I Nutr Sci Vitaminol, 49, 340-345, 2003

Inhibition of Activities by Citrus Pectin

Takahiro TsuJITA 1,Maho SUMIyosH 1,Li-Kun HAN2, Tsutomu FUJIWARA3, Junji TSUJITA3and Hiromichi OKUDA2

1Division of Medical Bioscience, Department of Bioscience,Integrated Center for Sciences,Ehime University, Shigenobu, Onsen-gun, Ehime 791-0295, Japan 2Division of Food and Health Environment, Faculty of Environmental and SymbioticScience, Prefectural University of Kumamoto, Tsukide 3-chome, Kumamoto, Kumamoto 862-8502, Japan 3Research Laboratory, NihonshokkenLtd., Tomitashinko,Imabari, Ehime 799-1582, Japan (Received March 4, 2003)

Summary The oral administration of pectin to rats reduced and delayed the peak plasma triacylglycerol concentration. Pectin inhibited the of trioleoylglycerol emulsified with soybean phosphatidylcholine by pancreatic, carboxylester, and lingual in a con centration-dependent manner. However, the effective concentration of pectin for lingual lipase was 100 times lower than that for pancreatic lipase. Pectin did not inhibit the tribu tyrin and p-nitrophenylbutyrate-hydrolyzing activities by pancreatic and carboxylester lipase. When low molecular weight pectin was assayed, pectin at a molecular weight of 90,000 (MW 90) most strongly inhibited three lipase activities. When the effect of pH on pectin inhibition was analyzed using pancreatic lipase, strong inhibition was observed at an acidic pH (below pH 7.0). In the assay system, the pancreatic lipase protein levels in the supernatant and fat layer were estimated by Western blotting with an anti-pancreatic lipase antibody. Pectin reduced the amount of pancreatic lipase protein in the fat layer in a concen tration-dependent manner and concomitantly increased that in the supernatant. These results suggest that pectin may interact with emulsified substrates and inhibit the adsorp tion of lipase to the surface of substrate emulsion. Key Words pectin, pancreatic lipase, lingual lipase, triacylglycerol

It is well known that pectin lowers plasma and liver terol esters (7), and catalyzes the hydrolysis of water cholesterol in rats with diet-induced hyperlipidemia (1 soluble substrates such as methyl butyrate and p-nitro 3). Pectin intake resulted in lower plasma LDL choles phenylbutyrate. Typical substrates for these terol concentrations, whereas plasma VLDL and HDL are long-chain triacylglycerols, which are separated cholesterols were not affected by pectin intake in guinea from the aqueous medium by the surface phase. In con pigs (4). In humans, pectin also reduced total choles trast to their substrates, these lipases are water-soluble terol and LDLcholesterol concentrations, but HDL cho proteins. Thus, for catalysis, these enzymes must be lesterol was not significantly influenced by pectin (5). adsorbed into (or penetrate) the lipid surfaces, and Though numerous studies have shown that pectin has therefore, the quality of the substrate lipid surface is an been shown to lower cholesterol concentrations in sev important factor for lipase activity. Enzymes are some eral animal and human studies, few studies have ex times activated or denatured by surface adsorption (8). amined the effect of pectin on the triacylglycerol In the gut, triacylglycerol is emulsified with phospho metabolism. In this study, we investigated in detail, the lipids and bile salts. There are some reports that dietary effect of pectin on dietary triacylglycerol digestion and fiber affects lipase activity (9,10). The experiments pre absorption. sented in this report were undertaken to elucidate the In mammals, dietary triacylglycerol digestion is mode of inhibition of digestive lipases by pectin. mediated by three main enzymes: preduodenal (lingual MATERIALS AND METHODS or gastric), carboxylester, and pancreatic lipases. Under acidic conditions in the stomach, fat is hydrolyzed by Materials. The substrates and reagents used preduodenal lipase(s), which leads to the hydrolysis of were obtained as follows. Pectin (from citrus, molecular 10-30% of the dietary triacylglycerols to glycerols weight about 750,000) was obtained from Wako Pure (mainly diacylglycerol) and free fatty acids (6). Pancre Chemical Industries (Osaka, Japan) and low-molecular atic lipase hydrolyzes triacylglycerols to 2-monoacylg weight pectin (molecular weights of about 40,000, lycerol and free fatty acids. Carboxylester lipase has a 90,000, and 145,000) prepared from citrus pectin was broad substrate specificity, acting readily on triacylglyc obtained from Nihonshokken Co., Ltd. (Imabari, Japan). erols, diacylglycerols, monoacylglycerols, and choles Trioleoylglycerol, taurocholic acid and colipase were from Sigma (St. Louis, MO, USA). Phosphatidylcholine E-mail: [email protected] (PC, from soybean) was from Nihon Shoji (Tokyo,

