Characterization of trypsinogens 1 and 2 in two human pancreatic adenocarcinoma cell lines; CFPAC-1 and CAPAN-1.

Proteins with trypsin-like immunoreactivity (first detected by a specific immunoenzymatic assay) were isolated from CAPAN-1 and CFPAC-1 cell culture-conditioned media by chromatography on an immunoadsorbent prepared with a polyclonal antibody directed against trypsin 1. The adsorbed proteins were devoid of free trypsin activity but trypsin activity was present after enterokinase activation demonstrating that the immunoreactive trypsin present in cell supernatants corresponds to trypsinogens. When characterised by Western blotting using a monoclonal antibody directed against human trypsin 1 two protein bands corresponding to trypsinogen 1 (23 kDa) and trypsinogen 2 (25 kDa) gave a positive reaction. These results demonstrate the presence of trypsinogens 1 and 2 in CAPAN-1 and CFPAC-1 cells and in their culture-conditioned media.

Immunocytochemical demonstration of a pancreatic secretory protein of unknown function in human duodenum.

We have previously isolated from human pancreatic juice a secretory glycoprotein of 19 KD (P19), devoid of known enzymatic activity. P19 gave by proteolysis a protein of 14 KD (P14), at first named protein X and also called pancreatic thread protein or pancreatic stone protein. Specific rabbit immunosera prepared against P19 and P14 were applied to localize these proteins in human small intestine. By comparison, antibodies directed against some human pancreatic enzymes (amylase, lipase, chymotrypsin, trypsinogen 1, trypsinogen 2, and trypsin 1) were also tested. Positive immunoreactivity was observed on Paneth cells with antisera directed against trypsinogens, trypsin 1, and P19-related proteins. In addition, antisera directed against P19-related proteins stained the columnar cells located in the crypts of Lieberkühn. These original findings are a further indication of the resemblance between Paneth and pancreatic acinar cells but show that their functional analogy is only partial. On the other hand, the presence of P19-related proteins on non-mature columnar cells suggests that this differential distribution is a consequence of differentiation.

Characterization and purification of a kallikrein from human pancreatic juice and immunological comparison with other kallikreins.

A prekallikrein has been demonstrated in human pancreatic juice and the active enzyme has been purified from this material. The purification procedure included filtration on Sephadex G-100, chromatography on DEAE-cellulose and affinity chromatography on trypsin-inhibitor Sepharose. The purified kallikrein appeared to be homogeneous by polyacrylamide gel electrophoresis at pH 8.3 and by immunoelectrophoresis. Human pancreatic kallikrein is immunologically different from human plasma kallikrein and from pancreatic kallikreins of other species (hog, cat, rat and dog). Human pancreatic kallikrein has common antigenic determinants with human urinary and submandibular kallikreins but probably not with parotid kallikrein.

Some properties of human pancreatic kallikrein and comparison with human trypsins and porcine kallikrein.

The properties of human pancreatic kallikrein purified from pancreatic juice were investigated. The enzyme is very stable at pH 8 but is rapidly inactivated at pH 2.6. It is a glycoprotein with a molecular weight of 35 000 as determined by gel filtration on Sephadex G-200. Contrary to the two human trypsins, human kallikrein like porcine pancreatic kallikrein is unable to hydrolyse casein and Met-Lys-bradykinin. Human pancreatic kallikrein is inactivated by diisopropyl fluorophosphate but not by chloro (N-p-toluolsulfonly-L-lysyl)methane. The enzyme does not react with various proteinase inhibitors (secretory pancreatic trypsin inhibitors, ovomucoid, lima bean and soybean trypsin inhibitors) but is inhibited by the Kunitz pancreatic trypsin inhibitor.

Characterization and N-terminal sequence of a degradation product of 14,000 molecular weight isolated from human pancreatic juice.

Chromatography of human pancreatic juice has allowed the isolation of an inactive protein of 14,000 Mr (protein X) and the determination of its amino acid composition and N-terminal sequence. Protein X was found to be immunologically identical to the protein extracted from precipitates present in the pancreatic juices of patients with chronic calcifying pancreatitis and recently shown to be a degradation product of trypsinogen 1. The same chromatography performed in the presence of lima bean trypsin inhibitor has permitted the isolation of precursors of approximately congruent to 19,000 Mr which can be transformed into protein X "in vitro" by chymotrypsin hydrolysis. These results emphasize the easy activation of human pancreatic zymogens and the possible consequences due to proteolysis in pancreatic disease.

