Available online at www.sciencedirect.com

www.elsevier.com/locate/bcdf

B i o a c t i v e C a r b o h y d r a t e s a n d D i e t a r y F i b r e 4 ( 2 0 1 4 ) 2 7 – 3 8

http://dx.doi.org/102212-6198/& 2014 El

nCorresponding autE-mail address: c

Inhibition of digestive enzyme activities by pecticpolysaccharides in model solutions

Mauricio Espinal-Ruiz, Fabian Parada-Alfonso,Luz-Patricia Restrepo-Sanchez, Carlos-Eduardo Narvaez-Cuencan

Departamento de Química, Facultad de Ciencias, Universidad Nacional de Colombia, AA 14490 Bogotá DC, Colombia

a r t i c l e i n f o

Article history:

Received 25 March 2014

Received in revised form

19 June 2014

Accepted 19 June 2014

Keywords:

Pectic polysaccharide

Methylation degree

Digestive enzymes

Hydrophobic interactions

Non-competitive inhibition

.1016/j.bcdf.2014.06.003sevier Ltd. All rights rese

hor. Tel.: þ57 1 3165000x1enarvaezc@unal.edu.co (

a b s t r a c t

The presence of dietary fiber (e.g., pectic polysaccharides, PPs) in the gastrointestinal tract

may decrease the caloric intake and reduce the risk of developing cardiovascular diseases.

These phenomena are governed by several mechanisms, such as the regulation of the rate

of nutrient absorption and the alteration of the normal activity of the gastrointestinal tract

enzymes. In this study, we evaluated the effect of PPs with five methylation degrees (MD)

on the activities of lipase, α-amylase, alkaline phosphatase, and protease. The MD of the

PPs ranged (in mol/mol) from 87.4% (high-methylated PP, HMPP) to 7.1% (low-methylated

PP, LMPP). The enzymatic activities were evaluated in model solutions after incubation

with PPs. The Michaelis–Menten constant remained unmodified whereas the apparent

maximum velocity (Vmaxapp) decreased with increasing PP concentrations. The Vmaxapp

represented 13.3%, 38.6%, 41.9%, and 44.4% of the Vmax (without PPs) for lipase, α-amylase,

alkaline phosphatase, and protease, respectively, when they were inhibited with 100 μgmL�1 HMPP. Kinetic analyses showed that all of the tested PPs behaved as non-competitive

inhibitors of digestive enzymes. Increasing both the concentration and MD of the PPs

reduced the enzymatic activities by decreasing the non-competitive inhibition constant

(Ki). In plotting Ki versus MD, a straight line was obtained, with slopes of 1.943, 1.558, 1.344,

and 1.165 μg mL�1%�1 for lipase, α-amylase, alkaline phosphatase, and protease, respec-

tively. Among them, lipase was most likely to be inhibited by the PPs. Our results suggested

that PPs might be able to suppress digestion by inhibiting digestive enzymes.

& 2014 Elsevier Ltd. All rights reserved.

rved.

4458; fax: þ57 1 3165220.C.-E. Narváez-Cuenca).

1. Introduction

The main function of the gastrointestinal tract is the absorp-tion of nutrients derived from food digestion. This functionis controlled by a series of digestive processes that occurin different sections of the gastrointestinal tract. Thesedigestive processes are controlled by the secretion of diges-tive enzymes and their associated cofactors and by thestability of the pH and temperature conditions of the gastro-intestinal tract (Dawson, 1993). Lipases, α-amylases, alkaline

phosphatases, and proteases are the main gastric and pan-creatic enzymes present in the gastrointestinal tract(Rothman, 1977). These enzymes are responsible for thehydrolysis of the triglycerides, carbohydrates, and proteinsthat are consumed in diet, which are carriers of a great caloriccontent. It has been postulated that the presence of any typeof dietary fiber in the gastrointestinal tract may result in thedecrease of the total caloric intake (Amarowicz, Kmita-Glazewska, & Kostyra, 1990). This phenomenon is governedby several mechanisms, such as the regulation of the nutrient

B i o a c t i v e C a r b o h y d r a t e s a n d D i e t a r y F i b r e 4 ( 2 0 1 4 ) 2 7 – 3 828

absorption rate, perturbations of the intestinal physiologicalconditions, the encapsulation of minerals and vitaminsneeded for metabolic processes, and alterations of the normalactivities of the digestive enzymes (Kumar & Chauhan, 2010).

Recent epidemiological studies have shown that the con-sumption of dietary fiber is associated with the reduction ofthe risk of developing chronic cardiovascular diseases. Con-sequently, it has been recommended to increase the intake ofproducts of plant origin that have high dietary fiber levels,particularly soluble dietary fiber, together with other phyto-chemical constituents (Kris-Etherton et al., 2002). Pecticpolysaccharides (PPs) are a type of soluble dietary fiber thatcannot be digested in the gastrointestinal tract due to theirresistance to the hydrolytic action of digestive enzymes(Holloway, Tasman-Jones, & Maher, 1983). PPs also promotebacterial fermentation in the large intestine, improving theproliferation of intestinal microbiota that is beneficial forhuman health (Louis, Scott, Duncan, & Flint, 2007).

In the gastrointestinal tract, PPs form a complex three-dimensional matrix with fibrous and amorphous character-istics. Physicochemical properties, such as the methylationdegree (MD), acetylation degree, molecular weight distribu-tion, distribution of non-methylated galacturonic acid resi-dues, methylation and acetylation distribution patterns, andgel-forming capacity (Brownlee, 2011), as well as the structureof the three-dimensional matrix, play important roles in thehomeostatic and therapeutic functionality of PPs in humannutrition (Kay, 1982). Enhancement of the gastrointestinalviscosity, inhibition of digestion and the absorption of nutri-ents, control of gastrointestinal motility and immunity,regulation of the activity of the colonic microbiota, andregulation of a systemic stimulus associated with the feelingof satiety are among the therapeutic properties and physio-logic effects that are beneficial for human health that havebeen attributed to PPs (Brownlee, 2011).

Regulation of the gastrointestinal tract enzymatic activityby dietary fiber from different plant materials was previouslyreported (Dunaif & Schneeman, 1981; Isaksson, Lundquist, &Ihse, 1982a, 1982b; Ikeda & Kusano, 1983; Tsujita et al., 2007).The dietary fiber materials contribute to inhibiting the activ-ity of gastrointestinal tract enzymes. According to the aforecited work, the occurrence of physical interactions (such asionic interactions, hydrogen bonding, dispersive forces, andhydrophobic interactions) might play an important role in thecapacity of the dietary fiber to inhibit enzymatic activities.

However, the experiments conducted in those studies, didnot identify a kinetic mechanism underlying the inhibition ofsuch enzymes. Inhibition of the activity of the enzymes byingested PPs might play an important role in reducing thequantity of free fatty acids, monosaccharides, and aminoacids that can be absorbed at the gastrointestinal tractlevel (Ikeda & Kusano, 1983). Identifying the mechanism bywhich PPs act as inhibitors of digestive enzymes in modelsolutions might be useful in understanding the physiologicalphenomena involved in the in vivo regulation that PPs exerton nutrient absorption at the gastrointestinal tract level(Brownlee, 2011).

The aim of this study was, therefore, to evaluate in modelsolutions the effects of both the concentration and MD of PPson the activities of lipase, α-amylase, alkaline phosphatase,

and protease (chymotrypsin). This study also aimed todetermine a kinetic mechanism by which PPs inhibit theactivity of enzymes, as well as to evaluate theoretically,through molecular docking calculations, the influence ofstructural parameters on the enzyme–PP surface interaction.

