Effects of the proteinaceous moiety on the emulsifying properties of sugar beet pectin

ARTICLE IN PRESS

FOODHYDROCOLLOIDS

0268-005X/$ - s

doi:10.1016/j.fo

�CorrespondE-mail addr

Food Hydrocolloids 21 (2007) 1319–1329

www.elsevier.com/locate/foodhyd

Effects of the proteinaceous moiety on the emulsifyingproperties of sugar beet pectin

Takahiro Funamia,�, Guoyan Zhanga, Mika Hiroea, Sakie Nodaa, Makoto Nakaumaa,Iwao Asaia, Mary K. Cowmanb, Saphwan Al-Assafc, Glyn O. Phillipsd

aHydrocolloid Laboratory, San-Ei Gen F.F.I., Inc., 1-1-11, Sanwa-cho, Toyonaka, Osaka 561-8588, JapanbOthmer Department of Chemical & Biological Sciences & Engineering, Polytechnic University, Six Metrotech Center, Brooklyn, NY 11201, USA

cGlyn O. Phillips Hydrocolloid Research Centre, North East Wales Institute, Plas Coch, Mold Road, Wrexham LL11 2AW, UKdPhillips Hydrocolloids Research Ltd., 45 Old Bond Street, London W1S 4QT, UK

Received 20 January 2006; accepted 4 October 2006

Abstract

The role of the proteinaceous moiety in emulsifying was investigated using pectin from sugar beet as a model polysaccharide.

Physicochemical and macromolecular characteristics of sugar beet pectin were examined with or without an enzymatic modification

using multiple acid-proteinases. The enzymatic modification decreased the total protein content from 1.5670.15% to 0.1370.02% by

the Bradford method without significant change in ferulic acid or most constitutional sugars. It also decreased the weight-average

molecular weight (Mw) from 517728 to 254720 kg/mol and the z-average root-mean-square radius of gyration from 43.670.8 to

35.070.6 nm. Emulsifying properties of the polysaccharide with or without the enzymatic modification were evaluated by emulsion

droplet size and creaming stability of O/W emulsions (pH 3.0) containing 15w/w% middle-chain triglyceride and 1.5w/w% sugar beet

pectin as main constituents. The modification increased the average diameter (d3,2) of emulsion droplets from 0.5670.04 to

3.0070.25mm immediately after the preparation, suggesting a decrease in the emulsifying activity. It caused the creaming of the

emulsions during incubation at 60 1C, which was in line with the finding that macroscopic phase separation occurred only in the presence

of the modified pectin after storage at 20 1C for a day, suggesting a decrease in the emulsion stabilizing ability. The modification also

decreased significantly the amount of the pectin fraction that adsorbed onto the surface of emulsion droplets from 14.5872.21% to

1.2270.03% and the interfacial concentration of the polysaccharide from 1.4270.23 to 0.4570.05mg/m2, where the proteinaceous

materials in the pectin molecules activated the oil-water interface. Results from the present study suggest an important role of the

proteinaceous moiety to explain the emulsifying properties of sugar beet pectin as in the case of gum arabic and soy soluble

polysaccharide.

r 2006 Elsevier Ltd. All rights reserved.

Keywords: Sugar beet pectin; Emulsifying properties; Proteinaceous moiety; Enzymatic modification; Oil-water interface

1. Introduction

Many functions are expected to food hydrocolloids,including gelling, thickening, stabilizing, dispersing, waterholding, foaming, and emulsifying, etc. Some natural foodhydrocolloids exhibit emulsifying properties (emulsifyingactivity and emulsion stabilizing ability), including pectin(Akhtar, Dickinson, Mazoyer, & Langendorff, 2002;

ee front matter r 2006 Elsevier Ltd. All rights reserved.

odhyd.2006.10.009

ing author. Tel.: +816 6333 0521; fax: +81 6 6333 2076.

ess: [email protected] (T. Funami).

Leroux, Langendorff, Schick, Vaishnav, & Mazoyer,2003; Williams et al., 2005), gum arabic (Ray, Bird,Iacobucci, & Clark, 1995), microcrystalline-cellulose (Garti& Reichman, 1993), galactomannans (Garti & Reichman,1994), and soy soluble polysaccharide (Nakamura, Yoshida,Maeda, Furuta, & Corredig, 2004; Nakamura, Takahashi,Yoshida, Maeda, & Corredig, 2004; Nakamura, Yoshida,Maeda, & Corredig, 2006). Emulsifying properties of thesehydrocolloids are normally regarded as a function of theproteinaceous moiety because the character of polysacchar-ide itself is predominantly hydrophilic though the steric and

www.elsevier.com/locate/foodhyd

dx.doi.org/10.1016/j.foodhyd.2006.10.009

mailto:[email protected]

ARTICLE IN PRESST. Funami et al. / Food Hydrocolloids 21 (2007) 1319–13291320

mechanical stabilization effects of the carbohydrate moietyare not negligible on the emulsion. Actually, emulsifyingactivity of gum arabic at the oil–water interface is attributedto a small amount of protein fraction that is covalentlybound to highly branched polysaccharide structures, de-tected as a high molecular mass (Dickinson, Elverson, &Murray, 1989; Dickinson, Galazka, & Anderson, 1991;Randall, Phillips, & Williams, 1988), suggesting that thearabinogalactan-protein (AGP)-type fraction plays majorrole in emulsifying (Al-Assaf, Katayama, Phillips, Sasaki,& Williams, 2003; Islam, Phillips, Sljivo, Snowden, &Williams, 1997). Similarly, emulsifying properties of soysoluble polysaccharide are ascribed to the proteinaceousmoiety that is covalently bound to a high molecular weightfraction of the carbohydrate backbone, adsorbing onto theoil–water interface as an anchor, whereas the carbohydratemoiety contributes to emulsion stability due to a stericrepulsion, preventing aggregation or coalescence of theemulsion droplets through the formation of a hydrated layer(Nakamura, Yoshida, et al., 2004; Nakamura, Takahashi, etal., 2004; Nakamura et al., 2006).

