Proteinase Inhibitors Induce Selective Stimulation of ...

Human and Clinical Nutrition

Proteinase Inhibitors Induce Selective Stimulation of Human Trypsin and Chymotrypsin Secretion1'2 JA7Y/YEE. RESELAND, HALVOR HOLM,*3 MORTEN B. JACOBSEH,? TRONO G. JENSSEN** AND LARS E. HANSSENr MATFORSK, Norwegian Food Research Institute, N-1430 As, and Biotechnology Centre of Oslo, University of Oslo, Norway; institute for Nutrition Research, University of Oslo, N-0316 Oslo, Norway; 'Medical Department A, Rikshospitalet, N-0027 Oslo, Norway; and **Medical Department, University Hospital of Troms0, N-9012 Troms0, Norway

1The present study received financial support from The Research Council of Norway, Norwegian Dairies (J. E. Reseland), FREÕAChoc olate Factory Medical Foundation (H. Holm) and The Norwegian Cancer Society |L. E. Hanssen]. 2The costs of publication of this article were defrayed in part by

INDEXING KEY WORDS: •cholecystokinin release •feed-back regulation •pancreatic secretion •proteinases •soybean inhibitors •humans

the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 USC section 1734 solely to indicate this fact. 3To whom correspondence should be addressed. 4Abbreviations used: BBI, Bowman-Birk inhibitor; CCK, chole

Inhibitors of digestive enzymes are widely distrib uted in food. Soybeans are an increasingly important

cystokinin; CPA, carboxypeptidase A; CPB, carboxypeptidase B; KTI, Kunitz trypsin inhibitor; PSTI, pancreatic secretory trypsin inhibitor.

0022-3166/96 $3.00 ©1996 American Institute of Nutrition. Manuscript received 15 June 1994. Initial review completed 26 July 1994. Revision accepted 6 November 1995.

634

Downloaded from jn.nutrition.org by guest on July 14, 2011

nutritional source of protein containing protease or trypsin inhibitors. The two main families of soybean trypsin inhibitor are the Kunitz trypsin inhibitor (KTI)4 (Kunitz 1947a, 1947b) and the Bowman-Birk inhibitor ¡BBI)(Birk 1985, 1987). Kunitz trypsin inhibitor is a protein of 21 kDa, including two disulfide bridges; BBI has a molecular weight of 8 kDa with a high content of cysteine, forming seven disulfide bridges. Both inhib itors are known to inhibit human trypsin as well as chymotrypsin (Krogdahl and Holm 1979). A predominantly humoral regulation of pancreatic secretion via cholecystokinin (CCK) is well docu mented in rats (Fölsch 1990, Louie et al. 1986). Both inhibition of tryptic (EC 3.4.21.4) activity in the rat duodenum (Miyasaka and Green 1984) and diversion of pancreatic juice (Green and Lyman 1972) may increase plasma CCK, which in turn causes an increased parallel pancreatic secretion (Liener et al. 1988, Louie et al. 1986). Inhibition of human intraduodenal tryptic activ ity has been found to induce an enhanced pancreatic secretion (Holm et al. 1992, Layer et al. 1990). Though extensively studied, the precise mechanism regulating pancreatic secretion in humans remains unknown. Layer et al. (1990) found no increase in plasma CCK

ABSTRACT Among the variety of signals stimulating pancreatic secretion, cholecystokinin (CCK) and re lated hormones are assumed to be responsible for modulating proteinase output. In some species, intraduodenal tryptic activity has to be abolished to demon strate feedback-induced CCK release. The aim of this study was to investigate in vivo effects of modest inhi bition of ¡ntraduodenalproteolytic enzymes on the se cretion patterns of pancreatic enzymes and plasma CCK concentrations. Two inhibitors (Kunitz trypsin in hibitor and Bowman-Birk inhibitor) were applied. Inter mittent sampling of plasma and duodenal juice was performed during ¡ntraduodenalsaline and inhibitor in stillations in six healthy volunteers. Enzyme activities and concentrations were determined in the duodenal samples and expressed as percentages of basal val ues. Instillation of Kunitz trypsin inhibitor caused an increase in trypsin and the pancreatic secretory trypsin inhibitor (PSTI), without changes in plasma CCK. This result demonstrates, for the first time, that pancreatic exocrine secretion of trypsin and chymotrypsin is regu lated by different mechanisms. Bowman-Birk inhibitor additionally stimulated the secretion of chymotrypsin and carboxypeptidase A and B and increased plasma CCK. Elastase 1 and amylase secretions were not in creased by either instillations. Although the inhibitors have similar in vitro inhibition patterns, their in vivo effects are different. The nonparallel secretion of proteinases (trypsin, chymotrypsin and elastase 1) sup ports the view of a complex system involved in feed back regulation of human pancreatic exocrine secre tion, including signals other than CCK. J. Nutr. 126: 634-642, 1996.