340 Lipase Inhibition by Pectin 341

Japan). Lipid assay kits ( E-Test, NEFA C - Test, and Total cholesterol E-Test) were from Wako Pure Chemical Industries. Animals. Male Wistar King strain rats (5wk-old) were obtained from Charles River Japan (Yokohama, Japan), and housed for 1wk in a 12h alternate light/ dark cycle in a temperature and humidity-controlled room. The animals were given free access to food and water. After adaptation to the lighting conditions for 1 wk, the healthy animals were used in the present exper iments. The experiment protocol was approved by the Animal Studies Committee of Ehime University. Estimation of plasma triacylglycerol levels after the oral administration of lipid emulsion in rat. A suspension of 6 mL corn oil, 80 g cholic acid, and 2mg cholesterol oleate in 6mL water was sonicated for 5min. Male Wistar King rats, weighing 200g, were starved over night, divided into two groups and 1.0mL of corn oil suspension was administered to each rat via a stomach tube. One group received this suspension containing a 1.0mL pectin (50mg) solution and the control group received the suspension containing 1.0mL of water or corn starch (50mg). After pectin administration, blood samples were collected from the tail vein or artery into heparinized microcapillary tubes at regular intervals and centrifuged immediately at 6,500•~g for 5min. Plasma triacylglycerol and free concentra tions were determined using the Triglyceride E-Test and

NEFA C-Test, respectively. Enzyme source. Pancreatic lipase from rat pancreas

(3,200 Ulmg protein, at pH 6.8) and carboxylester Fig. 1. Effects of pectin on rat plasma triacylglycerol (A) lipase from porcine pancreas (159U/mg protein, at pH and free fatty acid (B) levels after the oral administra

6.8) was purified as described previously (11, 12). Crys tion of lipid emulsion. Lipid emulsion alone (•›) and talline lipase from Pseudomonas florescence was from lipid emulsion containing pectin (50mg) (•œ) were Amano Pharmaceutical Co. (Nagoya, Japan), and was orally administered. The results are expressed as mean•}SE of eight experiments. *p<0.05, significantly purified further as described previously (4,200U/mg different from lipid emulsion only group. protein, at pH 6.8) (13). A lingual lipase fraction was prepared from rat tongues. The entire lingual serous glandular region was homogenized in cold 25mM Lingual lipase activity was assayed in 0.1M citrate potassium phosphate buffer, pH 6.3, containing 0.9% potassium diphosphate buffer, pH 5.4, using PC as an NaCI. The homogenate was centrifuged at 100,000•~g emulsifier. The rate of tributyrin hydrolysis was mea for 60min and the supernatant, which was used as the sured using a pH-stat by the titration of liberated enzyme solution, was stored at -80•Ž (0.28U/mg pro butyric acid with 0.02N NaOH, as described previously tein, at pH 5.0). One unit (U) is defined to produce 1.0ƒÊ (15). mol of oleic acid from trioleoylglycerol per minute at 37•Ž. Pancreatic lipase distribution assay. An anti-pancre atic lipase antiserum was raised in rabbits using puri

Enzyme assay. Lipase activity was determined by fied rat pancreatic lipase as described previously (11). measuring the rate of release of oleic acid from trio Pancreatic lipase was incubated with trioleoylglycerol leoylglycerol. Briefly, a suspension of trioleoylglycerol PC emulsion containing various amount of pectin.