Pectin is widely used in the food industries utilizing itsgelling (as in jams, jellies, and marmalades), thickening,and stabilizing properties (as in acidified milk beverages)(Voragen, Pilnik, Thibault, Axelos, & Renard, 1995).Pectin products commercially available are extracted fromcitrus peel (lemon, lime, grapefruit, orange, etc) and applepomace in most cases, but pectin is also obtained fromsugar beet pulp as a byproduct during the extraction ofsugar. As common features of pectin molecules (Voragenet al., 1995), the backbone consists of a-(1-4)-linkedD-galacturonic acid units interrupted by the insertion of(1-2)-linked L-rhamnopyranosyl residue in adjacent oralternate positions; whereas the side chains consist mainlyof D-galactose and L-arabinose as found in galactan,arabinogalactan, and arabinan with a considerable mole-cular length, linked glycosidically to O4 and/or O3 of theL-rhamnopyranose. Physicochemical differences betweenpectin from sugar beet and that from ‘‘conventional’’sources include that sugar beet pectin has a higherproportion of neutral-sugar side chains (rich in hairyregions) (Williams et al., 2005), a higher content of acetylgroup at O2 and O3 positions within the galacturonicbackbone (Rombouts & Thibault, 1986), a higher contentof phenolic esters in the side chains especially arabinoseand galactose (Colquhoun, Ralet, Thibault, Faulds, &Williamson, 1994; Guillon, Thibault, Rombouts, Voragen,& Pilnik, 1989; Ralet, Thibault, Faulds, & Williamson,1994; Rombouts & Thibault, 1986), and a higher content ofthe proteinaceous materials bound to the side chainsthrough covalent linkages (Williams et al., 2005). Pectinfrom sugar beet does not form gels thermally even in thepresence of high concentration of soluble solids (e.g.,sugar) at low pH (o3–4) conditions (different from highmethoxyl pectins forming gel matrixes through a combina-tion of hydrophobic interactions within the methoxylgroups and hydrogen bondings within the hydroxyl

groups) (Oakenfull & Scott, 1984) or in the presence ofcalcium ions (different from low methoxyl pectins forminggel matrixes through coordinate bonds), which may beattributed to the higher proportion of acetyl groups(Pippen, McCready, & Owens, 1950) and of the lateralchains (Keenan, Belton, Matthew, & Howson, 1985). Themain application of sugar beet pectin in the food industriesis as an emulsifier rather than as a gelling or stabilizingagent.We can access the following three papers that provide us

with unique insights into the emulsifying properties ofpectin (not only from sugar beet but also from citrus). Thefirst one (Leroux et al., 2003) elucidated that sugar beetpectin is effective in stabilizing rapeseed oil or orange oilemulsions. It also suggests that molecular weight, proteincontent, and acetyl content of the polysaccharide influencesignificantly its emulsifying properties and that theproteinaceous materials associated with the pectin mole-cules play a key role in stabilizing the emulsions. Thesecond one (Akhtar et al., 2002) elucidated that depoly-merized citrus pectin with an average molecular weight of70 kg/mol exhibits the best functionality to stabilize O/Wemulsions containing rapeseed oil or D-limonene under anacidified condition (i.e. pH 4.7). The third one (Williams etal., 2005) elucidated using fractionations of sugar beetpectin through hydrophobic affinity chromatography thatthe accessibilities of protein and ferulic acid to the surfaceof the oil droplets influence the emulsifying properties ofthe polysaccharide but that no simple relationship is foundbetween the emulsifying properties and each of thesefactors. All these papers suggest that the emulsifyingproperties of pectin are not determined by a single factor,which still remains some ambiguous aspects of theemulsifying mechanism. Also, presence of the arabinoga-lactan (AG) side-chain fraction, which functions biologi-cally to connect the rhamnogalacturonan backbone tohemicellulose and/or cellulose (Hwang, Pyun, & Kokini,1993), and of the AGP-type side-chain fraction, which isdetected with a high molecular weight in the size exclusionchromatography (Oosterveld, Voragen, & Schols, 2002),has been elucidated in pectin. To accumulate the knowl-edge on the emulsifying properties of pectin, especiallypectin from sugar beet, whose usage is somewhat behindconventional ones, should be emphasized to spread itsapplication areas in the food industries.We targeted sugar beet pectin as a model polysaccharide

in the present study. The initial phase of the study involvedthe determination of physicochemical and macromolecularcharacteristics of the polysaccharide with or without anenzymatic modification using multiple acid-proteinases toremove the proteinaceous materials. The secondary phasecovered the evaluation of O/W type emulsions in thepresence of the polysaccharide with or without theenzymatic modification. The objective of the study is toexplain the emulsifying properties of sugar beet pectin byfocusing on the role of the proteinaceous materialsespecially bound to the carbohydrate moiety.

ARTICLE IN PRESST. Funami et al. / Food Hydrocolloids 21 (2007) 1319–1329 1321

2. Materials and methods

2.1. Materials

Sugar beet pectin (batch G33–215 manufactured by CPKelco, Denmark), extracted from dry materials using anacidified medium, was subjected to the following enzymaticmodification after purification through alcohol precipita-tion. Samples of sugar beet pectin (5w/v%) were solubi-lized in distilled de-ionized water at 80 1C with mechanicalstirring at 1500 rpm for 30min in a batch size of 1000mLand precipitated by adding 3 volumes of 95w/v% ethanolinto the solution. The samples precipitated were lyophilizedafter the evaporation of the solvent under vacuum at 35 1C,followed by manual pulverization in a mortar. The overallpercentage recovery in this purification step was �90%.Sugar beet pectin thus purified was stored in a desiccator at20 1C until use.

Reagent grade of ferulic acid, neutral sugars includingrhamnose (Rha), arabinose (Ara), galactose (Gal), glucose(Glc), and xylose (Xyl), and galacturonic acid (GalA) wereused without further purifications (Wako Pure ChemicalInd., Ltd., Osaka, Japan).