INHIBITOR-STIMULATED

PROTEINASE SECRETION

during intraduodenal instillation of camostate (72% in hibition of tryptic activity). When both tryptic and chymotryptic (EC 3.4.21.1) activities were abolished by a mixture of KTI, BBI and camostate, a nonparallel pan creatic secretion of enzymes was found concomitant with a significant increase in plasma CCK (Reseland et al. 1995). We wanted to study the intraduodenal effect of individual and moderate inhibition on pancreatic exocrine secretion.

MATERIALS AND METHODS

Amersham, U.K.) as a nonabsorbable marker to enable calculation of dilution of the test solutions by endoge nous duodenal juice. Steady-state concentration of the marker was achieved within 30-40 min in an initial saline period lasting 60 min. The duodenal juice and blood samples drawn during the last 10 min of this saline period were designated as basal values for each individual. These samples were used as reference val ues (100%) for the subsequent samples. Inhibitor solu tions of approximately 0.1 mmol/L were used: 60 mg BBI or 180 mg KTI (Sigma Chemical, St. Louis, MO) in 100 mL of saline was instilled on different days over a period of 60 min. The inhibitor solutions were followed by a second saline instillation for another 30 min (100 mL/h). The samples of duodenal juice were divided into small tubes and kept at -20°C for short-term storage and at -70°C for long-term storage. Enzyme assays. As previously described (Holm et al. 1988), Afa-benzoyl-DL-arginine p-nitroanilide (BAPNA) and Na-benzoyl-L-tyrosine ethyl ester (BTEE) (Sigma Chemical) were used as substrates in the trypsin and chymotrypsin assays, respectively. Total proteolytic ac tivity was assayed using casein (Merck, Darmstadt, Ger many) (Krogdahl and Holm 1979) and elastase 1 (EC 3.4.21.11 ) activity using succinyl-(L-alainine)3-p-nitroan-

635

ilide (Peptide Institute, Osaka, Japan) according to the method of Bieth et al. (1974). Carboxypeptidase A (CPA) (EC 3.4.17.1) activity was assayed according to the method of Hartsuck and Lipscomb (1971) using Afa-hippuryl-L-phenylalanine (Sigma Chemical), and carboxypeptidase B (CPB) (EC 3.4.17.2) activity was assayed according to the method of Folk (1971) using Mr-hippuryl-L-arginine (Sigma Chemical). Amylase (EC 3.2.1.1) activity was measured with Phadebas Amylase Testkit (Pharmacia, Uppsala, Sweden). Inhibitory effect of Bowman-Bilk inhibitor and Kunitz trypsin inhibitor on pancreatic enzymes. Aliquots of the two inhibitors were added to basal duode nal juice in vitro, in a ratio similar to the average dilu tion of the test solutions by endogenous juice in vivo. Enzyme activities in these mixtures were compared with the initial activities in basal duodenal juice. Autolysis of enzymes. Aliquots of basal duodenal juice from all the participants were pooled, mixed, di vided into three samples and incubated in vitro at 37°C. The autodigestion was monitored for 60 min by taking aliquots every 10 min for measurements of tryptic and chymotryptic activities and immunoreactivity. Immunological analysis. Trypsin and chymotryp sin were quantified by rocket immunoelectrophoresis as described previously (Holm et al. 1992). The antisera to human trypsin and chymotrypsin (Calbiochem, La Jolla, CA) were used in final dilutions of 1:200 and 1:150, respectively. Purified human trypsin and human chymotrypsin (Calbiochem) were used as reference pro teins. Relative changes in a-amylase concentrations were measured by scanning of Western immunoblots, as described by Reseland et al. (1995). Elastase 1 was quantified with the Pancreatic Elastase 1 (stool speci men) Testkit from ScheBo Tech (Wettenberg, Ger many). Prior to the analysis, the duodenal samples were diluted 1:10,000 in Tris-buffered saline. Pancreatic se cretory trypsin inhibitor (PSTI) was quantified using the Spectria TATI kit (Farmos Diagnostica, Turku, Fin land), a kit developed for measuring tumor-associated trypsin inhibitor (TATI) (Stenman et al. 1991). The analyses of elastase 1 and PSTI were performed ac cording to the manufacturers' instructions, but the in cubation step involving diluted duodenal juice was per formed at 4°Cinstead of at room temperature (Mason and Williams 1986) to avoid autodigestion of proteins. For the quantification of elastase 1, the incubation time was doubled. Cholecystokinin was measured in periph eral plasma in a sensitive radioimmunoassay according to the method of Burhol et al. (1982). Immunoreactivity is defined as being similar to concentration when an immunological technique involving reference or stan dard proteins is employed. Statistical analysis. Results are given as means and SEM of observations from six persons. However, the activity of elastase 1 and the concentrations of elastase 1, PSTI and amylase are given as results from pooled