(80mg), PC (10mg) and taurocholic acid (5mg) in 9 After 10-min incubation, the supernatant and lipid mL of 0.1M N-tris(hydroxymethyl)-2-aminoethane layer were separated by centrifugation at 6, 500•~g for

sulfonic acid (TES) buffer (pH 7.0) containing 0.1M 10min and then suspended in Laemmli sample buffer NaCI was sonicated for 5min. The assay system was containing .1 and 20% (w/v) SDS, respectively. An ali comprised of the following components in a total vol quot of each suspension (10ƒÊL) was subjected to SDS ume of 200ƒÊL: 25ƒÊ enzyme solution, 50ƒÊL of pectin PAGE. For Western blotting, the proteins were trans solution, 0.5ƒÊmol trioleoylglycerol, 0.053ƒÊmol tauro ferred to a PVDF membrane (Bio-Rad Laboratories, CA,

cholic acid, 0.07ƒÊmol PC, 20ƒÊmol TES, and 20ƒÊmol USA) blocked with 5% (wv) skim milk and incubated NaCI. Colipase (0.25ƒÊg) were added to the assay mix with the antibody (16). Immunoreactivity was visual ture for pancreatic lipase. The amount of released oleic ized with alkaline -conjugated goat anti acid was estimated as described previously (14). rabbit IgG and Attophos (ICN Pharmaceuticals Inc., 342 TSUJITA T et al.

Fig. 2. Effect of pectin on pancreatic (A), carboxylester (B), and lingual (C) lipase activity. Triacylglycerol emulsified with

PC was used as substrates. (•›) pectin; (•œ) corn starch; (• ) polydextrose. The activity is expressed as the percentage of ac

tivity in the absence of polysaccharide. The results are expressed as means±SE of four experiments.

OH, USA), and the enhanced chemifluorescence inten sity was determined using a FlorImager, Florescence Imaging Analyzer (Amersham Pharmacia Biotech UK Ltd., Bucks, UK). Analysis of data. Statistical analysis was performed by the Fisher's Protected LSD test to determine the sig nificance of differences using Super ANOVAsoftware.

RESULTS Plasma triacylglycerollevels after oral administration of lipid emulsion in rat Figure 1 shows the change of the plasma triacylglyc erol concentrations over time when a corn oil suspen sion with or without pectin was administered orally to rats. Two hours after pectin administration, the plasma triacylglycerol concentrations decreased significantly as compared to the controls, and then increased after 6h. Therefore, the peak plasma triacylglycerol concentra Fig. 3. Effect of various molecular weight pectins on tion was reduced and delayed by pectin administration. pancreatic lipase activity. MW 750, MW-145, MW-90 However, the plasma triacylglycerol level was not affect and MW 40 have molecular weights of about 750,000, ed by the oral administration of cornstarch (data not 145,000, 90,000, and 40,000 pectins, respectively. Pectin (5mg/mL) was added in an incubation mixture. shown). Plasma free-fatty acid level was not affected by The activity is expressed as the percentage of activity in the oral administration of pectin (Fig. 1B). the absence of pectin. The results are expressed as

Lipase inhibition by pectin mean•}SE of four experiments. *p<0.05, significantly

Pectin inhibited pancreatic lipase activity dose different from control (absence of pectin). dependently at concentrations of 1-5mg/mL in the assay system using trioleoylglycerol emulsified with PC: pectin on pancreatic lipase activity. Commercial citrus at 5mg/mL, it inhibited trioleoylglycerol hydrolysis by pectin (molecular weight of about 750,000) and three approximately 70% (Fig. 2A). Cornstarch slightly inhib low molecular weight pectins (molecular weights of ited pancreatic lipase activity: at 5mg/mL, it inhibited about 40,000, 90,000 and 145 000) inhibited pancre trioleoylglycerol hydrolysis by only 12%. Polydextrose atic lipase activity. The pectin with a molecular weight did not inhibit lipase activity. Pectin inhibition of trio of 90,000 (MW 90) was the strongest in inhibiting leoylglycerol hydrolysis by carboxylester lipase and lin lipase activity: 5mg/mL pectin inhibited hydrolysis by gual lipase was observed (Fig. 2B, C). Pectin (up to 10 90%. Similarly, MW90 pectin was the strongest for mg/mL) did not affect the tributyrin-hydrolytic activi inhibiting carboxylester and lingual lipase activities ties of pancreatic and carboxylester lipases (data not (data not shown). MW 90 pectin at 5 g/mL inhibited shown), nor did it affect the p-nitrophenyl butyrate Pseudomonas fluorescence, Candida cylindicea, Mucor hydrolytic activity of the latter (data not shown). lipolytics, and Risopus delmar lipase activities by 45, 86, Low molecular weightpectin 48, and 52%, respectively. However, it did not inhibit Figure 3 shows the effect of low molecular weight Aspergillus sp. lipase activity. As shown in Fig. 4, MW Lipase Inhibition by Pectin 343