2.2. Enzymatic modification of sugar beet pectin

Aqueous solutions of sugar beet pectin (1w/v%) wereprepared by dissolving the polysaccharide in distilled de-ionized water at 60 1C with mechanical stirring at 1000 rpmfor 15min in a batch size of 100mL. Stock solution ofpepsin (E.C.3.4.23.1, Wako Pure Chemical, 2700 unit/mg)or a food grade acidic proteinase (Amano Enzyme Inc.,Aichi, Japan, 7000 unit/g) was prepared by dissolving eachenzyme in 50mM citrate buffer (pH 3.0) and in distilled de-ionized water, respectively. Solutions of each enzyme atvarious concentrations were added into the pectin solu-tions, and the mixtures were incubated at 37–40 1C for 14 h.The mixture was heated at 100 1C for 10min to stop theenzymatic reaction, and the reaction solutions obtainedwere dialyzed against distilled de-ionized water at 20 1Cthrough a dialysis membrane with a 10 kg/mol molecularweight cut off, followed by lyophilization. The lyophilizedsamples were stored in a desiccator at 20 1C until use.

2.3. Physicochemical characteristics of sugar beet pectin

2.3.1. Constitutional sugars

Sugar beet pectin was hydrolyzed by a combination ofacid and enzyme pectinase according to the previous report(Garna, Mabon, Wathelet, & Paquot, 2004) with somemodifications. Sugar beet pectin (50mg) was solubilized in10mL of distilled de-ionized water, mixed with 1mL of0.2w/v% pectinase (E.C.3.2.1.15. Amano Enzyme Inc.,1200 unit/g), and incubated at 40 1C for 24 h. One milliliterof 3M H2SO4 was then added to the mixture and heated at98 1C for 2 h. Hydrolyzed samples obtained were neutra-lized with 1mL of 6M NaOH and diluted to a definite

concentration (100 ppm) with distilled de-ionized water.Sample solutions (25 mL) were injected to the high-performance anion-exchange chromatography coupledwith pulsed amperometric detection (HPAEC-PAD) sys-tem, DIONEX DXc-500 (Nippon Dionex K.K., Osaka,Japan), where a CarboPac PA-1 column (copolymer ofstyrene vinyl benzene and divinyl benzene; 4mm in innerdiameter and 250mm in length) was used as a separationcolumn. A mobile phase was changed in a gradient mode inorder: 5mM NaOH (0–20min), 5–100mM NaOH(20–30min), 100mM NaOH+0–100mM CH3COONa(30–50min), and 200mM NaOH (50–60min). Analyseswere carried out at a constant temperature of 30 1C and ata flow rate of 1mL/min. The pH of the mobile phase waskept constant using a post-column system by circulating500mM NaOH. Data were presented as means7SD oftriplicate.

2.3.2. Ferulic acid

Sugar beet pectin (10mg) was decomposed in 5mL of0.5M NaOH under N2 gas at 20 1C for 24 h and acidifiedwith 0.75mL of 6M HCl. Ferulic acid was extracted fromthe mixture in two steps using 4mL of ethyl acetate as asolvent and the residue was reconstituted in 10mL ofdistilled de-ionized water after the evaporation of thesolvent. Absorbance of test solutions was read at 310 nm at20 1C using a V-560 spectrophotometer (JASCO Interna-tional Co., Ltd., Tokyo, Japan) to determine the content offerulic acid. Data were presented as means7SD oftriplicate.

2.3.3. Protein (Bradford method)

Bradford protein assay kit (Pierce, USA) was used todetermine the total protein content in sugar beet pectinwith or without the enzymatic modification through anabsorbance reading at 595 nm as a result of the reactionbetween protein and Coomassie Brilliant Blue G-250.Bovine serum albumin was used as a standard substancefor calibration within a working range between 0 and750 mg/mL. It should be noted that this assay did notnecessarily represent absolute values of protein concentra-tion and was used to confirm the action of the enzymesbecause the assay is generally insensitive to low molecularweight proteins or polypeptides and can be influenced bybound materials with protein. Data were presented asmeans7SD of triplicate.

2.4. Macromolecular characteristics of sugar beet pectin

2.4.1. Size exclusion chromatography coupled with

multiangle laser light-scattering (SEC-MALS)

The SEC system consisted of a PU-980 HPLC pump(JASCO International Co., Ltd.), a Shodex OHpak SB-Gguard column (SHOWA DENKO K.K., Tokyo, Japan),and a Shodex OHpak SB-806M HQ separation column(SHOWA DENKO K.K.). Nominal exclusion limit of theseparation column was 20.0� 103 kg/mol. An on-line

ARTICLE IN PRESST. Funami et al. / Food Hydrocolloids 21 (2007) 1319–13291322

degasser JASCO DG-980–50 was used to remove airbubbles from the eluent, aqueous solutions of 0.05MNaNO3 containing 0.01w/v% NaN3 as a preservative,circulated throughout the system at a flow rate of 0.5mL/min. The MALS measurements were carried out at 25 1Cusing a DAWN-DSP (Wyatt Technology Co., CA, USA)with linearly polarized light of l ¼ 632.8 nm (He-Ne laser)to determine the weight-average molecular weight Mw, thez-average root-mean-square radius of gyration Rg, and thepolydispersity index. Scattering intensity was determined atangles from 261 to 1321 simultaneously with multipledetectors. Pectin solutions (100 mL) at 0.05w/v%, preparedby two passes through a high-pressure homogenizerNanomizer (Yoshida Kikai Kogyo, Aichi, Japan) at50MPa, was injected to the SEC system after filteringthrough cellulose–acetate membrane filters of 0.45 mm poresize. The increase in refractive index RI with concentration(dn/dc in mL/g), determined using an OPTILAB DSP(Wyatt Technology Co.), was 0.131mL/g, equivalent to theprevious report (Corredig & Wicker, 2001), and overallpercentage recovery was deduced from RI (RI-930, JASCOInternational Co., Ltd.) reading. UV (UV-970, JASCOInternational Co., Ltd.) signal was detected at a wave-length of 214 nm to prevent overlapping with the signalsfrom ferulic acid. The measurements were completedwithin 1 h after the preparation of test solutions. Datawere presented as means7SD of triplicate.