Downloaded from jn.nutrition.org by guest on July 14, 2011

Healthy volunteers. Six healthy female volunteers, aged 20-31 y, participated in this study. All partici pants gave written informed consent before the study. The Regional Ethics Committee, Rikshospitalet (Oslo, Norway) approved the experimental protocol. Instillation of inhibitors. The procedure for intra duodenal instillation and sampling with a three-lumen tube was previously described (Holm et al. 1988). In brief, a Lagerlöf tube (Rusch, Waiblingen, Germany) was placed under fluoroscopic control to allow contin uous aspiration of gastric juice and intermittent sam pling of duodenal juice. Duodenal juice (3-6 mL) was sampled every 10 min. A Portex vinyl tube (800/000/ 180/800) (Portland Plastics, Kent, U.K.), attached to the Lagerlöftube, was placed with its end close to the pa pilla of Vater for continuous instillation of saline or isoosmotic test solutions (100 mL/h). Saline and inhibitor solutions contained [57Co]cyanocobalamin (370 kBq/L,

IN HUMANS

RESELAND ET AL.

636 TABLE 1

In vitro inhibition of enzyme activities in basal duodenal juice after addition of Kunitz trypsin inhibitor (KTI) or Bowman-Birk inhibitor (BBI) in a ratio similar to the average dilution of the instilled inhibitor solution by endogenous juice in vivo1 Inhibition by KTI

Enzyme

Inhibition by BBI

TrypsinChymotrypsinElastase 1Carboxypeptidase ACarboxypeptidase BAmylase967453202987448701

rapidly than immunoreactivity (P < 0.002) (Fig. 1). The losses of immunoreactive material the first 10 min of incubation were on the order of 7 and 5% for trypsin (Fig. la) and chymotrypsin (Fig. Ib), respectively. The enzyme activities decreased on the order of 20 and 35% during the same period. Instillation of Kunitz trypsin inhibitor. The instil lation of KTI (10-60 min) increased trypsin concentra tion to an average of 178% of the basal value, from 2.3 ±0.6 to 4.1 ±0.5 g/L (P = 0.03) (Fig. 2a). The chymotrypsin concentration did not change signifi cantly from the basal value of 3.8 ±0.3 g/L. The esti mated relative changes in output (mg/min) of trypsin and chymotrypsin were not significantly different from the measured changes in concentration (Table 2) during KTI instillation. Reductions of intraduodenal tryptic

1 Values are means of three replicates.

a function of instillation time was performed for CCK, PSTI and the enzymes (Minitab Release 7.2, Minitab, Pennsylvania State University, PA). Differences with P =s 0.05 were considered significant.

100

Trypsin V activity o concentration

80