specific activities of other tissue kallikreins determined with this sensitive kallikrein substrate. The enzyme is a glycoprotein consisting of 237 amino acid residues and containing three to four glucosamine molecules. Its amino acid composition is not identical with that reported for guinea-pig coagulating-gland kallikrein, but is remarkably similar to that of the porcine tissue kallikreins. Apparent Mr values are 29000 (sodium dodecyl sulphate/polyacrylamide-gel electrophoresis) or 34000 (gel filtration). The amino acid sequence of the first 31 N-terminal residues was determined and was found to be closely homologous with that of other tissue kallikreins.     

Horse urinary kallikrein, I. Complete purification and characterization

The isolation procedure for horse urinary kallikrein was considerably improved by the introduction of two new purification steps: a) removal of mucoproteins and concentration of the urine by ultrafiltration and b) affinity chromatography on benzamidine-Sepharose conjugate. The homogeneity of the enzyme preparations, regarding their protein moiety, was demonstrated by: 1) a single symmetric peak on DEAE-Sephadex chromatography, with constant values for A280/A260 ratios, esterolytic and amidolytic specific activities; 2) a single band, although dispersed, on gel-electrophoresis at pH 8.3, also in the presence of sodium dodecyl sulfate, and 3) a unique sequence for the six amino-terminal residues. The isolated enzyme was shown to be a single chain glycoprotein (alpha-kallikrein), similar to human urinary and porcine-pancreatic kallikreins regarding the protein moiety molecular mass, amino-acid composition, and partial amino-terminal sequence; differences were found in their total sugar content and even more conspicuously in their carbohydrate composition. In contrast to porcine pancreatic beta-kallikrein, horse urinary kallikrein was not substrate-activated and unlike other alpha-kallikreins, did not present the biphasic time-course in benzoyl-L-arginine ethyl ester hydrolysis. The specificity constants (kcat/Km) for ester and 4-nitroanilide substrates were lower for horse urinary than for pancreatic beta-kallikrein and as observed with the latter enzyme, were affected by NaCl.     

Purification and chemical studies on rat urinary kallikrein.

A highly purified kallikrein was obtained from rat urine by chromatography on DE-32 cellulose, affinity chromatography on Bio-gel P-200-Aprotinin and gel filtration over Sephadex G-100 coarse and superfine. A molecular weight of 32,000 by sodium dodecyl sulfate polyacrylamide disc gel electrophoresis was estimated. The aminoacid composition and the esterase activity of the purified material were determined. Biological characterization of the purified kallikrein was tested by liberation of a kinin from rat plasma kininogen, by direct action on the isolated rat uterus and by the lowering of rat arterial pressure after intravenous injection of the enzyme. The preparation of insoluble derivative of Aprotinin is described herein. The polymer used as insoluble support (Bio-gel P-200) was before changed to its corresponding azide, which reacts with Aprotinin; the product maintained the binding property of the Aprotinin with urinary kallikrein.     

Kallistatin: a novel human tissue kallikrein inhibitor. Purification, characterization, and reactive center sequence.

A novel human tissue kallikrein inhibitor designated as kallistatin has been purified from plasma to apparent homogeneity by polyethylene glycol fractionation and successive chromatography on heparin-Agarose, DEAE-Sepharose, hydroxylapatite, and phenyl-Superose columns. A purification factor of 4350 was achieved with a yield of approximately 1.35 mg per liter of plasma. The purified inhibitor migrates as a single band with an apparent molecular mass of 58 kDa when analyzed on SDS-polyacrylamide gel electrophoresis under reducing conditions. It is an acidic protein with pI values ranging from 4.6 to 5.2. No immunological cross-reactivity was found by Western blot analyses between kallistatin and other serpins. Kallistatin inhibits human tissue kallikrein's activity toward kininogen and tripeptide substrates. The second-order reaction rate constant (ka) was determined to be 2.6 x 10(4) M-1 s-1 using Pro-Phe-Arg-MCA. The inhibition is accompanied by formation of an equimolar, heat- and SDS-stable complex between tissue kallikrein and kallistatin, and by generation of a small carboxyl-terminal fragment from the inhibitor due to cleavage at the reactive site by tissue kallikrein. Heparin blocks kallistatin's complex formation with tissue kallikrein and abolishes its inhibitory effect on tissue kallikrein's activity. The amino-terminal residue of kallistatin is blocked. Sequence analysis of the carboxyl-terminal fragment generated from kallistatin reveals the reactive center sequence from P1' to P15', which shares sequence similarity with, but is different from known serpins including protein C inhibitor, alpha 1-antitrypsin, and alpha 1-antichymotrypsin. The results show that kallistatin is a new member of the serpin superfamily that inhibits human tissue kallikrein.     