2.5. Preparation of O/W emulsions

Sugar beet pectin (1.5 g) was dispersed in 80 g of distilledde-ionized water using a Polytron-type homogenizer at24000 rpm for 1min, into which 1mL of 10w/v% benzoicacid (as a preservative) and the same amount of 10w/v%citric acid (to adjust the final pH of the emulsions at 3.0)were added with mixing using the homogenizer at8000–10000 rpm for 1min. Distilled de-ionized water and15 g of middle-chain triglyceride (MCT) (Nisshin OillioGroup, Ltd., Tokyo, Japan), synthesized from triglycerideand a mixture of C8 fatty acid and C10 fatty acid at a ratioof 75:25 and with a density of �0.95 g/mL, were added toweigh 100 g, then mixed again with the homogenizer at24000 rpm for 1min to prepare crude emulsions. The crudeemulsions were applied to two passes through the high-pressure homogenizer at 50MPa and subjected to thefollowing measurements to know the emulsifying proper-ties of sugar beet pectin.

2.6. Emulsifying properties of sugar beet pectin

2.6.1. Emulsion droplet size and interfacial characteristics of

sugar beet pectin

Droplet size distribution of the O/W emulsions wasdetermined by a computer-controlled laser diffractionapparatus SALD-2100J (Shimadzu Co., Kyoto, Japan)and was analyzed by an installed software Wing SALD-2100 ver. 1.10. The emulsions, diluted to an appropriate

range for the apparatus with distilled de-ionized water at20 1C using a magnetic stirrer, were subjected to thedetermination immediately after the preparation. Averagedroplet diameters of the emulsions were determined as d3,2and d4,3, representing the surface–volume mean diameterand the volume-moment mean diameter, respectively. Thespecific surface area Sv of the whole emulsion wascalculated as

Sv ¼ 6f=d3;2ðm2=mL emulsionÞ,

where f represents the volume fraction of the dispersedphase (Puppo et al., 2005). The volume fraction f wasdetermined separately by density measurement using aDMA 48 density meter (Anton Paar, Tokyo, Japan):

f ¼ ðraq � remÞ=ðraq � roilÞ,

where raq, rem, and roil represent the density of aqueousphase, emulsion, and oil, respectively (Buffo, Reineccius, &Oehlert, 2001). Interfacial concentration of sugar beetpectin at the surface of emulsion droplets G was thencalculated as:G (mg/m2) ¼ adsorbed pectin concentration C (mg/mL

emulsion)/Sv (m2/mL emulsion) (Puppo et al., 2005).Method to determine the C will be described later inSection 2.6.4. Data were presented as means7SD oftriplicate.

2.6.2. Creaming stability

Creaming stability of the O/W emulsions was measuredby the backscattering of light from the emulsions as afunction of height using a Turbiscan LAb Expert (EkoInstruments Co., Ltd., Tokyo, Japan). Backscattered lightfluxes were detected at 880 nm after incubation at 60 1C for30, 60, 90, and 120min, and changes in the scatteringintensity after each incubation time were determined:

D Backscattering ð%Þ

¼ Data at the initial ð%Þ

�Data after each incubation time ð%Þ.

This was a kind of acceleration test at the relatively hightemperature to know the difference in emulsion stabilizingability of sugar beet pectin before and after the enzymaticmodification. One representative datum was shown as aresult for each treatment with the reproducibility confirmedfrom repeated experiments using three different emulsionsamples.

2.6.3. Interfacial tension

Interfacial tension was determined by pendant dropprocedure using a Drop Master 700 (Kyowa InterfaceScience Co., Ltd, Saitama, Japan) at 20 1C. Tests werecarried out using MCT as a dispersed phase and aqueoussolutions of sugar beet pectin with or without theenzymatic modification or instead distilled de-ionizedwater (as a reference) as a dispersion medium to evaluatethe surface activity. Aqueous solutions of sugar beet pectin


were prepared by dissolving the polysaccharide in distilledde-ionized water using the Polytron-type mixer at24000 rpm for 1min in a batch size of 100mL, followedby two passes through the high-pressure homogenizer at50MPa. Concentration of the polysaccharide in themedium was 1.76w/w%, and the pH of the solution wasadjusted at 3.0 using citric acid. The MCT was droppedinto each dispersion medium using an injection syringe(Fig. 1), and the shape factor b of the pendant drophanging from the syringe top was determined by ds and deon the images. Interfacial tension g was then calculated as(Hansen & Rfdsrud, 1991):

g ¼ DrgR20=b,

where Dr, g, and R0 represent difference in density betweenthe dispersed phase and the dispersion medium, gravita-tional constant, and radius of the pendant drop at apex,respectively. Density of each liquid was determinedseparately using the DMA 48 density meter. Data werepresented as means7SD of triplicate.

2.6.4. Mass distribution of sugar beet pectin within

emulsions

Mass distribution of sugar beet pectin within the O/Wemulsions was determined using the SEC-(MALS)-RIsystem (Section 2.4.1) by calculating concentrations ofthe polysaccharide either in an aqueous phase or adsorbedonto the surface of emulsion droplets using 0.131mL/g as adn/ds value. The emulsions were subjected to the determi-nations immediately after the preparation. Each pectinfraction was recovered based on the procedures reportedpreviously for the isolation of fat globules from milk(Patton & Huston, 1986) and for the separation of proteinin oil or creaming phase (Puppo et al., 2005) with somemodifications. In detail, the O/W emulsions (5 g) wereplaced into a centrifugation tube (40mL), and mixed with20 g of solution A (36w/w% sugar and 6w/w% NaCl indistilled de-ionized water) using a Bortex-type mixer for1min. The mixtures were centrifuged at 25000 g for 30minat 10 1C to recover the lower aqueous phase with an aid ofmicropipette. The residues (i.e. intermediate creamingphase and upper oil phase) were centrifuged again at thesame condition after mixing with the solution A (totalweight 25 g) to recover the lower aqueous phase, which was

Fig. 1. Determination of the interfacial tension by pendant drop procedure. T

hanging from the injection needle (syringe) as in (a). In the present study, since t

of sugar beet pectin (dispersion media), a pendant drop was formed upward t

repeated three times. Recovered samples (i.e. pectinfraction in an aqueous phase of the original wholeemulsions) were collected together and subjected to thedetermination of concentrations after diluting to anappropriate range with 0.05M NaNO3. The residues weremixed with solution B (0.1w/w% SDS and 5w/w% NaClin distilled de-ionized water) (total weight 25 g) using thePolytron-type homogenizer at 10000 rpm for 2min. Here,the surfactant SDS functions to displace the proteinaceousmaterials from the oil–water interface and transfers themto an aqueous phase, which has been shown by an atomicforce microscopy (Mackie, Gunning, Wilde, & Morris,2001). The mixtures were centrifuged at 25000 g for 30minat 10 1C to recover the lower aqueous phase. The residues(i.e. intermediate creaming phase and upper oil phase) werecentrifuged again at the same condition after mixing withthe solution B (total weight 25 g) to recover the loweraqueous phase, which was repeated three times. Recoveredsamples (i.e. pectin fraction adsorbed onto the surface ofemulsion droplets) were collected together and subjected tothe determination of concentrations in a similar manner asaforementioned. Weight percentage of each fraction toinjected was determined as a mass ratio. Data werepresented as means7SD of triplicate.

