David A. W. Grant

Hydrolysis of artificial substrates by enterokinase and trypsin and the development of a sensitive specific assay for enterokinase in serum

The activities of highly purified human enterokinase (enteropeptidase, EC 3.4.21.9) and bovine trypsin were tested against three synthetic substrates alpha-N-Benzoyl-L-arginine ethyl ester HCl, alpha-N-Benzoyl-DL-arginine-p-nitroanilide HCl and alpha-N-Benzoyl-DL-arginine-2-naphthylamide HCl. There was no detectable hydrolysis of these substrates by enterokinase whereas the kinetic parameters obtained for trypsin were in close agreement with those previously described by other workers. The values for Km and kcat were dependent on the Ca2+ concentration. Hydrolysis of glycine-tetra-L-aspartyl-L-lysyl-2-naphthylamide (Gly(Asp)4-Lys-Nap) by these protease was also studied. Enterokinase-catalysed hydrolysis obeyed simple steady-state kinetics and values for Km of 0.525 mM and 0.28 mM and for kcat of 21.5 s-1 and 28.3 s-1 were obtained in 0.1 mM and 10 mM Ca2+, respectively. Trypsin-catalysed hydrolysis was complex and the response to Ca2+ was sigmoidal partly due to the lability of trypsin at low Ca2+ concentrations. A sensitive specific assay for enterokinase was developed and applied to the measurement of the enzyme in serum; interference by nonspecific arylamidases was eliminated by the addition of Zn2+.

Induced mucosal penetration and transfer to portal blood of luminal horseradish peroxidase after exposure of mucosa of guinea pig small intestine to ethanol and lysolecithin

The effect of luminal 150 mmol saline, 0.05–0.2% (w/v) lysolecithin, and 5–20% (v/v) ethanol was studied on the mucosal morphology of the proximal small intestine in conscious guinea pigs as well as on the mucosal penetration and transfer to portal venous blood of luminal horseradish peroxidase (HRP). No ultrastructural evidence of mucosal damage was identified in any of the lysolecithin-perfused animals compared with saline controls. Ten and 20% ethanol (v/v) resulted in the appearance of fluid-filled spaces between enterocytes and in cytoplasmic lipid deposits and an increased number of autophagic vesicles within the cells themselves. Tight junctions remained intact. These changes after luminal 5% ethanol (v/v) were much less conspicuous. In the presence of saline, luminal HRP was largely confined to the brush border. Both lysolecithin and ethanol (5% v/v) rapidly induced mucosal penetration of HRP which was seen in cytoplasmic vesicles within enterocytes, between enterocytes, and in the lamina propria. Peak portal venous blood levels of HRP studied in multiple samples over 3 hr were one log unit greater than saline controls. Absorption of HRP was proportional to the luminal concentration of lysolecithin in the range tested. These studies show that mucosal penetration and absorption of functional exogenous macromolecules may be induced, in the absence of morphological evidence of mucosal damage, by luminal constituents which may perturb the structure of enterocyte membranes.

Cleavage of peptide hormones by α2-macroglobulin-trypsin complex and its relation to the pathogenesis and chemotherapy of acute pancreatitis

Access to the active site of alpha 2-macroglobulin bound proteinases by macromolecular substrates and inhibitors is likely to be influenced by steric favourability and flexibility, as well as by molecular size. Hydrolysis of pure porcine cholecystokinin-pancreozymin 39, a flexible single chain peptide, by alpha 2-macroglobulin-trypsin complex resulted in a rapid up to 6-fold increase in cholecystokinin bioactivity; alpha 2-macroglobulin-trypsin rapidly abolished the bioactivity of endogenous parathormone in human plasma. Inhibition of both reactions was completed by low concentrations of antipain and leupeptin; Trasylol (aprotinin), a single chain peptide with three disulphide bonds, was an ineffective inhibitor even in massive molar excess. These findings may provide an explanation for the disordered calcium homeostasis in severe acute pancreatitis and for the failure of Trasylol to reduce mortality; they suggest that sterically favourable low molecular weight inhibitors may provide effective specific chemotherapy for the disease.

The generation of lysolecithin by enterokinase in trypsinogen prophospholipase A2 lecithin mixtures, and its relevance to the pathogenesis of acute necrotising pancreatitis

The cascade enterokinase-trypsinogen-prophospholipase A2 lecithin, generating trypsin, phospholipase A2 and lysolecithin, respectively, was studied in vitro using a novel phospholipase A2 assay. The rate of enterokinase catalysed activation of trypsinogen was maximal at 4 mmol/1 glycodeoxycholic acid; higher concentrations of bile salt progressively inhibited enterokinase activity. Net phospholipase A2 activity in reaction mixtures was critically dependent on the trypsin/prophospholipase A2 molar ratio. Lecithin hydrolysis by phospholipase A2 was dependent on the bile salt/lecithin molar ratio and was optimal at 1.25 to 1. The addition of enterokinase to lecithin and bile salt mixtures, containing trypsinogen and prophospholipase A2 at presumed pathophysiological concentrations, resulted in the generation of concentrations of lysolecithin lytic for pancreatic acinar cells within 5 min. These findings would support the concept that the entry of bile containing active enterokinase into the pancreatic duct system in vivo may in some cases be involved in the initiation of necrotising acute pancreatitis in man.