The role of glandular kallikrein in the formation of a salivary proline-rich protein A by cleavage of a single bond in salivary protein C.

An enzyme was purified from human parotid saliva that can cleave a single arginine-glycine peptide bond between residues 106 and 107 in human salivary proline-rich protein C, hereby giving rise to another proline-rich protein A, which is also found in saliva. The enzyme was purified 2400-fold. It cleaved salivary protein C at the rate of 59 micrograms of protein/h per microgram of enzyme and had amino acid composition, molecular weight and inhibition characteristics similar to those reported for human salivary kallikrein. Confirmation that the enzyme was kallikrein was demonstrated by its kinin-generating ability. Histochemical evidence indicates that a post-synthetic cleavage of protein C by kallikrein would have to take place during passage of saliva through the secretory ducts. In secreted saliva, cleavage of salivary protein C can only be observed after 72 h incubation. In addition, there is no effect of salivary flow rate on the relative amounts of proteins A and C in saliva. On the basis of the experimental observations, it is proposed that in vivo it is unlikely that kallikrein secreted from ductal cells plays a significant role in converting protein C into protein A.     

Enzyme- and immuno-histochemical localization of kallikrein. II. The human kidney

Localization of kallikrein in the human kidney was investigated by two markers: kallikrein-like activity and kallikrein antigenicity. Kallikrein-like activity was demonstrated enzyme-histochemically by using a synthetic substrate for kallikrein, pro-phe-arg-naphthyl-ester. Kallikrein antigenicity was demonstrated by the unlabelled antibody peroxidase-antiperoxidase method using an antiserum against human urinary kallikrein. The kallikrein-like activity was localized in the proximal tubular cells without any corresponding kallikrein antigenicity. Neither kallikrein-like activity nor kallikrein antigenicity was noticed in any other tubular cell. These results are contrary to those in the ductal cells of the human parotid gland where the kallikrein-like activity and the kallikrein antigenicity were identical in their locations. The peroxidase-antiperoxidase method revealed, for the first time, kallikrein antigenicity both in the interstitium and in the basement membrane region of Bowman's capsule and of all the tubules, possibly representing circulating glandular kallikreins deposited in the renal tissue. Thus, the present findings are consistent with the hypothesis that the urinary (renal) kallikreins are derived from circulating glandular kallikreins.     

Purification and characterization of a kallikrein from human submaxillary glands

A tissue kallikrein was purified over 1500-fold from the postmicrosomal supernatant of human submaxillary glands. The purified enzyme gave a single band, corresponding to an apparent molecular weight of 42,000 on SDS-polyacrylamide gel electrophoresis. This enzyme cross-reacted with the anti-human urinary kallikrein antiserum. The purified enzyme was characterized in comparison with the purest human urinary kallikrein preparation. Both enzymes hydrolyzed the synthetic substrate, Ac-Phe-Arg-OMe, most effectively. Aprotinin, TLCK, and PMSF suppressed the enzyme activities, while SBTI, LBTI, and alpha 1-antitrypsin had no effect at all. The purified enzyme generated kinin from the natural substrate, kininogen. It was concluded therefore that the purified enzyme is a typical tissue kallikrein.     

Dog renal kallikrein: purification and some properties.