2.7. Statistics

Experimental data were analyzed by t-test to know thestatistical difference between the control pectin and the modi-fied pectin with a significance defined at po0:05 at both sides.

3. Results and discussion

3.1. Characterizations of sugar beet pectin with or without

the enzymatic modification

Effects of the enzymatic modification were investigatedon ferulic acid and constitutional sugars. Ferulic acid is animportant phenolic compound bound to sugar beet pectinto explain the emulsifying properties of the polysaccharidedue to its hydrophobicity as well as the proteinaceousmoiety (Williams et al., 2005). No significant difference wasfound in the content of ferulic acid before (12.670.5mg/g)and after (12.270.4mg/g) the enzymatic modification

wo parameters, ds and de, determine the shape factor of a pendant drop

he density of oil (dispersed phase) was lower than that of aqueous solutions

he needle as in (b).

ARTICLE IN PRESS

Table 1

Ferulic acid and constitutional sugars of sugar beet pectin with or without the enzymatic modification

Treatments Ferulic acid (mg/g) Constitutional sugarsa (mg/g)

Rha Ara Gal Glc Xyl GalA

Control pectin 12.670.5 72.871.9 91.773.4 154.076.6 110.371.2 19.070.0 425.7715.4

Modified pectin 12.270.4 89.672.2* 95.778.4 149.774.8 115.073.0 17.774.8 383.9710.6*

Data were presented as means7SD of triplicate.

A value in the same column with an asterisk is significantly different from control by t-test (po0:05).aRha: rhamnose; Ara: arabinose; Gal: galactose; Glc: glucose; Xyl: xylose; GalA: galacturonic acid.

0A

bso

rba

nce

3

1

2

200 500300 400

Wavelength (nm)

200 500300 400

0

Ab

so

rba

nce

3

1

2

Wavelength (nm)

a

b

Fig. 2. UV spectra of reagent ferulic acid and sugar beet pectin with or

without the enzymatic modification. (a), 0.001w/v% reagent ferulic acid

(solid) and 0.1w/v% sugar beet pectin without the enzymatic modification

(dashed); (b), 0.08w/v% sugar beet pectin with (dashed) or without (solid)

the enzymatic modification.

T. Funami et al. / Food Hydrocolloids 21 (2007) 1319–13291324

(Table 1), and these values were almost equivalent to theprevious report (Levigne, Ralet, & Thibault, 2002). Also,no significant difference was found in the composition ofmost of the constitutional sugars before and after theenzymatic modification (Table 1). Galacturonic aciddecreased upon the proteolysis, however, which may bedue to a lower stability of the uronic acid than those of theother constitutional sugars, leading to the decompositionduring the proteolysis. Rhamnose increased upon theproteolysis, on the contrary, which seems to be difficultto interpret. Increase in rhamnose may be a relative onedue to a decrease in galacturonic acid and/or an indicationthat the neutral sugar is out of the periphery of the pectinmolecules, leading to a lower degree of ‘‘denaturation’’ incomparison with other constitutional sugars.

Changes in protein content before and after the enzy-matic modification were qualified by UV spectra. Since thewavelength scanning of pure ferulic acid (in an aqueoussolution) and sugar beet pectin (in an aqueous solution)indicated that UV peaks of the polysaccharide at around290 and 320 nm were both attributed to the absorbanceof ferulic acid and that another peak at 214 nm (A214)should be used as a measure of the proteinaceous materials(Fig. 2). The A214 decreased markedly after the enzymaticmodification, indicating a decrease in protein content. Thiswas validated quantitatively by the Bradford method,elucidating that the total protein content decreased from1.5670.15% to 0.1370.02% through the modification.

The enzymatic modification altered the macromolecularcharacteristics of sugar beet pectin (Table 2). The modifi-cation decreased the Mw from 517728 to 254720 kg/moland the Rg from 43.670.8 to 35.070.6 nm, suggesting thatthe modification makes the molecular conformation of thepolysaccharide short and compact. The determination ofthe Mw for the control pectin was somewhat larger than theprevious report (Levigne et al., 2002); 70–355 kg/molthough the macromolecular characteristics determineddepend on raw material, extraction method, and analyticalconditions, etc. The polydispersity increased from 2.3270.17 to 5.2170.79 through the modification, indicatingenlarged chain length distribution with an increase in theheterogeneity, which may be cased by digestion of proteincores connecting carbohydrate fractions with differentmolecular weights. UV peak shifted to a lower molecularweight regime (i.e. a larger elution volume) after the

enzymatic modification on the SEC-MALS profile (Fig. 3),suggesting that the proteinaceous materials bound to thecarbohydrate moiety and detected as a high molecularweight fraction should be digested successfully by themodification. Decrease in the Mw of sugar beet pectin wasless drastic in comparison with gum arabic (Connolly,Fenyo, & Vandevelde, 1988; Osman, Menzies, Williams,Phillips, & Baldwin, 1993) upon the proteolysis, which wasindicative of less-developed wattle blossom type structures(originally proposed for gum arabic by Connolly, Fenyo, &Vandevelde, 1987; Randall et al., 1988; Williams & Phillips,2000) of sugar beet pectin, though the proteolysis condi-tions were not necessarily the same between the studies.