Fig. 4. Effect of MW 90 pectin on pancreatic (A), carboxylester (B), and lingual (C) lipase activities. The activity is ex pressed as the percentage in the absence of pectin.

Fig. 5. Effect of pH on the rate of hydrolysis of trio

leoylglycerol emulsified with PC by pancreatic lipase.

The enzyme activities were determined in the presence

(•›) and absence (•œ) of MW 90 pectin (5mg/mL). Tris HCl (pH 8.0-9.0), potassium phosphate (pH 6.5-7.5), Fig. 6. Pancreatic lipase distribution. Pancreatic lipase and sodium acetate (pH 5.0-6.0) were used for the as was incubated with trioleoylglycerol-PC emulsion con say The innic strength was 0.1. taining various amounts of MW-90 pectin. After 10 - min incubation, the supernatant and lipid layer were 90 pectin inhibited pancreatic and carboxylester lipase separated by centrifugation as described in Materials activities at concentrations of 1-5mg/mL. Pectin and Methods, and the proteins were separated by SDS strongly inhibited lingual lipase activity at 10-50ƒÊg/ PAGEusing gel containing 8% acrylamide. A; A repre mL (Fig. 4C). As lingual lipase activity was determined sentative immunoblot showing pancreatic lipase pro at pH 5.4, the effect of pH on pancreatic lipase activity tein with various amounts of MW 90 pectin. B; was determined (Fig. 5). The optimum pH for the tri Pancreatic lipase immunoreactive protein in the super oleoylglycerol-hydrolytic activity of pancreatic lipase natant (shaded column) and fat layer (open column) was between 8.0 and 9.0. MW 90 pectin (5mg/mL) tabulated as a percentage fraction of the density. Each inhibited this activity at all pH values, but it showed column represents the mean±SE of four separate ex strongest inhibition at acidic pHs (below pH 7.0). periments. Pancreatic lipase distribution Pancreatic lipase was incubated with trioleoylglyc tion-dependent manner and concomitantly increased erol-PC emulsion containing various amounts of pectin. that in the supernatant (Fig. 6B). After 10-min incubation, the supernatant and lipid DISCUSSION layer were separated by centrifugation. The pancreatic lipase protein levels in the supernatant and fat layer It is well known that dietary fat is not directly ab were estimated by Western blotting with anti-pancre sorbed from the intestine unless it has been subjected to atic lipase antibody. Figure 6A shows representative the action of lipases (lingual, pancreatic, and carboxyl immunoblots of pancreatic lipase in the supernatant ester). The two main products formed by the hydrolysis and fat layer. MW-90 pectin reduced the amount of of lipases are fatty acid and 2-monoacylglycerol. These pancreatic lipase protein in the fat layer in a concentra lipolytic products are dispersed as micelles and carried 344 TSUJITA Tet al.