2. Materials and methods

2.1. Chemicals

The enzymes porcine pancreatic α-amylase, from Sus scrofa(16 U mg�1 type VI-B, E.C. 3.2.1.1) porcine pancreatic lipase,from Sus scrofa (100 U mg�1 type II, E.C. 3.1.1.3) bovinepancreatic protease (chymotrypsin), from Bos taurus (5 Umg�1 type I, E.C. 3.4.21.1), and bovine intestinal mucosaalkaline phosphatase, from Bos taurus (10 U mg�1 type I, E.C.3.1.3.1); the artificial substrates 2-chloro-p-nitrophenyl-α-D-maltotrioside (G3CNP), p-nitrophenyl palmitate (pNPPA), p-nitrophenyl acetate (pNPA), and p-nitrophenyl phosphate(pNPP); the reaction products p-nitrophenol (pNP) and 2-chloro-p-nitrophenol (CNP); and the protein determinationreagent brilliant blue G-250 (Coomassie Blue); as well asbovine serum albumin were purchased from Sigma-Aldrich(St. Louis, MO, USA). A high-methylated citrus pectic poly-saccharide (HMPP) was purchased from CIMPA (Bogotá,Colombia). Ox bile extract with cholic acid content higherthan 55% (w/w) was purchased from MP Biomedicals (Solon,OH, USA). Other chemicals were purchased from Merck(Darmstadt, Germany).

2.2. Pectic materials

2.2.1. Preparation of pectic polysaccharidesAlkaline de-esterification of the HMPP was performed asdescribed by Dongowski (1997) to obtain PPs with differentMD levels. One gram of HMPP was mixed with 50 mL of 0.25 MNaOH (pH of the mixture Z10.0) and stirred for 0, 9, 15, 30, or45 min at 25 1C. The mixture was neutralized to pH 7.0 with3 M HCl and then 150 mL of 80% (v/v) ethanol were added toinduce PP precipitation. The partially de-esterified pecticmaterials were filtered and washed with 100 mL of 80% (v/v)ethanol. The PPs were dried at 70 1C for 5 h and then the MD,total uronic acid content, acetylation degree, and molecularweight distribution were evaluated.

2.2.2. Characterization of the pectic polysaccharides2.2.2.1. Determination of total uronic acid content. The totaluronic acid content was determined according to van denHoogen et al. (1998). An aliquot of 400 μL of each PP solution(100 μg mL�1) was mixed with 2 mL of concentrated sulfuricacid (98%, w/w) containing 120 mM sodium tetraborate andincubated for 60 min at 80 1C. After cooling to room tempera-ture, the background absorbance of the samples was mea-sured at 540 nm. Then, 400 μL of m-hydroxydiphenyl reagent(100 μL of 100 mgmL�1 m-hydroxydiphenyl in dimethyl sulf-oxide, mixed with 4.9 mL of 80% (v/v) sulfuric acid) was addedand mixed with the samples. After 15 min, the absorbanceof the pink-colored samples was measured at 540 nm.A calibration line was obtained using galacturonic acid at

B i o a c t i v e C a r b o h y d r a t e s a n d D i e t a r y F i b r e 4 ( 2 0 1 4 ) 2 7 – 3 8 29

concentrations ranging from 0.1 to 1.0 μg mL�1 (7 data points,r2¼0.999). Total uronic acid content was expressed as themoles of uronic acid residues per 100 g of PP.

2.2.2.2. Determination of methylation and acetylation degrees.The MD and acetylation degree were determined according toVoragen, Schols, and Pilnik (1986). A model LC-20AT liquidchromatograph (Shimadzu Corporation, Kyoto, Japan)equipped with an Aminex HPX-87H column (300 mm�7.8mm�9 μm; Bio-Rad Labs, Richmond, CA, USA) was used. Thecolumn was operated at 18 1C, at a flow rate of 0.3 mL min�1,with 5 mM sulfuric acid as the eluent. The components elutedfrom the column were detected using a RID-10A refractiveindex detector (Shimadzu Corporation) at 40 1C. Each PP(30 mg) was suspended in 1 mL of 0.4 M NaOH and stirred at18 1C for 2 h. The suspension was then centrifuged (4000g;4 1C) for 30 min and 20 μL of the clear supernatant wasinjected into the column. The amounts of methanol andacetic acid were determined using the external standardmethod. Calibration lines were obtained with methanol andacetic acid at concentrations ranging from 5 to 100 μg mL�1 (7data points, r2¼0.999 for methanol and r2¼0.997 for aceticacid). The MD and acetylation degree were expressed asmoles of methyl and acetyl esters, respectively, per 100 molof uronic acid.

2.2.2.3. Determination of molecular weight distribution.The molecular weight distribution was determined usinghigh performance size exclusion chromatography (HPSEC)according to the method of Houben, Jolie, Fraeye, Van Loey,and Hendrickx (2011), using a LC-20AT liquid chromatographmodel (Shimadzu Corporation), equipped with a mixed-bedcolumn of TSK-gel (GMPW, 300 mm�7.5 mm, pore size 100–1000 Å, particle size 17 μm; Tosoh Biosciences, Stuttgart,Germany). A 20 μL injection loop was used. Elution wasperformed using 50 mM NaNO3 at a flow rate of 0.7 mL min�1

for 25 min at 35 1C. A RID-10A refractive index detector(Shimadzu Corporation) at 40 1C was used to monitor theeluents. Dextran standards with molecular weights rangingfrom 1.1 to 400 kDa (Sigma-Aldrich, St. Louis, MO, USA) wereused to estimate the molecular weight distribution of the PPs.A straight calibration curve was obtained when plotting thelog of the molecular weight versus the elution time.

2.3. Kinetics of the inhibition of digestive enzyme activitiesby pectic polysaccharides in model solutions

2.3.1. Experimental designThe kinetic inhibition profile for each enzyme by each PP wasobtained according to the methodology proposed by Kakkar,Boxenbaum, and Mayersohn (1999). A randomized compositefactorial 62�5 design was used for lipase, α-amylase, alkalinephosphatase, and protease (chymotrypsin). The experimentalfactors were derived using six substrate concentration points(ranging from 2 to 10 mM pNPPA for lipase, from 0.2 to 1.0 mMG3CNP for α-amylase, from 0.2 to 1.0 mM pNPP for alkalinephosphatase, and from 0.4 to 2.0 mM pNPA for protease), sixconcentration points for each PP (ranging from 20 to 100 μgmL�1), and five types of each PP, with MDs ranging (in mol/mol) from 7.1% (low-methylated PP, LMPP) to 87.4% (high-

methylated PP, HMPP). The inhibitory mechanism of each PPtoward the enzyme activities was evaluated by kinetic ana-lysis using the Lineweaver–Burk plot.

2.3.2. Determination of the inhibition of digestive enzymeactivities by the PPs2.3.2.1. Determination of the activity of lipase. The enzymaticactivity of lipase was determined according to Tsujita et al.(2007). A reaction mixture consisting of 500 μL of pNPPA atconcentrations ranging from 6 to 30 mM, 500 μL of each PPsolution at concentrations ranging from 60 to 300 μg mL�1,and 500 μL of a working solution of lipase at 3000 UmL�1

(one unit is the amount of enzyme required to convert 1 μmolof pNPPA) was prepared. All of the solutions were preparedin 50 mM Tris–HCl buffer pH 7.0 containing 150 mM NaCl,1 mM CaCl2, 100 μg mL�1 Tween-20, and 3.5 mg mL�1 bileacid extract (equivalent to 5.0 mM cholic acid in the totalmixture). The mixture was incubated at 37 1C and the absor-bance of the pNP produced was measured at 415 nm every 5 sthroughout 120 s.