The contents of kallikrein [EC 3.4.21.8] in the kidneys of various animals were estimated and the activity was found to be most potent in dogs. The dog renal kallikrein (DRK) was located mainly in the kidney cortex. Following the activation of a dog kidney cortex homogenate with acetone, kallikrein was purified about 2,000-fold with an overall yield of 18% by diethylaminoethyl (DEAE)-cellulose adsorption, acetone fractionation, and chromatography on Sephadex G-75 and DEAE-Sephadex A-50. The final purified preparation of dog renal kallikrein had a vasodilator activity of 65.5 KU per A280, and appeared to be homogeneous both in disc electrophoresis and ultracentrifugal analysis. Its molecular weight was estimated to be approximately 3.8 X 10(4) from the sedimentation coefficient obtained by ultracentrifugation, and by Sephadex gel filtration. However, isoelectric fractionation of the purified DRK preparation gave three isoelectric point, 3.9, 4.1, and 4.3. The DRK had an optimum pH of about 8.6 and was stable at pH 8. This enzyme was hardly inhibited by Trasylol, soybean trypsin inhibitor, ovomucoid trypsin inhibitor or potato kallikrein inhibitors. These properties were compared with those of kallikrein from other sources; DRK appeared to be similar to urinary kallikrein.     

Enzyme- and immuno-histochemical localization of kallikrein

Summary Localization of kallikrein in the human kidney was investigated by two markers: kallikrein-like activity and kallikrein antigenicity. Kallikrein-like activity was demonstrated enzyme-histochemically by using a synthetic substrate for kallikrein, pro-phe-arg-naphthyl-ester. Kallikrein antigenicity was demonstrated by the unlabelled antibody peroxidase-antiperoxidase method using an antiserum against human urinary kallikrein. The kallikrein-like activity was localized in the proximal tubular cells without any corresponding kallikrein antigenicity. Neither kallikrein-like activity nor kallikrein antigenicity was noticed in any other tubular cell. These results are contrary to those in the ductal cells of the human parotid gland where the kallikrein-like activity and the kallikrein antigenicity were identical in their locations. The peroxidase-antiperoxidase method revealed, for the first time, kallikrein antigenicity both in the interstitium and in the basement membrane region of Bowman's capsule and of all the tubules, possibly representing circulating glandular kallikreins deposited in the renal tissue. Thus, the present findings are consistent with the hypothesis that the urinary (renal) kallikreins are derived from circulating glandular kallikreins.     

Enzyme- and immuno-histochemical localization of kallikrein. I. The human parotid gland

Localization of kallikrein in the human parotid gland was investigated simultaneously by two markers: kallikrein-like (enzyme) activity and kallikrein antigenicity. Kallikrein-like activity was histochemically demonstrated by using a synthetic substrate, pro-phe-arg- naphthylester . Kallikrein antigenicity was demonstrated by an unlabelled antibody peroxidase-antiperoxidase method, where monospecific antiserum against highly purified urinary kallikrein was used as the primary antiserum. The results showed that kallikrein-like activity and kallikrein antigenicity were identical in their locations in the ductal cells, being localized in the luminal part of the striated ducts and to a lesser degree in the excretory ducts. This indicates the presence of active kallikrein in these regions. No enzyme activity nor antigenicity was observed either in acini or in intercalated ducts. Moreover, the peroxidase-antiperoxidase method revealed kallikrein antigenicity for the first time extracellularly in the basement-membrane region of acini and of ducts as well as in the interstitium surrounding ducts and major vessels.     

Enzyme-and immuno-histochemical localization of kallikrein

Summary Localization of kallikrein in the human parotid gland was investigated simultaneously by two markers: kallikrein-like (enzyme) activity and kallikrein antigenicity. Kallikrein-like activity was histochemically demonstrated by using a synthetic substrate, pro-phe-arg-naphthylester. Kallikrein antigenicity was demonstrated by an unlabelled antibody peroxidase-antiperoxidase method, where monospecific antiserum against highly purified urinary kallikrein was used as the primary antiserum. The results showed that kallikrein-like activity and kallikrein antigenicity were identical in their locations in the ductal cells, being localized in the luminal part of the striated ducts and to a lesser degree in the excretory ducts. This indicates the presence of active kallikrein in these regions. No enzyme activity nor antigenicity was observed either in acini or in intercalated ducts. Moreover, the peroxidase-antiperoxidase method reveated kallikrein antigenicity for the first time extracellularly in the basement-membrane region of acini and of ducts as well as in the interstitium surrounding ducts and major vessels.     

Inhibition of tissue kallikrein by protein C inhibitor. Evidence for identity of protein C inhibitor with the kallikrein binding protein.