3.2. Emulsifying properties of sugar beet pectin with or

without the enzymatic modification

Without storage, average droplet diameters d3,2 and d4,3of the O/W emulsions were 0.5670.04 and 0.7170.02 mm,


respectively, in the presence of the control pectin, but thesediameters became 5–6 times larger in the presence of themodified pectin; 3.0070.25 mm for d3,2 and 4.0770.19 mmfor d4,3, respectively (Table 3). Difference was also found inthe distribution pattern of emulsion droplet size (Fig. 4).The emulsions exhibited a monomodal distribution with asharp peak in the presence of the control pectin. In thepresence of the modified pectin, on the other hand, thedistribution was enlarged with an increase in the frequencyof fractions with extremely large (e.g., 43 mm) dropletsizes. These results suggest that sugar beet pectin loses mostpart of the emulsifying activity after the modification as inthe case of gum arabic (Randall et al., 1988) and soysoluble polysaccharide (Nakamura et al., 2006; Nakamura,Yoshida, et al., 2004).

Table 2

Macromolecular characteristics of sugar beet pectin with or without the

enzymatic modification

Treatments Mw (kg/mol) Rg (nm) Polydispersity (-)

Control pectin 517728 43.670.8 2.3270.17

Modified pectin 254720* 35.070.6* 5.2170.79*


A value in the same column with an asterisk is significantly different from

control by t-test (po0:05).Mw: weight-average molecular weight; Rg: z-average root-mean-square

radius of gyration.

Fig. 3. SEC-MALS profiles of sugar beet pectin with or without the enzymatic

v%). Separation column: Shodex OHpak SB-806M HQ (SHOWA DENKO);

Temperature: 25 1C. Area was indicated by two lines to calculate the macrom

No marked change was seen in the vertical density of theemulsions during incubation at 60 1C for 30 to 120min inthe presence of the control pectin (Fig. 5). In the presenceof the modified pectin, on the other hand, the emulsiondroplets moved upward at around 43–44mm height andbecame packed closely. These results indicated that thecreaming occurred markedly only in the presence of themodified pectin due to its lower emulsion stabilizing abilitythan the control pectin, which was in line with the findingthat macroscopic phase separation occurred only in thepresence of the modified pectin between lower aqueousphase and upper oil (and middle creaming) phase(s) afterstorage at 20 1C for a day.The interfacial tension between MCT and aqueous

solutions of sugar beet pectin increased significantly from

modification. (a), Control pectin (0.05w/v%); (b), Modified pectin (0.05w/

Mobile phase: 0.05M NaNO3+0.01w/v% NaN3; Flow rate: 0.5mL/min;

olecular characteristics.

Table 3

Average droplet size of O/W emulsions in the presence of sugar beet pectin

with or without the enzymatic modification

Treatments d3,2 (mm) d4,3 (mm)

Control pectin 0.5670.04 0.7170.02

Modified pectin 3.0070.25* 4.0770.19*



control by t-test (po0:05).The emulsions (pH 3.0), containing 15w/w% MCT (middle-chain

triglyceride) and 1.5w/w% sugar beet pectin as main constituents, were

subjected to the determinations immediately after the preparation.

ARTICLE IN PRESS

0

5

10

15

20

25

30

35

0.10

5

0.12

9

0.15

9

0.19

6

0.24

1

0.29

7

0.36

50.

45

0.55

4

0.68

20.

84

1.03

5

1.27

51.

57

1.93

3

2.38

1

2.93

2

3.61

1

4.44

7

5.47

7

6.74

6

8.30

8

10.2

31

12.6

01

15.5

18

19.112

23.5

38

28.9

88

35.7

01

43.9

68

54.1

49

Droplet size (μm)

Nu

mb

er

fre

qu

en

cy (

%)

Fig. 4. Droplet size distribution (histogram) of O/W emulsions in the presence of sugar beet pectin with or without the enzymatic modification. The

emulsions (pH 3.0), containing 15w/w% MCT (middle-chain triglyceride) and 1.5w/w% sugar beet pectin as main constituents, were subjected to the

determination immediately after the preparation. Gray: The emulsions with the control pectin; Black: The emulsions with the modified pectin.

-10

-5

0

5

10

Height (mm)

Δ B

acksca

tte

rin

g (

%)

-10

-5

0

5

10

0 15 30 450 15 30 45

Height (mm)

Δ B

acksca

tte

rin

g (

%)

60min

90min

120min

30min

60min

90,120min

30min

a b

Fig. 5. Creaming stability of O/W emulsions in the presence of sugar beet pectin with or without the enzymatic modification. (a), The emulsions with the

control pectin; (b), The emulsions with the modified pectin. Backscattered light fluxes from the emulsions were detected at 880 nm after incubation at 60 1C

for 30, 60, 90, and 120min.


17.271.8 to 35.671.3mN/m after the enzymatic modifica-tion (Table 4), suggesting that the proteinaceous materialsin the pectin molecules increase the surface activity at theoil–water interface. The determination for sugar beet pectinwas smaller than that for distilled de-ionized water (i.e.43.070.8mN/m) even after the modification, suggestingthat the carbohydrate moiety in the pectin molecules itselfexhibits some surface activity though it is much lower thanthat of the proteinaceous materials. Also, the determina-tion for the control pectin (i.e. without the enzymaticmodification) was almost equivalent to the previous report(Leroux et al., 2003); 19.4mN/m in spite of some

deviations in the method to determine the parameter(e.g., the Du Nouy ring vs. pendant drop, oil source, pectinconcentration, emulsion pH, etc).The enzymatic modification decreased significantly the

amount of sugar beet pectin that adsorbed onto the surfaceof emulsion droplets from 14.5872.21% to 1.2270.03%(Table 5). Total recovery reached to almost 100% for eachemulsion. The determination of 14.58% was almostequivalent to the previous report (Leroux et al., 2003);�10–15% in spite of some deviations in the processingparameters to prepare the test emulsions (e.g., oil source,oil load, homogenization pressure, etc). The enzymatic


modification also decreased significantly the interfacialconcentration of sugar beet pectin from 1.4270.23 to0.4570.05mg/m2 (Table 6), suggesting that the loss of theproteinaceous materials lowers the accessibility of thepolysaccharide to emulsion droplets and that the emulsify-ing properties of the polysaccharide are limited by the bulkconcentration of available surface active materials thatcover the oil–water interface. The determination of theinterfacial concentration for the control pectin corre-sponded to �15% of that for depolymerized citrus pectin(Akhtar et al., 2002), leading to an estimation that theemulsifying activity of sugar beet pectin is 6–7 times aslarge as that of citrus pectin on the equivalent weight baseat the oil–water interface though emulsion systems testedwere different between the studies.