2.3.2.2. Determination of the activity of α-amylase. The enzy-matic activity of α-amylase was determined according toMorishita, et al. (2000). A reaction mixture consisting of500 μL of G3CNP at concentrations ranging from 0.6 to3.0 mM; 400 μL of 13.1 mg mL�1 bile acid extract (equivalentto 18.7 mM cholic acid in the total mixture), 500 μL of each PPsolution at concentrations ranging from 60 to 300 μg mL�1,and 100 μL of a working solution of α-amylase of 150 UmL�1

(one unit is the amount of enzyme required to convert 1 μmolof G3CNP) was prepared. All of the solutions were prepared in20 mM phosphate buffer pH 7.0 containing 10 mM NaCl and15% (v/v) glycerol. The mixture was incubated at 37 1C and theabsorbance of the CNP produced was measured at 405 nmevery 5 s throughout 60 s.

2.3.2.3. Determination of the activity of alkaline phosphatase.The enzymatic activity of alkaline phosphatase was deter-mined according to Chaudhuri, Chatterjee, Venu-Babu,Ramasamy, and Thilagaraj (2013). A reaction mixture con-sisting of 500 μL of pNPP at concentrations ranging from 0.6 to3.0 mM, 400 μL of 13.1 mg mL�1 bile acid extract (equivalentto 18.7 mM cholic acid in the total mixture), 500 μL of each PPsolution at concentrations ranging from 60 to 300 μg mL�1,and 100 μL of a working solution of alkaline phosphatase of12 UmL�1 (one unit is the amount of enzyme required toconvert 1 μmol of pNPP) was prepared. All of the solutionswere prepared in 20 mM Tris–HCl buffer pH 7.0 containing15 mM NaCl. The mixture was incubated at 37 1C and theabsorbance of the pNP produced was measured at 415 nmevery 5 s throughout 60 s.

2.3.2.4. Determination of the activity of protease. The enzy-matic activity of protease (chymotrypsin) was determinedaccording to Verma and Ghosh (2013). A reaction mixtureconsisting of 500 μL of pNPA at concentrations ranging from1.2 to 6.0 mM, 400 μL of 13.1 mgmL�1 bile acid extract(equivalent to 18.7 mM cholic acid in the total mixture),500 μL of each PP solution at concentrations ranging from 50to 300 μg mL�1, and 100 μL of a working solution of protease

B i o a c t i v e C a r b o h y d r a t e s a n d D i e t a r y F i b r e 4 ( 2 0 1 4 ) 2 7 – 3 830

of 45 UmL�1 (one unit is the amount of enzyme required toconvert 1 μmol of pNPA) was prepared. All of the solutionswere prepared in 50 mM phosphate buffer pH 7.0 containing20 mM NaCl. The mixture was incubated at 37 1C and theabsorbance of the pNP produced was measured at 415 nmevery 5 s throughout 60 s.

2.3.2.5. Determination of the specific enzymatic activities. Todetermine the enzymatic activities, the slope of the straightlines obtained when plotting the absorbance versus time dataobtained from the enzymatic activity experiments was inter-polated in the calibration lines for each reaction product. Thecalibration lines were obtained using pNP at concentrationsranging from 5 to 50 μM (7 data points, r2¼0.999) for lipase,alkaline phosphatase, and protease and using CNP at con-centrations ranging from 5 to 60 μM (7 data points, r2¼0.998)for α-amylase. The protein concentrations of the workingenzyme solutions were determined using the Bradfordmethod as modified by Zor and Selinger (1996), using bovineserum albumin as the standard. The specific enzymatic activitiesof lipase, alkaline phosphatase, and protease were expressed asμmol of pNP released per minute per mg protein (μmol pNPmin�1 mg�1 protein), while the specific enzyme activity ofα-amylase was expressed as the μmol of CNP released perminute per mg protein (μmol CNPmin�1 mg�1 protein).

2.4. Theoretical studies of digestive enzyme inhibitionby pectic polysaccharides

2.4.1. Structure of the gastrointestinal enzymesThe structure of each enzyme was obtained from the ProteinData Bank (http://www.rcsb.org/pdb, consulted on March 5,2013). Porcine pancreatic lipase from Sus scrofa (EC 3.1.1.3, pdbcode 1ETH), porcine pancreatic α-amylase from Sus scrofa (EC3.2.1.1, pdb code 1DHK), bovine pancreatic chymotrypsinfrom Bos taurus (EC 3.4.21.4, pdb code 1S0Q), and humanplacental alkaline phosphatase from Homo sapiens (EC 3.1.3.1,pdb code 3MK1) were used for the molecular dockingcalculations.

2.4.2. Structure of the artificial substratesThe molecular structure of the artificial substrates pNPPA,G3CNP, pNPA, and pNPP was fully optimized using theHartree–Fock method, as implemented in GAMESS (GeneralAtomic and Molecular Electronic Structure System) versionMay 01 of 2012 (Schmidt et al., 1993) in the 6-311G(d,p) basisset. Solvent effects were simulated by placing the substratesin dielectric medium simulating water, using the IEF-PCMmodel (Cossi, Barone, Mennucci, & Tomasi, 1998). The geo-metric optimization and the energy calculations were per-formed using this medium.

2.4.3. Structure of the pectic polysaccharidesPP fragments α-D-GalpA-(1-4)-[α-D-GalpA-(1-4)]18-α-D-GalpA-(1-1)–OH with MDs ranging (in mol/mol) from 0% to 100%were constructed in the biomolecular force field GLYCAM 06for carbohydrates (Kirschner et al., 2008). Energy minimiza-tion of the PP fragments was performed using the AMBER 9force field, and the assigning of partial charges was con-ducted using Gasteiger function partial charges. In all cases,

the pdb format files that were built were employed in themolecular docking calculations.

2.4.4. Molecular docking protocolThe basic docking protocol was performed using the defaultsettings provided by AutoDock Tools, according to Neuhaus(2010). The enzymes, artificial substrates and PP fragmentswere converted to pdbqt format in Autodock Tools. TheLamarckian Genetic Algorithm with a population size of 150dockings and five million energy evaluations was used. Allother parameters, e.g., the crossover rate and mutation rate,were obtained using the default settings. The grid size forspecifying the search space was set at 21�21�21 Å3 in acentered position with default grid-point spacing of 0.375 Å.Autodock 4.0 was launched from Autodock Tools on Devian-Linux operating system, and the docking logs were analyzedusing the graphical user interface of Autodock Tools. Thedocked energy was defined as the sum of the intermolecularand the internal energies. For a representative dockinginstance, the orientation or pose with the lowest estimatedfree energy (ΔG) of binding, corresponding to the dockingenergy and unbound free energy of the system, was chosen ineach calculation. The binding free energy ratio (BFER) wasdefined as the ratio of the free energy binding of eachenzyme–PP complex and the free energy binding of theenzyme–substrate complex.

2.5. Data analysis

All measurements were conducted with three analytical andthree technical replicates, for a total of nine replicates (n¼9).The mean values and their standard deviations werereported. Comparisons among the mean values were per-formed by one-way variance analysis (ANOVA) and usingFisher's least significant difference test (po0.05), using the Rprogram (version 2.13.1, July 08 of 2011).

3. Results and discussion

3.1. Pectic polysaccharide materials

Alkaline hydrolysis of HMPP was performed to obtain PPswith different MD, thus the PPs obtained had MDs (%mol/mol)ranging from 87.475.4% (HMPP) to 7.172.7% (LMPP) (Table 1).In addition to the MD, the acetylation degree and molecularweight distribution of the obtained PPs were also evaluatedbecause these molecules are potentially susceptible theeffects of the alkaline hydrolysis conditions, and thesefeatures determine the structure, three-dimensional confor-mation, and functional properties of the PPs (Mohnen, 2008).The alkaline treatment did not affect (po0.05) the acetylationdegree (average of acetylation degree was 4.970.4% (mol/mol)) nor the molecular weight values (HPSEC profiles of theobtained PPs did not show significant differences; Fig. 1).