We studied the inhibition of tissue kallikrein by protein C inhibitor (PCI), a relatively unspecific heparin-dependent serine protease inhibitor present in plasma and urine. PCI inhibited the amidolytic activity (cleavage of H-D-valyl-L-leucyl-arginine-p-nitroaniline) of urinary kallikrein with an apparent second order rate constant of 2.3 x 10(4) M-1 s-1 and formed stable complexes (85 kDa) with urinary kallikrein as judged from silver-stained sodium dodecyl sulfate-polyacrylamide gels. Complex formation was time-dependent and was paralleled by a decrease in the intensity of the main PCI protein band (Mr = 57,000) and an increase in the intensity of the lower Mr (54,000) PCI form (cleaved inhibitor). Heparin interfered with the inhibition of tissue kallikrein by PCI and with the formation of tissue kallikrein-PCI complexes in a dose-dependent fashion and completely abolished PCI-tissue kallikrein interaction at 300 micrograms/ml. This is in contrast to findings on the interaction of PCI with all other target proteases studied so far (i.e. stimulation of inhibition by heparin) but is similar to the reaction pattern of 125I-labeled tissue kallikrein with so called kallikrein binding protein described in serum and other systems. To study a possible relationship between PCI and this kallikrein binding protein we incubated 125I-labeled urinary kallikrein in serum and in PCI-immunodepleted serum in the absence and presence of heparin and analyzed complex formation using sodium dodecyl sulfate-polyacrylamide gel electrophoresis. In normal serum, formed complexes co-migrated with complexes of purified PCI and 125I-kallikrein and were less intense in the presence of heparin. No complex formation at all was seen in PCI-depleted serum. Our data indicate that PCI may be a physiologically important endogenous inhibitor of tissue kallikrein and provide evidence that PCI may be identical to the previously described kallikrein binding protein.     

Biochemical characterization and substrate specificity of rat prostate kallikrein (S3): comparison with tissue kallikrein, tonin and T-kininogenase.

A tissue kallikrein-like enzyme encoded by S3 mRNA was purified to homogeneity from rat prostate gland. The apparent molecular mass of the prostate enzyme is 32 kDa as determined by sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE). The intact 32 kDa enzyme is split into two bands of lower molecular mass, 18 and 14 kDa, under reducing conditions on SDS-PAGE. NH2-terminal amino acid sequence analyses of the intact enzyme and heavy and light chains revealed the identity to the translated sequence of a prostate kallikrein cDNA (S3). Isoelectric focusing indicated that the prostate enzyme is a basic protein with pI of 7.30-7.45. Specific activities of the prostate kallikrein toward angiotensin I, angiotensinogen and rat low M(r) kininogen as well as tripeptide chromogenic substrates were compared with those of tissue kallikrein, tonin and T-kininogenase. The kinin-releasing activity is inhibited by leupeptin, antipain, benzamidine and soybean trypsin inhibitor. A sensitive and specific radioimmunoassay for the rat prostate kallikrein shows that the immunoreactive kallikrein levels in prostate and submandibular gland were 23.78 +/- 2.62 micrograms/mg protein (n = 5) and 12.29 +/- 2.25 micrograms/mg protein (n = 5), respectively. The results indicate that the prostate kallikrein S3 is expressed at high levels in both prostate and submandibular glands.     

Immunohistochemical localization of glandular kallikrein in the endocrine and exocrine human pancreas

Fab fragments from two new monospecific anti-human tissue kallikrein sera were examined for their capacity to inhibit the functional activities of purified human urinary kallikrein and purified human pancreatic kallikrein. Fragments from a new anti-urinary kallikrein serum and from an anti-pancreatic kallikrein serum yielded mixed inhibition of kinin-generating activity and minimal inhibition of esterolytic activity. In contrast to the previously described "active site directed" anti-urinary kallikrein, these new antisera demonstrated little specificity for epitopes near the enzymatic site of urinary or pancreatic kallikrein. When used to localize kallikrein antigen in human pancreas obtained at surgery, IgG fractions of the new anti-kallikrein sera yielded moderate acinar and ductal staining in the absence of pretreatment of the tissue with trypsin or pronase. Short incubation with 0.125 mg/ml of either enzyme permitted the discrete localization of islet beta cell kallikrein antigen, while increased pronase concentrations decreased kallikrein antigen in both islets and exocrine tissue and led to islet destruction. Both antibody specificity and tissue preparation influence kallikrein localization in human pancreas.