The Mw of the pectin fraction that adsorbed onto thesurface of emulsion droplets was 4–8 times as large as thatof the pectin fraction existing in an aqueous phase of the

Table 6

Interfacial behavior of sugar beet pectin in O/W emulsions

Treatments Specific surface area (m2/mL emulsion) Adsorbed

Control pectin 1.5670.11 2.1970.21

Modified pectin 0.4070.03* 0.187o0


A value in the same column with an asterisk is significantly different from co

The emulsions were subjected to the determinations immediately after the pre

Table 5

Mass distribution of sugar beet pectin within O/W emulsions

Treatments Mass ratioa (%)

Fraction in an aqueous phase Fraction a

Control pectin 84.7373.51 14.5872.2

Modified pectin 98.7370.40* 1.2270.03


A value in the same column with an asterisk is significantly different from co

The emulsions were subjected to the determinations immediately after the preaWeight percentage of each fraction to injected.

Table 4

Surface activity of sugar beet pectin with or without the enzymatic

modification

Treatments Interfacial tension (mN/m)

Control pectin 17.271.8

Modified pectin 35.671.3*

Distilled de-ionized water 43.070.8*



control by t-test (po0:05).Tests were carried out using middle-chain triglyceride (MCT) as a

dispersed phase and an aqueous solution (pH 3.0) of each pectin (1.76w/

w%) as a dispersion medium.

emulsions (data not shown). Although it was possible thatimpurities in the test specimens caused an overestimationof the MALS determinations (Section 2.4.1), this result wasanalogous to the finding in soy soluble polysaccharide(Nakamura et al., 2006; Nakamura, Yoshida, et al., 2004);soy soluble polysaccharide consists of two fractions withdifferent Mw, and the higher Mw fraction (310 kg/mol)exhibits better emulsifying properties than the wholepolysaccharide. The lower Mw fraction (20 kg/mol), onthe other hand, affects negatively the adsorption behaviorof the polysaccharide in spite of the higher protein content.As supported by a shift of the UV signal to a lowermolecular weight regime after the enzymatic modification(Section 3.1), it is hypothesized that the proteinaceousmaterials especially bound to the carbohydrate moiety anddetected as a high molecular weight fraction are funda-mental to the adsorption at the oil–water interface as in thecase of gum arabic (Al-Assaf et al., 2003; Randall et al.,1988) and soy soluble polysaccharide (Nakamura et al.,2006; Nakamura, Yoshida, et al., 2004).

3.3. Emulsifying mechanism of sugar beet pectin

Deduced mechanism was drawn schematically (Fig. 6) toexplain the emulsifying properties of sugar beet pectin. Theproteinaceous moiety in the pectin molecules contributes toform O/W emulsions with a fine microstructure through adecrease in the interfacial tensions between water and oilphases by adsorbing onto the surface of emulsion dropletsas an anchor as in the case of soy soluble polysaccharide(Nakamura et al., 2006; Nakamura, Yoshida, et al., 2004).Loss of most part of protein results in a decrease in theemulsifying activity, making the emulsion droplets large,

concentration (mg/mL emulsion) Interfacial concentration (mg/m2)

1.4270.23

.01* 0.4570.05*

ntrol by t-test (po0:05).paration.

dsorbed onto the surface of emulsion droplets Total (recovery)

1 99.3175.72

* 99.9570.37

ntrol by t-test (po0:05).paration.

ARTICLE IN PRESS

Emulsion droplets

Proteinaceous moiety

Carbohydrate moiety

Aqueous phase

a b

Fig. 6. Schematic drawing to explain the emulsifying properties of sugar beet pectin in O/W emulsions. (a), Illustration of the emulsions with the control

pectin; (b), Illustration of the emulsions with the modified pectin.


coarse, and sometimes flocculate. Although this function ofsugar beet pectin is similar to that of gum arabic and morepreferably to soy soluble polysaccharide as mentioned, asmaller amount is required to activate the oil–waterinterface and also to stabilize the emulsions for sugar beetpectin (e.g., 1–3w/w%) in comparison with gum arabic(e.g., 10–30w/w%) or soy soluble polysaccharide (e.g.,5–10w/w%) when evaluated using the same emulsionsystem as in the present study (data not shown). Thisdifference should be discussed further not only fromprotein content and amino acid composition but also fromthe molecular relationship between the proteinaceousmoiety and the carbohydrate one, linking together toaffect the hydrophilic–hydrophobic balance of the wholepolysaccharide. That is, contributions of the carbohydratemoiety should be considered not only to the emulsionstabilizing ability through a kind of steric effects but also tothe emulsifying activity by altering the hydrophobiccharacter of the proteinaceous materials as seen in gumarabic; different fractions with an equivalent proteincontent do not show the same emulsifying properties(Ray et al., 1995), and in soy soluble polysaccharide; afraction with the higher protein content does not necessa-rily show better emulsifying properties (Nakamura, Yoshi-da, et al., 2004). Hydrophobic characters attributable toferulic acid, acetyl groups, and their combinations with theproteinaceous materials should be considered in the case ofsugar beet pectin, which makes the situation morecomplicated in comparison with other polysaccharides.Proteolysis of sugar beet pectin also results in a decrease inthe emulsion stabilizing ability, making the emulsiondroplets unstable and flocculate upon storage. Since theemulsion stabilizing ability of sugar beet pectin ispredominantly ascribed to the carbohydrate moiety, itsfunction will be compared by estimating the thickness of

hydrated layer formed around the surface of emulsiondroplets between the samples before and after the enzy-matic modification or with other polysaccharides usinghemicellulase and pectinase treatments that have beenapplicable to soy soluble polysaccharide (Nakamura et al.2006; Nakamura, Takahashi, et al., 2004).