Alkaline hydrolysis was performed at low temperature(pHZ10, 45 min, 25 1C) because it has been established thatalkaline hydrolysis of the α-(1,4) glycosidic bond of PPs shouldnot be performed at temperatures above 60 1C to preventβ-elimination from becoming the primary mechanism of

5 10 15 20

Elution Time (min)

RI R

espo

nse

Fig. 1 – High performance size exclusion chromatography(HPSEC) elution profiles of the pectic polysaccharides (PPs)after 0 (continuous line) and 45 min (dotted line) of de-esterification with 0.25 M NaOH at 25 1C, to obtain PPs withmethylation degrees of 87.4% and 7.1% (mol/mol),respectively.

Table 1 – Effect of the alkaline hydrolysis of the high-methylated pectic polysaccharide on the methylation andacetylation degrees.

Time(min)

Methylation(%mol/mol)1

Acetylation(%mol/mol)2

0 87.475.4ª 4.470.3ª9 64.675.3b 4.870.3a

15 39.175.0c 5.370.1a

30 28.473.8d 4.970.4a

45 7.172.7e 5.470.4a

Different letters within the same column indicate significantdifferences as calculated using Fisher's least significant differencetest (po0.05).1 Expressed as moles of methyl esters per 100 mol of uronic acid.2 Expressed as moles of acetyl esters per 100 mol of uronic acid.

B i o a c t i v e C a r b o h y d r a t e s a n d D i e t a r y F i b r e 4 ( 2 0 1 4 ) 2 7 – 3 8 31

degradation (Krall & McFeeters, 1998). We found that themolecular weight distribution was not affected by this pro-cess (Fig. 1). This result is most likely because hydrolysis ofthe α-(1,4) glycosidic bond of the PPs, which is required todecrease the molecular weight, is highly efficient underextreme temperature and acidity conditions (pHr2.0, 3 h,100 1C) (Ramaswamy, Kabel, Schols, & Gruppen, 2013). It iswell known that the alkaline hydrolysis conditions may affectboth the MD and acetylation degree, depending on thetemperature, alkali concentration, and the initial number ofavailable methyl and acetyl groups present in the PP. Never-theless, in this study, we found that the alkaline conditionsdid not affect significantly the acetylation degree, most likelydue to the low number of initial acetylation sites availablecompared to the high number of initial methylation sitespresent in HMPP (Garna, Mabon, Nott, Wathelet, & Paquot,2006; Sundar-Raj, Rubila, Jayabalan, & Ranganathan, 2012).In contrast to the stability of the acetylation degree andmolecular weight during the alkaline treatment, there was a

statistically significant decrease (po0.05) of the MD over time.After 45 min of alkaline treatment, almost complete de-esterification of HMPP was obtained, yielding an LMPP witha MD of 7.172.7% (mol/mol).

The MD was the only structural parameter that differen-tiated the PPs used in the enzyme inhibition experiments,meaning that the observed differences in the biologicalfunctionalities of the PPs are most likely due to differencesin their MD. The structural importance of MD in the PPs lies inthe fact that the methyl-ester groups neutralize the negativecharges present in the free carboxyl groups, decreasing theirpolar nature and increasing their hydrophobicity (Mohnen,2008). Thus, the MD is a parameter that affects the biologicalfunctionality of the PPs and it can be a factor that determinesthe interaction of the PPs with other biologically relevantbiomolecules, such as proteins (Benjamin, Lassé, Silcock, &Everett, 2012).

3.2. Kinetics of the inhibition of the digestive enzymaticactivities by the pectic polysaccharides

The effect of the MD and of the concentration of each PP onthe activities of the enzymes lipase, α-amylase, alkalinephosphatase, and protease (chymotrypsin) was evaluated.The kinetic profiles of the in vitro digestion of artificialsubstrates by the enzymes in the presence of HMPP areshown in Fig. 2. Increasing the concentration of HMPPdecreased (po0.05) the activity of lipase, α-amylase, alkalinephosphatase, and protease. The same trend was observedusing each of the other PPs. Because the increase in theconcentration of each PP in the reaction medium caused adecrease in the activity of each enzyme, one might supposethat the tested PPs displayed inhibitory behavior toward theenzymatic activities and that the PPs therefore might behavekinetically as a defined type of enzyme inhibitor (competitive,noncompetitive or uncompetitive). Inhibitory behaviortoward enzymes was previously reported for other types ofdietary fiber, such as cellulose (Dunaif & Schneeman, 1981),agar–agar, carboxymethyl cellulose, sodium alginate, xylan,inulin (Ikeda & Kusano, 1983) pectin, mucilages, polyethyleneglycol (Isaksson et al., 1982a), and dietary fiber extracted fromwheat bran, oat bran, and alfalfa (Dunaif & Schneeman, 1981).In all cases, increasing the substrate concentration did notovercome the inhibitory effect of the PPs (the maximumvelocity, Vmax, of each enzyme inhibited by the PPs did notequal the Vmax of the uninhibited enzyme at any substrateconcentration). Lipase was the enzyme most affected by theinhibitory effect of each PP among the enzymes studied(Fig. 2a).

That HMPP utilized the kinetic mechanism of non-competitive inhibition of the activities of the studiedenzymes was confirmed using the double-reciprocal plotmethod of Lineweaver–Burk (Fig. 3). All of the tested PPsbehaved as non-competitive inhibitors of the enzymes. Thisbehavior suggested that PPs can interact with both the freeenzyme and the enzyme–substrate complex at a domaindifferent from the catalytic site of the enzymes (Rehm &Becker, 1988). It was also possible to verify that the Vmax ofthe enzymes inhibited with PPs did not equal the Vmax of thenon-inhibited enzyme at any substrate concentration

0 2 4 6 8 100.0

0.4

0.8

1.2

pNPPA (mM)

Lipa

se A

ctiv

ity( μ

molpN

P m

in-1

mg-1

pro

tein

)

0.0 0.2 0.4 0.6 0.8 1.00

2

4

6

G3CNP (mM)

α-A

myl

ase

Act

ivity

( μm

ol C

NP

min

-1 m

g-1 p

rote

in)

0.0 0.2 0.4 0.6 0.8 1.00

10

20

30

pNPP (mM)

Phos

phat

ase

Act

ivity

( μm

olpN

P m

in-1

mg-1

pro

tein

)

0.0 0.5 1.0 1.5 2.00

3

6

9

pNPA (mM)

Prot

ease

Act

ivity

( μm

olpN

P m

in-1

mg-1

pro

tein

)

Fig. 2 – Effect of the concentration of high-methylated pectic polysaccharide (HMPP) (methylation degree of 87.4%) on thekinetic behaviors of lipase (a), α-amylase (b), alkaline phosphatase (c), and protease (d). The concentrations of HMPP were 0 (●),20 (○), 40 (▲), 60 (△), 80 (■), and 100 μg mL�1 (□).

B i o a c t i v e C a r b o h y d r a t e s a n d D i e t a r y F i b r e 4 ( 2 0 1 4 ) 2 7 – 3 832

because the apparent maximum velocity (Vmaxapp) decreasedsignificantly as the PP concentrations increased (Fig. 4; onlythe results from using HMPP are shown). The presence of thePPs in the reaction medium did not significantly modify theES affinity or the efficiency of ES complex formation (kineti-cally represented by the Michaelis–Menten constant, Km),which are distinctive behaviors of non-competitive enzymeinhibition.

The variance analysis (ANOVA, with po0.05) performed foreach kinetic profile (from Fig. 4) revealed that neither the MDlevels nor the PP concentration affected the Km value. Theobserved kinetic mechanism of non-competitive inhibitionexplains the fact that this type of inhibition cannot be overcomeby increasing the substrate concentration because in this typeof inhibition the PPs do not compete for the catalytic site of theenzyme. This result is important in understanding the enzyme–PP interaction phenomenon because it could be expected thatan interaction of enzymes with the PPs could be only superficial(e.g., at an allosteric regulation site) without any compromise ofthe catalytic site of the enzyme.