Proteolysis of human trypsinogen 1. Pathogenic implication in chronic pancreatitis.

SDS electrophoresis on polyacrylamide gels of purified trypsinogen 1 has shown the occurrence of a proteolysis in some molecules during long storage at -20 degrees C. This proteolyzed trypsinogen gives a positive reaction with an antiserum directed against the precipitate protein, major protein of about 14 000 molecular weight extracted from precipitates present in the pancreatic juice of patients with chronic pancreatitis. The autoactivation of proteolyzed trypsinogen 1 liberates a polypeptide of 14 000 molecular weight which is immunologically identical to the precipitate protein. These results show that the major protein present in pancreatic precipitates (and pancreatic stones) of patients with chronic pancreatitis is a degradation product of trypsinogen 1 liberated by a proteolysis which necessarily requires a premature zymogen activation in the disease.

Immunolocalization of kallikrein in porcine pancreas at the ultrastructural level.

The immunological localization of kallikrein in porcine pancreatic tissue was investigated by an immunoferritin method and presented by electronmicroscopy. Kallikrein was found in the zymogen granules of the pancreatic acinar cells as well as in the golgian area of the same cells. On the same section, all the zymogen granules of all acinar cells gave a positive reaction. Kallikrein was not found in the centroacinar or interstitial cells.

Immunocytochemical localization of pancreatic kallikrein in human acinar cells.

The immunological localization of kallikrein in human pancreatic tissue was studied at the optical and subcellular levels. Data obtained by light microscopy, using indirect immunofluorescence and immunoperoxidase techniques, demonstrate the presence of kallikrein in pancreatic acinar cells. Ultrastructural localization was performed by using the immunocytochemical protein A-gold technique. Kallikrein was found at the level of the rough endoplasmic reticulum and mainly in the zymogen granules of pancreatic acinar cells. Kallikrein was not found in the centro-acinar or duct cells.

Lactoferrin-binding sites at the surface of HT29-D4 cells. Comparison with transferrin.

The binding of 125I-lactoferrin to HT29-D4 cells, a clone of HT29 cells, was studied and compared to the binding of 125I-transferrin to the same cells. The binding of the two iron-transport proteins is saturable and reversible suggesting the presence of specific receptors for each protein. Scatchard analysis suggests the existence of binding sites for lactoferrin with the relatively high equilibrium dissociation constant, Kd1 of 408 nM. Additionally, the cell is capable of binding large amounts of lactoferrin with very low affinity, probably in a non-receptor intermediate fashion. The dissociation constant of transferrin and its receptor was calculated 9.29 nM which corresponds well to values found in the literature. In contrast to lactoferrin, the cell was capable of binding only low amounts of transferrin in a non-receptor intermediate fashion. After chemical crosslinking of lactoferrin to the cell surface, the radiolabeled lactoferrin was found in a complex of molecular mass 300 kDa. Crosslinking of transferrin resulted in a complex of much higher molecular mass. These data clearly show a binding site for lactoferrin different from the transferrin receptor. Only if competition experiments were performed with a high molar excess of both ligand proteins did a small percentage of either of the two ligands crossreact with the receptor for the other, possibly due to a structural similarity of the two glycoproteins.

Comparison of immunological and enzymatic properties of human urinary and pancreatic kallikrein.

Human urinary and human pancreatic kallikreins were compared by radioimmunoassay, immunoelectrophoresis, by their enzymatic activity using the substrate Ac-Phe-Arg-OEt and by their biological activity employing the dog blood pressure assay. The only significant difference observed was a higher anodal mobility of the urinary kallikrein under the conditions of immunoelectrophoresis.

Effect of lactoferrin on the growth of a human colon adenocarcinoma cell line--comparison with transferrin.

Lactoferrin was examined for its effect on the growth of a human colon adenocarcinoma cell line (HT 29) in culture and its action was compared to that produced by transferrin and two different iron solutions (ferrous sulfate and ferric chloride). When transferrin was replaced by either iron solutions the cell grew in proportion to the quantity added and the maximal effect obtained was identical to that produced by transferrin alone. When transferrin was replaced by lactoferrin the cells were unable to proliferate for a long time. However, in the presence of low-concentration iron solutions, lactoferrin stimulated the cell growth, and the effect was more pronounced with the ferric chloride solution.

Protein X, a proteolysis product of human pancreatic juice. Immunological relationship with trypsinogen 1.