to the site of fat absorption. Lipid absorption takes place was effective in repressing the level of the blood choles in the apical part of the plasma membrane of epithelial terol content (25). Yamaguchi et al. reported the rela cells or enterocytes lining the gut. Pectin inhibited trio tionship between molecular weights of pectin and leoylglycerol hydrolysis by digestive lipases (Fig. 2). hypocholesterolemic effects (26). In this paper, we However, it did not inhibit the hydrolysis of monomeric determined the effect of the molecular weight of pectin substrates such as tributyrylglycerol or p-nitrophenyl on lipase activities. MW 90 pectin was strongest for butyrate by the lipases (data not shown). These results inhibiting lipase activities (Fig. 3). MW 90 pectin might suggest that pectin may have interacted with emulsified have the appropriate viscosity and solubility for lipase substrate and inhibited the adsorption of pancreatic li inhibition as compared with other molecular weight pase to the surface of substrate emulsion. In fact, pectin pectins. Therefore, MW-90 pectin will be useful as a reduced the amount of lipase protein in substrate emul food and drink additive to inhibit lipid adsorption. sion in a concentration-dependent manner (Fig. 6). REFERENCES For lipase catalysis, the surface characteristics of sub strate lipid emulsion are the most important factors. Li 1) Hexeberg S, Hexeberg E, Willumsen N, Berge PK. 1994. pases are sometimes activated or denaturated by A study on in heart and liver of choles adsorption (or penetration) onto the substrate surface terol and pectin-fed rats. Br J Nutr 71: 181-192. 2) Tinker LF, Davis PA, Schneeman BO. 1994. Prune fiber (17). Most lipase inhibitors reported thus far are ad or pectin compared with cellulose lowers plasma and sorbed onto the lipid surface, thus blocking the surface liver lipids in rats with diet-induced hyperlipidemia. J and affecting lipase activity. Several proteins, hydrocar Nutr 124: 31-40. bons, fatty alcohols, and fatty acids also adsorb onto the 3) Anderson JW,Jones AE, Riddell-Mason S.1994. Ten dif substrate interface and inhibit lipase activity (18, 19). ferent dietary fibers have significantly different effects The responssssiblemechanism for inhibition by pectin is on serum and liver lipids of cholesterol-fed rats. J Nutr likely similar to that of these inhibitors. However, lipase 124: 78-83. inhibition by pectin was affected by the pH: pectin was 4) Shen H, He L, Price RL, Fernandez ML. 1998. Dietary strongly inhibitie at acidic pHs, below pH 7.0 (Fig. 5). soluble fiber lowers plasma LDL cholesterol concentra Therefore, it strongly inhibited the lingual lipase (acidic tions by altering lipoprotein metabolism in female lipase): the effective concentration of pectin for lingual guinea pigs. J Nutr 128: 1434-1441. lipase was 100 times lower than that for pancreatic li 5) Brown L, Rosner B, Willett WW, Sacks FM. 1999. Cho lesterol-lowering effects of dietary fiber: a meta-analysis. pase (Fig. 4). Other inhibitors (basic proteins and Am J Clin Nutr 69: 30-42. chitin-chitosan) did not inhibit triacylglycerol hydroly 6) Hamosh M, Klaevman HL, Wolf R0, Scow R0. 1975. sis by lingual lipase (14). Therefore, pectin may strongly Pharyngeal lipase and digestion of dietary triglyceride in inhibit triacylglycerol digestion in the stomach. man. J Clin Invest 55: 908-913. When corn oil suspension with or without pectin was 7) Redd ER, Brockman HL. 1984. Pancreatic carboxylester administered orally to rats, total triacylglycerol absorp lipase. In: Lipases (Borgstrom B, Brockman HL, eds), p tion was not changed (Fig. 1). However, the peak 185-204. Elsevier, Amsterdam. plasma triacylglycerol concentration was reduced and 8) Rietsch J, Pattus F, Desnuelle P, Verger R. 1977. Further delayed by pectin administration: 2h after pectin studies of mode of action of lipolytic enzymes. J Biol administration, plasma triacylglycerol concentrations Chem 252: 4313-4318. decreased significantly as compared to the controls, and 9) Hansen WE. 1987. Effect of dietary fiber on pancreatic lipase activity in vitro. Pancreas 2: 195-198. increased after 6h. These results suggest that pectin 10) Hendrick JA, Tadokoro T, Emenhiser C, Nienaber U, Fen may protect diet-induced hypertriglyceridemia. Bae et nema OR. 1992. Various dietary fibers have different al, reported that postprandial hypertriglyceridemia effects on lipase-catalyzed hydrolysis of tributyrin in caused endothelial dysfunction and might pave the way vitro. J Nutr 122: 269-277. for the development of atherosclerosis (20, 21). Fur 11) Tsujita T, Miyazaki T, Tabi R, Okuda H. 1996. Coenzyme thermore, it is well known that hypertriglyceridemia is A-independent monoacylglycerol acyltransferase from a risk factor for acute pancreatitis (22, 23). rat intestinal mucosa, J Biol Chem 271: 2156-2161. Pectin generally has the properties of low solubility, 12) Tsujita T, Okuda H. 1990. Effect of bile salts on the high viscosity, and strong gelatinization. When pectin interfacial inactivation of pancreatic carboxylester of various molecular weights was assayed for lipase lipase. J Lpid Res 31: 831-838. inhibitory activity, MW 90 pectin was the strongest 13) Nakagawa A, Tsujita T, Okuda H. 1984. Effect of Briji 58 on the hydrolysis of methylbutyrate by lipase from inhibitor, and MW 40 and MW-750 pectin weakly Pseudomonasfluorescens. J Biochem96: 815-820. inhibited lipase activity (Fig. 3). As compared with MW 14) Tsujita T, Matsuura Y, Okuda H. 1996. Studies on the 90 pectin, MW-40 pectin has high solubility and a low inhibition of pancreatic and carboxylester lipase by viscosity. Contrary to this, MW 750 has low solubility protamine. J Lipid Res 37:1481-148 7. and high viscosity. Therefore, the solubility and viscos 15) Tsujita T, Nakagawa A, Shirai K, Saito Y, Okuda H. ity might be responsible for the observed inhibition of 1984. Methyl butyrate-hydrolyzing activity of hepatic lipases. Isaksson et al. assayed the effects of viscosity on triglyceride lipase from rat post heparin plasma. J Biol enzyme activities of duodenal juice and reported that Chem 25: 11215-11220. increasing viscosity of pectin reduced pancreatic lipase 16) Morimoto C, Kiyama A, Kameda K, Ninmiya H, Tsujita activity (24). Makady reported that high viscous pectin T, Okuda H. 2000. Mechanism of the stimulatory action Lipase Inhibition by Pectin 345