The decrease in the Vmaxapp (Eq. (1)) was dependent on theconcentration of each PP. According to the non-competitiveinhibition mechanism, this parameter corresponds to thefollowing:

Vmaxapp ¼ Vmax

1 þ ð PP½ �=KiÞð1Þ

where Ki is the non-competitive inhibition constant of aninhibited enzyme with each PP at a given concentration [PP].This constant governs the process of non-competitive inhibi-tion of each enzyme. Using Eq. (1), it was possible to find theKi constant for each PP with a given MD. Fig. 5 shows that alinear tendency, with a negative slope, was revealed uponplotting the Ki against MD. An increase in the MD caused asignificant decrease in the Ki. In the context of non-competitive enzymatic inhibition, the Ki can be interpretedas the inhibitor concentration required to reduce the enzy-matic activity by 50% (IC50, Cheng & Prussof, 1973), namely, aPP with a high Ki (LMPP) is less efficient as a non-competitiveinhibitor of enzymes than is a PP with a low Ki (HMPP).

All of the tested enzymes were inhibited by the PPs withany MD by means of the same molecular mechanism (non-competitive inhibition). The efficiency of the inhibition wasdetermined from the slope (m) of the straight line that wasobtained when plotting Ki versus MD (Fig. 5). Interestingly,each enzyme was inhibited with different efficiencies. Lipase(m¼�1.943 μg mL�1%�1) was more likely to be inhibited bythe PPs than were α-amylase (m¼�1.558 μg mL�1%�1), alka-line phosphatase (m¼�1.344 μg mL�1%�1), or protease(m¼�1.165 μg mL�1%�1). These differences could be attribu-ted to the interaction capacity of each enzyme and the PPs,which is mediated by the structural characteristics of each ofthe molecule involved in the interaction (Rodríguez-Patiño &Pilosof, 2011).

-1.2 -0.8 -0.4 0.0 0.4 0.80

2

4

6

8

10

1 / pNPPA (mM-1)

1 / L

ipas

e A

ctiv

ity

-3 0 3 60.0

0.5

1.0

1.5

1 / G3CNP (mM-1)

1 / α

-Am

ylas

e A

ctiv

ity

-4 -2 0 2 4 60.00

0.05

0.10

0.15

0.20

1 / pNPP (mM-1)

1 / P

hosp

hata

se A

ctiv

ity

-4 -2 0 2 40.0

0.1

0.2

0.3

0.4

0.5

1 / pNPA (mM-1)

1 / P

rote

ase

Act

ivity

Fig. 3 – Lineweaver–Burk plot of lipase (a), α-amylase (b), alkaline phosphatase (c), and protease (d) inhibited with high-methylated pectic polysaccharide HMPP (methylation degree of 87.4%). The kinetic profile obtained suggested that HMPPbehaves as a non-competitive inhibitor of digestive enzymes. The concentrations of HMPP were 0 (●), 20 (○), 40 (▲), 60 (△), 80(■), and 100 μg mL�1 (□).

B i o a c t i v e C a r b o h y d r a t e s a n d D i e t a r y F i b r e 4 ( 2 0 1 4 ) 2 7 – 3 8 33

Ikeda and Kusano (1983) have suggested two hypothesesto explain the inhibitory effect of dietary fiber on the activityof trypsin. The first hypothesis states that the inhibition ofenzymatic activity may result from the interaction of dietaryfiber with the substrate, preventing the enzyme–substrateinteraction, and the second hypothesis states that there is aninteraction between dietary fiber and trypsin, making theenzyme unable to perform its catalytic function. Regardingthese hypothesis, we found that the inhibitory effect ofthe PPs on the activities of the enzymes appears to bedue to a non-competitive enzyme–PP interaction. The non-competitive inhibitory mechanism of the PPs on the activitiesof the enzymes revealed in this study (Fig. 3) is most likelydue to the large size of the PPs, which generally have sizesand molecular dimensions comparable to those of theenzymes, suggesting that the interaction of a PP with anenzyme may be superficial (Zhao, Diao, & Zong, 2013). Never-theless, it is important to consider that the overall inhibitoryeffect of the PPs toward a gastrointestinal tract activity mightbe caused by several mechanisms, such as modification ofthe composition and structure of the interface, substratecoating with a PP layer or embedding of the substrates withinPP particles (McClements & Li, 2010; Miled, Beisson, de Caro,de Caro, Arondel, & Verger, 2001; Reis, Holmberg, Watzke,Leser, & Miller, 2009). Also, it has been previously establishedthat the inhibition of some digestive processes can be

enhanced by the increase of viscosity of the gastrointestinalfluids and by flocculation of lipids due to the presence ofpolysaccharides in the gastrointestinal tract (McClements,2000). At sufficiently high concentration, polysaccharidesform a three-dimensional network of interacting or entangledmolecules than traps substrates and enzymes and effectivelyinhibit their movements and interactions (Dickinson, 2009).

3.3. Nature of the enzyme–PP interaction

The interaction of PPs with enzymes may be controlled by thestructural parameters of both the PPs (MD, acetylation degree,and molecular weight distribution) and the digestiveenzymes (size, molecular weight, surface hydrophobicity,and three-dimensional structure). Table 2 shows some ofthe structural characteristics of the enzymes of interest inthis study. Size is a structural parameter that might beimportant in the interaction of each enzyme with the PPs.The size of the enzymes may be represented by severalstructural characteristics, such as the molecular weight,volume and accessible surface area. Lipase, with the highestdegree of inhibition by the PPs, is the enzyme with thehighest molecular weight, volume, and surface accessiblearea of those studied. The large size of the lipase moleculemight result in an increased probability of interaction withthe PPs and therefore an increased susceptibility of being

0 20 40 60 80 1000.0

0.5

1.0

1.5

0.0

0.5

1.0

1.5

2.0

Pectin (μg mL-1)

V max

App

( μm

olpN

P m

in-1

mg-1

pro

tein

)

Km

(mM

)

0 20 40 60 80 1000

2

4

6

8

10

0.0

0.2

0.4

0.6

0.8

1.0K

V

Pectin (μg mL-1)

V max

App

( μm

ol C

NP

min

-1 m

g-1 p

rote

in)

Km

(mM

)

0 20 40 60 80 1000

10

20

30

40

0.0

0.2

0.4

0.6

0.8

1.0K

V

Pectin (μg mL-1)

V max

App

( μm

olpN

P m

in-1

mg-1

pro

tein

)

Km

(mM

)

0 20 40 60 80 1000

2

4

6

8

10

0.0

0.2

0.4

0.6

0.8

1.0K

V

Pectin (μg mL-1)

V max

App

( μm

olpN

P m

in-1

mg-1

pro

tein

)

Km

(mM

)

Fig. 4 – Effect of the concentration of high-methylated pectic polysaccharide (HMPP, methylation degree of 87.4%) on the apparentmaximum velocity (Vmaxapp) and Michaelis–Menten constant (Km) of lipase (a), α-amylase (b), alkaline phosphatase (c), andprotease (d). In non-competitive inhibition, the Km remains constant whereas the Vmaxapp decreases when the concentration ofHMPP increases. An ANOVA analysis (po0.05) showed that the HMPP concentration did not significantly affect the Km.

0 20 40 60 80 1000

50

100

150

200

m = -1.943 ± 0.112

Methylation Degree (%)

Ki (

μg m

L-1)

0 20 40 60 80 1000

50

100

150

200

m = -1.558 ± 0.109

Methylation Degree (%)

Ki (

μg m

L-1)

0 20 40 60 80 1000

50

100

150

200

m = -1.344 ± 0.154

Methylation Degree (%)

Ki (

μg m

L-1)

0 20 40 60 80 1000

50

100

150

200

m = -1.165 ± 0.139

Methylation Degree (%)

Ki (

μg m

L-1)

Fig. 5 – Effect of the methylation degree (MD) of the pectic polysaccharides (PPs) on the non-competitive inhibition constant (Ki)of lipase (a), α-amylase (b), alkaline phosphatase (c), and protease (d). The efficiency of inhibition was measured as the slope(m) of the straight line obtained when plotting Ki versus MD. High-methylated PP had the highest ability to inhibit theactivities of the digestive enzymes.