Catalytically active enterokinase in human bile

Enterokinase activates trypsinogen very rapidly and is itself resistant to proteolysis and endogenous inhibitors in blood and pancreas. Using a novel one-stage specific catalytic assay capable of detecting enterokinase in the presence of trypsin inhibitors, we have positively identified catalytically active enterokinase in human bile in each of 14 patients studied. Since the presence of active enterokinase in human bile was not explicable by duodeno-biliary reflux, enterokinase must have followed the pathway from gut to blood to liver to bile, previously identified in greater detail experimentally. We suggest that entry into the pancreatic duct system of bile-borne active enterokinase from the common bile duct may trigger necrotising acute pancreatitis.

A differential effect between the acute and chronic administration of ethanol on the endocytotic rate konstant, ke, for internalisation of asialoglycoproteins by hepatocytes

The endocytotic rate constant, ke, originally described for the quantification of epidermal growth factor by fibroblasts (Wiley, H.S. and Cunningham, D.D. (1982) J. Biol. Chem. 257, 4222-4229) has been adapted to measure receptor-mediated endocytosis of asialoglycoproteins by hepatocytes. A ke value of 0.21 min-1 was obtained for the internalisation of beta-D-galactosyl bovine serum albumin by freshly isolated hepatocytes. The addition of ethanol to the incubation medium had a biphasic effect on ke. The value of ke was increased by up to 30% by low concentrations of ethanol, whereas higher concentrations progressively decreased ke and in 500 mM ethanol the ke value was 0.1 min-1. The amount of ligand bound to the cell surface was independent of the extracellular concentration of ethanol and the changes in ke were exclusively due to changes in the amount of internalised ligand. There was a progressive decrease in the value of ke in hepatocytes prepared from rats that were maintained on an ethanol-impregnated liquid diet for up to 20 days. The decrease was already apparent by day 2 when blood alcohol levels were only 50 mg%, indicating that the effect of chronic alcoholism on endocytosis are manifested at an early stage.

Intramucosal activation of pepsinogens in the pathogenesis of acute gastric erosions and their prevention by the potent semisynthetic amphipathic inhibitor pepstatinyl-glycyl-lysyl-lysine☆

This study was designed to test the proposal that a critical event in the formation of acute gastric erosions is the intramucosal activation of pepsinogens. Gastric erosions were induced in rats by controlled haemorrhagic shock. Free acid proteinase activity in homogenates of gastric mucosa taken to include erosions increased progressively throughout the period of hypotension. Free enzyme activity in homogenates of apparently normal mucosa between erosions remained substantially unchanged. Luminal pepstatin, the potent but hydrophobic specific acid proteinase inhibitor which does not penetrate gastric mucosa, did not prevent the development of erosions. The equally potent but amphipathic water-soluble analogue pepstatinyl-gly-lys-lys, which does penetrate gastric mucosa, substantially reduced mucosal ulceration in this system. These findings suggest that focal intramucosal pepsinogen activation at the site of acute experimental erosions is causally related to their development, and suggest the inclusion of potent water soluble acid proteinase inhibitors of appropriate molecular structure in their clinical prevention and treatment.

Displacement of endogenous enterokinase into portal venous blood and bile following luminal perfusion of proximal small intestine in guinea pigs

The displacement of endogenous enterokinase into portal venous blood or bile was studied in conscious guinea pigs both with the small intestine undisturbed and during gentle, intermittent luminal perfusion of a 25-cm segment of duodenum and proximal jejunum. Perfusates tested included water, 150 mM saline, 5% (v/v) ethanol, 0.2% (w/v) lysolecithin, and mixtures of ethanol and lysolecithin. Enterokinase activity was absent from portal venous blood of control guinea pigs with the intestine undisturbed but perfusion with luminal saline or water was consistently associated with substantial levels of active enterokinase in portal venous blood. Similar concentrations of enterokinase in portal blood were also detected in response to luminal ethanol and lysolecithin. The capacity of the normal liver rapidly to clear the enzyme from portal blood was demonstrated. Of the estimated total endogenous enterokinase displaced, 0.2–0.4% was recovered in catalytically active form from the pooled bile of luminally perfused but not control animals. The readiness with which enterokinase was displaced into the circulation in the absence of mucosal damage raises the unexpected possibility that the event may be physiological. Induced penetration of the mucosa and absorption of luminal components is clearly different from the release into portal venous blood of endogenous mucosal macromolecules.

Distribution of pepstatin and statine following oral and intravenous administration in rats. Tissue localisation by whole body autoradiography

The distribution in rats of two radiolabelled derivatives of the potent pepsin inhibitor, pepstatin A, has been monitored by whole-body autoradiography. There was a very poor absorption of pepstatinyl-[14C]glycine and N-[3H]acetylstatine across the gastric and intestinal mucosae after oral administration. Both inhibitors were rapidly cleared from the blood by the liver and kidneys after intravenous administration. Neither inhibitor was localised in the gastric mucosa at a concentration that could be expected to be effective in inhibiting intracellular pepsin activity.

A sensitive fluorometric assay for the simultaneous estimation of pepsin and pepsinogen in gastric mucosa

An assay for pepsin has been developed based on the fluorometric measurement of trichloroacetic acid-soluble peptides released from casein at pH 5.3. The increase in relative fluorescence was most sensitive in the range 10-50 micrograms pepsin/l and casein hydrolysis was not affected by the addition of up to a 1000-fold molar excess of pepsinogen. This assay has been used to measure the free and total acid proteinase content of biopsies (less than 5 mg) from different areas of the gastric mucosa of rat and man. Interference by the major lysosomal acid hydrolase, cathepsin D, could be eliminated by the differential stability of pepsin and cathepsin D at acid and neutral pH. The free acid proteinase activity of biopsies from the corpus were almost identical in these species whereas the total acid proteinase activity was approximately 5-fold greater in man.