Protein X (PX) previously isolated from human pancreatic juice is an inactive protein of 14 kDa which has been shown to be a degradation product liberated by proteolysis of 19 kDa precursors. Polyclonal antibodies against P19 and PX were prepared in rabbits by injection of the two proteins purified by SDS polyacrylamide gel electrophoresis. These antibodies reacted with a form of trypsin 1 (DFP-trypsin 1) which was shown to be partly proteolysed. Immunological studies were performed with pancreatic juice proteins and partially purified trypsinogen 1 using antibodies directed against PX, P19 and trypsin 1. The results of immunoprecipitation and immunoadsorbent chromatography show that these different antisera recognized a protein of 25 kDa. Immunoblotting has permitted to characterize this protein as a trypsinogen 1-like molecule which would be a form of inert protein generated by uncontrolled trypsinogen activation.

Salivary secretion in chronic pancreatitis with special reference to albumin and lactoferrin.

Saliva from one parotid gland was collected under citric acid stimulation in three groups consisting of 69 control subjects and two groups consisting of 25 patients with chronic and relapsing chronic pancreatitis. Mixed saliva was collected under mechanical stimulation from 10 patients and 6 control subjects. Flow rates and the contents of bicarbonate, amylase and protein were determined. In a subgroup of patients and controls albumin and lactoferrin were measured. Wide inter- and intraindividual variations of secretory values were observed and normal values (which are reported in detail) were highly dependent from conditions of stimulation. Secretory patterns were not significantly different between patients and control subjects. It is concluded that the alterations leading to an elevation of albumin and lactoferrin and a fall in other secretory values are restricted to the exocrine pancreas and do not affect salivary glands in chronic pancreatitis.

Inhibition of human pancreatic proteinases by mucus proteinase inhibitor, eglin c and aprotinin.

The kinetic investigation of the inhibition of human pancreatic trypsin 1, trypsin 2 and chymotrypsin A by mucus proteinase inhibitor, eglin c and aprotinin reveals that (i) the first protein is a potent inhibitor of chymotrypsin A (kass. = 1.4 x 10(6) M-1.s-1, Ki = 71 pM) but forms loose complexes with trypsin 1 (Ki = 0.5 microM) and trypsin 2 (Ki = 18 nM), (ii) eglin c does not inhibit the two trypsins but forms a tight complex with chymotrypsin A (kass. = 3.3 x 10(6) M-1.s-1, Ki < 0.1 nM) and (iii) aprotinin is a potent inhibitor of trypsin 1 (kass. = 1 x 10(6) M-1.s-1, Ki < 0.2 nM) and trypsin 2 (kass. = 2.4 x 10(5) M-1.s-1, Ki < 1 nM) but forms a loose complex with chymotrypsin A (Ki = 0.17 microM). These data, together with those published previously on human pancreatic elastase, suggest that a cocktail of aprotinin + eglin c might be a better intensive-care drug for acute pancreatitis than aprotinin alone, because it will efficiently inhibit all four human pancreatic proteinases. On the other hand, human gastric juice inactivates mucus proteinase inhibitor by pepsin-mediated cleavage. This indicates that the fraction of mucus proteinase inhibitor that reaches the stomach following aerosol delivery to cystic fibrosis patients does not reach the duodenum in an active form and, therefore, does not aggravate the pancreatic insufficiency of these patients.

"In vivo" and "in vitro" inhibition of human pancreatic chymotrypsin A by serum inhibitors.

The interaction of human pancreatic chymotrypsin A with serum inhibitors was assessed by enzyme immunoassay, enzymatic activity and inhibitory capacity measurements and electrophoretic analyses. In normal serum, chymotrypsin A was detected in four forms: one form (Mr approximately equal to 25,000) which might be chymotrypsinogen A and three forms complexed to the main inhibitors present in serum, alpha 2-macroglobulin (alpha 2-M), alpha 1-proteinase inhibitor (alpha 1-PI) and alpha 1-antichymotrypsin (alpha 1-Achy). As chymotrypsin A remains to 90% active when bound to alpha 2-M, the chymotrypsin A/alpha 2-M complex was quantified by an enzymatic assay. The kinetic parameters of the interaction of chymotrypsin A with alpha 1-PI and alpha 1-Achy were determined. Using these data the partition of chymotrypsin A between the different inhibitors in serum was calculated. In acute pancreatitis, the chymotrypsin A plasma level follows the progression of the disease and in this case as well as in normal serum alpha 1-PI is the major antagonist of chymotrypsin A.