B i o a c t i v e C a r b o h y d r a t e s a n d D i e t a r y F i b r e 4 ( 2 0 1 4 ) 2 7 – 3 834

Table 2 – Structural properties of lipase, α-amylase, alkaline phosphatase, and protease. Lipase has the greater probabilityof having surface interactions with hydrophobic pectic polysaccharides (such as high-methylated pectic polysaccharide,HMPP) due to their larger size and its large exposed surface area. The high content of hydrophobic amino acids confersupon the lipase enzyme its predominantly hydrophobic character and its high capacity for interaction with HMPP.

Enzyme MWa pIb Vc Asd THA (%)e ΔGtr w-o

f

Lipase 99.7 (896) 6.11 (�12.0) 170,960 31,063 58.1 (521) �74.5α-Amylase 55.4 (496) 6.17 (�5.2) 111,734 16,811 56.0 (278) �40.5Alkaline phosphatase 52.7 (484) 6.12 (�7.8) 76,503 14,611 52.5 (254) �35.7Protease 23.3 (223) 8.34 (þ6.3) 33,795 6,753 49.8 (111) �16.1

a MW: molecular weight (kDa). The number of total amino acids is in parentheses. The data was obtained from the Protein Data Bank.b pI: isoelectric point. The estimated charge at pH 7.0 is in parentheses. The data was obtained from the Protein Data Bank.c V: protein volume (Å3). Calculated according to Voss and Gerstein (2010).d As: accessible surface area (Å2). Calculated according to Miller, Janin, Lesk, and Chothia (1987).e THA: total hydrophobic amino acids (sum of Phe, Trp, His, Tyr, Ala, Ile, Leu, Val, Pro, and Gly). The percent of hydrophobic residues is given asnumber of hydrophobic residues per 100 residues. The number of hydrophobic residues is in parentheses. The data was obtained from theProtein Data Bank.

f ΔGtr w-o (kcal mol�1): free energy transfer from water to the bilayer interface (surface hydrophobicity). Calculated according to Eisenhaber(1996).

B i o a c t i v e C a r b o h y d r a t e s a n d D i e t a r y F i b r e 4 ( 2 0 1 4 ) 2 7 – 3 8 35

inhibited by the PPs. Thus, as the size of the enzyme decreases(lipase, As 31,063 Å2; α-amylase, As 16,811 Å2; alkaline phospha-tase, As 14,611 Å2; and protease, As 6,753 Å2), its ability tointeract with and to be inhibited by the PPs also decreases.

It was also observed that each enzyme was more susceptibleto inhibition by HMPP than by LMPP. This behavior could begoverned by the physical nature of the non-covalent intermo-lecular forces operating in the enzyme–PP interaction(McClements, 2006). The difference in the hydrophobicity ofLMPP and HMPP might also explain the non-covalent interac-tion of the PPs with each protein. Due to the presence ofnegatively charged free carboxyl groups, LMPP is hydrophilic innature and can interact more efficiently with proteins via anelectrostatic mechanism. In contrast, due to the presence ofmethylated carboxyl (methyl-ester) groups of neutral nature,HMPP is more hydrophobic and can interact more efficientlywith proteins via a hydrophobic mechanism. Lipase, α-amylase,and alkaline phosphatase, which have isoelectric points (pI)slightly below the pH (pH 7.0) of the model solution, would havea negative electrostatic charge (net charge of the enzymes at pH7.0 in all cases is very close to neutrality). The negativeelectrostatic nature of these enzymes could hinder their inter-action with LMPP due to negative electrostatic repulsions. Suchrepulsive electrostatic forces could explain the weak interactionbetween the enzymes and the LMPPs.

The total hydrophobic amino acids (THA) and the freetransfer energy of the enzyme from the aqueous phase (w)toward the organic (lipidic) phase (o, ΔGtr w-o) were calcu-lated to estimate the hydrophobicity (Eisenhaber, 1996) ofeach enzyme. Lipase (containing 58.1% THA) is the enzymewith the greatest number of hydrophobic amino acids (Phe,Trp, His, Tyr, Ala, Ile, Leu, Val, Pro, and Gly) present in itsstructure; whereas, as the hydrophobicity of the enzymesdecreased (α-amylase, 56.0% THA; alkaline phosphatase,52.5% THA; and protease, 49.0% THA), their ability to interactand be inhibited by HMPP also decreased. The parameter ΔGtr

w-o (Eq. (2)) is defined as

ΔGtr w-o ¼ �RT LnSoSw

� �ð2Þ

Where So corresponds to the solubility of the enzyme in theorganic (lipidic) phase and Sw corresponds to the solubility ofthe enzyme in the aqueous phase. In the case of hydrophobicenzymes, for which So4Sw, the ΔGtr w-o is negative, whereasin hydrophilic enzymes, for which SooSw, the ΔGtr w-o ispositive. The more negative the ΔGtr w-o, the greater the So,as well as the hydrophobicity of the enzyme, meaning that aprotein has a strong tendency to be transferred from withinthe aqueous solution to the interface.

Examining this parameter showed that lipase (ΔGtr w-

o¼�74.51 kcal mol�1) is the enzyme with the highest hydro-phobicity, whereas α-amylase, alkaline phosphatase, andprotease were the enzymes with low hydrophobicity. Inter-estingly, the enzymes with low ΔGtr w-o values (α-amylase,�40.46 kcal mol�1; alkaline phosphatase, �35.66 kcal mol�1;and protease, �16.07 kcal mol�1) were those which were lessinhibited by HMPP than lipase. This finding might indicatethat electrostatic interactions are less important in the over-all enzyme–PP interaction compared to hydrophobic interac-tions, which are generally dominant (Miled et al., 2001;McClements, 2006), and that the mechanism that could bedeterminant in the enzyme–PP interactions might be hydro-phobic in nature (McClements, 2006; Hur, Lim, Decker, &McClements, 2011).

The interfacial phenomena that occur at the oil–waterinterface of the simulated gastrointestinal fluid utilized inthis study also might govern the efficiency of the enzyme–PPinteraction. Lipase is the enzyme that is most affected byinterfacial characteristics, such as the substrate, bile salts,and lipase interfacial concentrations, as well as by its hydro-phobic character and its high susceptibility to be affected bythe highly hydrophobic HMPP molecule (McClements, 2006;Reis et al., 2009). Lipases are water-soluble enzymes with alimited activity toward substrates in aqueous media, but theyexhibit high activity when the substrate is at a concentrationhigh enough to form micelles in the presence of a surfactantor when the substrate is present in an emulsified medium(Miled et al., 2001).

This particular behavior occurs because lipases areenzymes that are resistant to the denaturing conditions of

B i o a c t i v e C a r b o h y d r a t e s a n d D i e t a r y F i b r e 4 ( 2 0 1 4 ) 2 7 – 3 836

interfaces, such as the high concentration of surfactantagents (Reis et al., 2009). It can be suggested, therefore, thatlipases are enzymes that are particularly susceptible to beinginhibited by the PPs, not only due to the direct effect theyhave on the activity of these enzymes, but also due to theeffect they have on the stability of the emulsion formedunder the simulated gastrointestinal tract conditions. It hasalso been suggested that the PPs can behave as anionicsurfactants that may compete with a lipase for a position inthe interface, displacing the enzyme and inhibiting its activ-ity by decreasing its interfacial concentration (Reis et al.,2009).