of okadaic acid on lipolysis in rat fat cells. J LipidRes 41: prandial hypertriglyceridemia on vascular responses in 199-204. patients with coronary artery disease: effects of ACE in 17) Rietsch J, Pattus F, Desnuelle P, Verger R. 1977. Further hibitors and fibrates. Atherosclerosis 158: 165-171. studies of mode of action of lipolysic enzymes. J Biol 22) Kimura W, Mossner J. 1996. Role of hypertriglyceri Chem 252: 4313-4318. demia in the pathogenesis of experimental acute pan 18) Gargouri Y, Julien R, Sugihara A, Verger R, Sarda L. creatitis in rats. Int J Pancreatol 20: 177-184. 1984. Inhibition on pancreatic and microbial lipase by 23) Tosles PP. 1990. Hyperlipidemic pancreatitis. Gastroen proteins. Biochim Biophys Acta 795: 326-331. terol Clin North Am 19: 783-791. 19) Ferreira GC, Patton JS. 1990. Inhibition of lipolysis by 24) Isaksson G, Lundquist I, Ihse I. 1982. Effect of dietary hydrocarbons and fatty alcohols. J Lipid Res 31: 889 fiber on pancreatic enzyme activity in vitro. Gastroenter -897. ology 82: 918-924. 20) Bae JH, Bassenge E, Kim KB, Kim YM, Kim KS, Lee HJ, 25) Mokady S. 19 73. Effect of dietary pectin and algin on Moon KC, Lee MS, Park KY,Schwemmer M. 2001. Post blood cholesterol level in growing rats fed a cholesterol prandial hypertriglyceridemia impairs endothelial func free diet. Nutri Metab 15: 290-294. tion by enhanced oxidant stress. AtheroscIerosis 155: 26) Yamaguchi F, Uchida S, Watabe S, Kojima H, Shimizu N, 517-523. Hatanaka C. 1995. Relationship between molecular 21) Bae JH, Bassenge E, Kim KB, Lee HJ, Park KR, Park CG, weights of pectin and hypocholesterolemic effects in Park KY,Lee MS, Schwemmer M. 2001. Impact of post rats. Biosci Biotechnol Biochem 59: 2130-2131.