In this study we did not conduct any emulsion preparationfor any enzyme test. For the lipase model, the chromogenicgroup of the pNPPA allows this compound to remain inaqueous solution. It was only necessary the addition of asmall amount of Tween 20 to improve the solubility of pNPPAin the buffer solution. It is known, nevertheless, that tocatalyze its enzymatic reaction, lipase must absorb to theoil–water interface so that it is in close proximity to lipiddroplets. Lipase generally does this as part of a complex withco-lipase and possibly with bile-salts (Singh, Ye, & Horne,2009). Competitive adsorption processes occur at the lipid-droplet surfaces among the enzyme complex, bile salts,phospholipids, digestion products, and other surface-activesubstances such as PPs, which could interfere with lipaseadsorbing to the droplet surfaces. Thus, in an emulsifiedsystem (not tested by us), lipid-droplet surfaces might becoated with PP layers that inhibit the direct access of thelipase/co-lipase complex to the lipid droplets (Singh et al.,2009; McClements & Li, 2010).

Fig. 6 – Structure of the enzyme–substrate-inhibitor complexes fophosphatase (c), or protease (d) when inhibited by the high-metof 100% mol/mol. In all cases, the PP fragments were docked on(catalytic) site of the enzymes.

3.4. Theoretical studies of digestive enzyme inhibitionby pectic polysaccharides

To evaluate the differences in the enzyme–PP interaction thatwere experimentally observed in this study, a theoreticalmodel was proposed. This model allowed calculating theinteraction energy for each enzyme–PP pair. The interactionenergies were calculated using molecular docking methodol-ogy, by searching for the site on the enzyme where theinteraction with the PP was most likely to occur. Fig. 6 showsthe structures of the enzyme–HMPP interaction. In all cases,the enzyme–PP interaction occurred at a site different fromthe catalytic site of each enzyme (where substrate is located),which is consistent with a non-competitive inhibitionmechanism. Due to the large size that PPs can have, theywere not expected to have a significant interaction with thecatalytic site of an enzyme, which generally represents only asmall fraction of the total surface area of an enzyme. Thesites of the enzymes for which the PPs had more affinity wereprotein domains with a relatively high abundance of super-ficial hydrophobic amino acids. For example, in the case oflipase, it was found that HMPP had high affinity for a highlyhydrophobic domain that includes Gly47, Leu136, Leu140,Trp131, Tyr142, Gly354, Tyr373, Gly415, Trp436, Val437, andLeu443, all of which are amino acids with a hydrophobicnature. This observation suggested that the hydrophobicinteractions in the enzyme–PP coupling are the interactionsthat could mostly govern the non-competitive enzyme inhi-bition that was experimentally observed.

As was experimentally observed, although all of theenzymes were inhibited by the PPs with any MD by means

rmed by the substrates and lipase (a), α-amylase (b), alkalinehylated pectic polysaccharide (PP) with a methylation degreethe surface of the enzymes at a site different from the active

0 40 80 1200

5

10

15

20

Methylation Degree (%)

BFE

R

Fig. 7 – Effect of the methylation degree (ranging from 0% to100% (mol/mol)) of pectic polysaccharides (PPs) on the freeenergy of binding of the enzyme–PP complexes. The bindingfree energy ratio (BFER) was defined as the ratio of the freeenergy of binding of an enzyme–PP complex and the freeenergy of binding of the enzyme–substrate complex.

B i o a c t i v e C a r b o h y d r a t e s a n d D i e t a r y F i b r e 4 ( 2 0 1 4 ) 2 7 – 3 8 37

of the same molecular mechanism, each enzyme was inhib-ited with different efficiencies. Fig. 7 shows the calculatedenzyme–PP interaction energies expressed as the BFER. TheBFER value corresponds to the relationship between the freeenergy of the enzyme–PP interaction and the free energy ofthe enzyme–substrate interaction. The BFER is a measure ofhow many fold greater is the affinity of enzyme for the PPthan for its respective substrate. It was observed that for anyMD, all the enzymes had greater affinity for the PPs than forthe substrates. The BFER increased as the MD increased.These results suggested that an increase in the MD promotesenzyme–PP interactions and increases the efficiency of thenon-competitive enzyme inhibition.

The trend of inhibition observed in our experiments (Fig. 5)was theoretically validated. For a PP with a theoretical MD of100%, lipase had 18.4 times greater affinity for the inhibitorthan for the substrate, whereas α-amylase (12.3 times), alka-line phosphatase (8.0 times), and protease (4.2 times) hadweaker interactions with the inhibitor (Fig. 7). Furthermore,for a PP with a theoretical MD of 0%, lipase had 6.2 timesgreater affinity for the inhibitor than for the substrate,whereas α-amylase (4.3 times), alkaline phosphatase (2.7times) and protease (1.4 times) were less inhibited (Fig. 7).Thus, it could be suggested that the enzyme–PPs hydrophobicinteractions are critical to the efficiency with which the PPsinhibit the activity of the enzymes via a non-competitivemechanism.

4. Conclusions

The kinetic analyses performed in this study suggested thatthe PPs function as non-competitive inhibitors of the gastro-intestinal tract enzymes. The high degree of methylation ofthe PPs and the hydrophobicity of the enzymes significantlycontribute to the enzyme–PP surface interactions and to theefficiency of the inhibitory effect of the PPs on the in vitroactivity of the enzymes in model solutions. The order of themagnitude with which the enzymes were inhibited by the PPs

was lipase4α-amylase4alkaline phosphatase4protease.Based on our results, we suggest that PPs might be able tosuppress nutrient absorption from the small intestine byinhibiting gastrointestinal tract enzymatic activities.

Acknowledgments

The authors are grateful to COLCIENCIAS and the UniversidadNacional de Colombia for providing a fellowship to MauricioEspinal-Ruiz that supported this study.

r e f e r e n c e s

Amarowicz, R, Kmita-Glazewska, H, & Kostyra, H (1990).Influence of fiber on the enzymatic digestion of casein.Fagopyrum, 10, 69–72.

Benjamin, O, Lasse, M, Silcock, P, & Everett, D W (2012). Effect ofpectin adsorption on the hydrophobic binding sites ofβ-lactoglobulin in solution and in emulsion systems.International Dairy Journal, 26, 36–40.

Brownlee, I A (2011). The physiological roles of dietary fiber. FoodHydrocolloids, 25, 238–250.

Chaudhari, G, Chatterjee, S, Venu-Babu, P, Ramasamy, K, &Thilagaraj, W R (2013). Kinetic behavior of calf intestinalalkaline phosphatase with pNPP. Indian Journal of Biochemistry &Biophysics, 50, 64–71.

Cheng, Y C, & Prussof, W H (1973). Relationship between theinhibition constant (Ki) and the concentration of inhibitorwhich causes 50 per cent inhibition (I50) of an enzymaticreaction. Biochemical Pharmacology, 22, 3099–3108.

Cossi, M, Barone, V, Mennucci, B, & Tomasi, J (1998). Ab initiostudy of ionic polarizable continuum dielectric model.Chemical Physics Letters, 286, 253–260.

Dawson, D C (1993). Gastrointestinal physiology: The molecularbasis of G.I. transport. Annual Review of Physiology, 55, 571–573.

Dickinson, E (2009). Hydrocolloids as emulsifiers and emulsionstabilizers. Food Hydrocolloids, 23, 1473–1482.

Dongowski, G (1997). Effect of the pH on the in vitro interactionsbetween bile acids and pectin. European Food Research andTechnology, 205, 185–192.

Dunaif, G, & Schneeman, B O (1981). The effect of dietary fiber onhuman pancreatic enzyme activity in vitro. The AmericanJournal of Clinical Nutrition, 34, 1034–1035.

Eisenhaber, F (1996). Hydrophobic regions on protein surfaces:Derivation of the solvation energy from their area distributionin crystallographic protein structures. Protein Science, 5,1676–1686.

Garna, H, Mabon, N, Nott, K, Wathelet, B, & Paquot, M (2006).Kinetic of the hydrolysis of pectin galacturonic acid chainsand quantification by ionic chromatography. Food Chemistry,96, 477–484.

Holloway, W D, Tasman-Jones, C, & Maher, K (1983). Pectindigestion in humans. The American Journal of Clinical Nutrition,37, 253–255.

Houben, K, Jolie, R P, Fraeye, I, Van Loey, A M, & Hendrickx, M E(2011). Comparative study of the cell wall composition ofbroccoli, carrot, and tomato: Structural characterization of theextractable pectins and hemicelluloses. Carbohydrate Research,346, 1105–1111.

Hur, S J, Lim, B O, Decker, E A, & McClements, D J (2011). In vitrohuman digestion models for food applications. Food Chemistry,125, 1–12.

B i o a c t i v e C a r b o h y d r a t e s a n d D i e t a r y F i b r e 4 ( 2 0 1 4 ) 2 7 – 3 838

Ikeda, K, & Kusano, T (1983). In vitro inhibition of digestiveenzymes by indigestible polysaccharides. Cereal Chemistry, 60,260–263.

Isaksson, G, Lundquist, I, & Ihse, A (1982a). Effect of dietary fiberon pancreatic enzyme activity in vitro. Gastroenterology, 82,918–924.

Isaksson, G, Lundquist, I, & Ihse, A (1982b). In vitro inhibition ofpancreatic enzyme activities by dietary fiber. Digestion, 24,54–59.

Kakkar, T, Boxenbaum, H, & Mayersohn, M (1999). Estimation of Ki

in a competitive enzyme-inhibition model: Comparisonsamong three methods of data analysis. Drug Metabolism andDisposition, 27, 756–762.

Kay, R M (1982). Dietary fiber. Journal of Lipid Research, 23, 221–242.Kirschner, K N, Yongye, A B, Tschampel, S M, Gonzalez-Outeirino,

J, Daniels, C R, Foley, L, et al. (2008). GLYCAM 06: Ageneralizable biomolecular force field. Carbohydrates. Journalof Computational Chemistry, 29, 622–655.

Krall, S. M., & McFeeters, R. F. (1998). Pectin hydrolysis: Effect oftemperature, degree of methylation, pH, and calcium onhydrolysis rates. Journal of Agricultural and Food Chemistry, 46,1311–1315.

Kris-Etherton, P. M., Hecker, K. D., Bonanome, A., Coval, S. M.,Binkoski, A. E., Hilpert, K. F., et al. (2002). Bioactive compoundsin foods: Their role in the prevention of cardiovascular diseaseand cancer. The American Journal of Medicine, 113, 71S–88S.

Kumar, A., & Chauhan, G. S. (2010). Extraction andcharacterization of pectin from Apple pomace and itsevaluation as lipase (steapsin) inhibitor. Carbohydrate Polymers,82, 454–459.

Louis, P., Scott, K. P., Duncan, S. H., & Flint, H. J. (2007).Understanding the effects of diet on bacterial metabolism inthe large intestine. Journal of Applied Microbiology, 102,1197–1208.

McClements, D. J. (2000). Comments on viscosity enhancementand depletion flocculation by polysaccharides. FoodHydrocolloids, 14, 173–177.

McClements, D. J. (2006). Non-covalent interactions betweenproteins and polysaccharides. Biotechnology Advances, 24,621–625.

McClements, D. J., & Li, Y. (2010). Review of in vitro digestionmodels for rapid screening of emulsion-based systems. Food &Function, 1, 32–59.

Miled, N., Beisson, F., de Caro, J., de Caro, A., Arondel, V., & Verger,R. (2001). Interfacial catalysis by lipases. Journal of MolecularCatalysis B: Enzymatic, 11, 165–171.

Miller, S., Janin, J., Lesk, A. M., & Chothia, C. (1987). Interior andsurface of monomeric proteins. Journal of Molecular Biology, 196,641–656.

Mohnen, D. (2008). Pectin structure and biosynthesis. CurrentOpinion in Plant Biology, 11, 266–277.

Morishita, Y., Iinuma, Y., Nakashima, N., Majima, K., Mizuguchi,K., & Kawamura, Y. (2000). Total and pancreatic amylasemeasured with 2-chloro-4-nitrophenyl-4-O-α-D-galactopyranosyl maltoside. Clinical Chemistry, 46, 928–933.

Neuhaus, F. C. (2010). Roles of Arg301 in substrate orientation and

catalysis in subsite 2 of D-alanine: D-Alanine (D-lactate) ligasefrom Leuconostoc mesenteroides: A molecular docking study.Journal of Molecular Graphics and Modelling, 28, 728–734.

Ramaswamy, U. R., Kabel, M. A., Schols, H. A., & Gruppen, H.(2013). Structural features and water holding capacities of

pressed potato fibre polysaccharides. Carbohydrate Polymers,93, 589–596.

Rehm, H., & Becker, C. M. (1988). Interpreting non-competitiveinhibition. Trends in Pharmacological Sciences, 9, 316–317.

Reis, P., Holmberg, K., Watzke, H., Leser, M. E., & Miller, R. (2009).Lipases at interfaces: A review. Advances in Colloid and InterfaceScience, 147-148, 237–250.

Rodrıguez-Patino, J. M., & Pilosof, A. M. R. (2011). Protein–

polysaccharide interactions at fluid interfaces. FoodHydrocolloids, 25, 1925–1937.

Rothman, S. S. (1977). The digestive enzymes of the pancreas: Amixture of inconstant proportions. Annual Review of Physiology,39, 373–389.

Schmidt, M. W., Baldridge, K. K., Boatz, J. A., Elbert, S. T., Gordon,

M. S., Jensen, J. H., et al. (1993). General atomic and molecularelectronic structure system. Journal of Computational Chemistry,14, 1347–1363.

Singh, H., Ye, A., & Horne, D. (2009). Structuring food emulsions inthe gastrointestinal tract to modify lipid digestion. Progress inLipid Research, 48, 92–100.

Sundar-Raj, A. A., Rubila, S., Jayabalan, R., & Ranganathan, T. V.

(2012). A review on pectin: Chemistry due to generalproperties of pectin and its pharmaceutical uses. ScientificReports, 1, 550–553.

Tsujita, T., Takaichi, H., Takaku, T., Sawai, T., Yoshida, N., &

Hiraki, J. (2007). Inhibition of lipase activities by basicpolysaccharides. Journal of Lipid Research, 48, 358–365.

van den Hoogen, B. M., van Weeren, P. R., Lopes-Cardozo, M., vanGolde, L. M. G., Barneveld, A., & van de Lest, C. H. A. (1998). Amicrotiter plate assay for the determination of uronic acids.

Analytical Biochemistry, 257, 107–111.Verma, S. K., & Ghosh, K. K. (2013). Activity, stability and kinetic

parameters of α-chymotripsin catalyzed reactions in AOT/isooctane reverse micelles with nonionic and zwitterionicmixed surfactants. Journal of Chemical Science, 125, 1–8.

Voragen, A. G. J., Schols, H. A., & Pilnik, W. (1986). Determination

of the degree of methylation and acetylation of pectins byHPLC. Food Hydrocolloids, 1, 65–70.

Voss, N. R., & Gerstein, M. (2010). 3V: Cavity, channel and cleftvolume calculator and extractor. Nucleic Acids Research, 38,W555–W562.

Zhao, G. Y., Diao, H. J., & Zong, W. (2013). Nature of pectin–protein–cathechin interactions in model systems: Pectin–

protein–cathechin interactions. Food Science and TechnologyInternational, 19, 153–165.

Zor, T., & Selinger, Z. (1996). Linearization of the Bradford proteinassay increases its sensitivity: Theoretical and experimental

studies. Analytical Biochemistry, 236, 302–308.