4. Conclusions

Our study suggests that the proteinaceous moiety insugar beet pectin especially bound to the carbohydratefraction plays an important role in emulsifying to activatethe oil–water interface by adsorbing favorably onto thesurface of emulsion droplets. The proteinaceous moietyfunctions to increase the surface activity of the poly-saccharide and its accessibility to emulsion droplets, similarto those found in other polysaccharides, including gumarabic and soy soluble polysaccharide.

References

Akhtar, M., Dickinson, E., Mazoyer, J., & Langendorff, V. (2002).

Emulsion stabilizing properties of depolymerized pectin. Food Hydro-

colloids, 16, 249–256.

Al-Assaf, S., Katayama, T., Phillips, G. O., Sasaki, Y., & Williams, P. A.

(2003). Quality control of gum arabic. FFI Journal, 208, 771–780.

Buffo, B. A., Reineccius, G. A., & Oehlert, G. W. (2001). Factors affecting

the emulsifying and rheological properties of gum acacia in beverage

emulsions. Food Hydrocolloids, 15, 53–66.

Colquhoun, I. J., Ralet, M.-C., Thibault, J.-F., Faulds, C. B., &

Williamson, G. (1994). Feruloylated oligosaccharides from cell-wall

polysaccharides, Part II. Structure identification of feruloyrated

oligosaccharides from sugar-beet pulp by NMR spectroscopy.

Carbohydrate Research, 263, 243–256.

Connolly, S., Fenyo, J.-C., & Vandevelde, M.-C. (1987). Heterogeneity

and homogeneity of an arabinogalactan-protein: Acacia senegal gum.

Food Hydrocolloids, 1, 477–480.


Connolly, S., Fenyo, J.-C., & Vandevelde, M.-C. (1988). Effect of a

proteinase on the macromolecular distribution of Acacia senegal gum.

Carbohydrate Polymers, 8, 23–32.

Corredig, M., & Wicker, L. (2001). Changes in the molecular mass

distribution of three commercial pectins after valve homogenization.


Dickinson, E., Elverson, D. J., & Murray, B. S. (1989). On the film-

forming and emulsion-stabilizing properties of gum arabic: dilution

and flocculation aspects. Food Hydrocolloids, 3, 101–114.

Dickinson, E., Galazka, V. B., & Anderson, D. M. W. (1991). Emulsifying

behaviour of gum arabic. Carbohydrate Polymers, 14, 373–392.

Garna, H., Mabon, N., Wathelet, B., & Paquot, M. (2004). New method

for a two-step hydrolysis and chromatographic analysis of pectin

neutral sugar chains. Journal of Agricultural Food Chemistry, 52,

4652–4659.

Garti, N., & Reichman, D. (1993). Hydrocolloids as food emulsifiers and

stabilizers. Food Structure, 12, 411–426.

Garti, N., & Reichman, D. (1994). Surface properties and emulsification

activity of galactomannans. Food Hydrocolloids, 8, 155–173.

Guillon, F., Thibault, J.-F., Rombouts, F. M., Voragen, A. G. J., &

Pilnik, W. (1989). Structural investigations of the neutral side-chains of

sugar-beet pectins. Part 2 Enzymic hydrolysis of the ‘‘hairy’’ fragments

from sugar-beet pectins. Carbohydrate Research, 190, 97–108.

Hansen, F. K., & Rfdsrud, G. (1991). Surface tension by pendant drop. I.

A fast standard instrument using computer image analysis. Journal of

Colloid and Interface Science, 141, 1–9.

Hwang, J., Pyun, Y. R., & Kokini, J. L. (1993). Sidechains of pectins:

some thoughts on their role in plant cell walls and foods. Food

Hydrocolloids, 7, 39–53.

Islam, A. M., Phillips, G. O., Sljivo, A., Snowden, M. J., & Williams, P. A.

(1997). A review of recent developments on the regulatory, structural

and functional aspects of gum arabic. Food Hydrocolloids, 11, 493–505.

Keenan, M. H. J., Belton, P. S., Matthew, J. A., & Howson, S. J. (1985). A 13C

NMR study of sugar beet pectin. Carbohydrate Research, 138, 168–170.

Leroux, J., Langendorff, V., Schick, G., Vaishnav, V., & Mazoyer, J.

(2003). Emulsion stabilizing properties of pectin. Food Hydrocolloids,

17, 455–462.

Levigne, S., Ralet, M.-C., & Thibault, J.-F. (2002). Characterisation of

pectins extracted from fresh sugar beet under different conditions

using an experimental design. Carbohydrate Polymers, 49, 145–153.

Mackie, B. A. R., Gunning, A. P., Wilde, P. J., & Morris, V. J. (2001).

Scratching the surface: imaging interfacial structure using atomic force

microscopy. In E. Dickinson, & R. Miller (Eds.), Food colloids.

Fundamentals of formulation (pp. 13–21). Cambridge: The Royal

Society of Chemistry.

Nakamura, A., Yoshida, R., Maeda, H., Furuta, H., & Corredig, M.

(2004). Study of the role of the carbohydrate and protein moieties of

soy soluble polysaccharides in their emulsifying properties. Agricultur-

al and Food Chemistry, 52, 5506–5512.

Nakamura, A., Takahashi, T., Yoshida, R., Maeda, H., & Corredig, M.

(2004). Emulsifying properties of soybean soluble polysaccharide. Food


Nakamura, A., Yoshida, R., Maeda, H., & Corredig, M. (2006). Soy

soluble polysaccharide stabilization at oil–water interfaces. Food


Oakenfull, D., & Scott, A. (1984). Hydrophobic interaction in the gelation

of high methoxyl pectins. Journal of Food Science, 49, 1093–1098.

Oosterveld, A., Voragen, A. G. J., & Schols, H. A. (2002). Characteriza-

tion of hop pectins shows the presence of an arabinogalactan-protein.

Carbohydrate Polymers, 49, 407–413.

Osman, M. E., Menzies, A. R., Williams, P. A., Phillips, G. O., &

Baldwin, T. C. (1993). The molecular characterisation of the

polysaccharide gum from Acacia senegal. Carbohydrate Research,

246, 303–318.

Patton, S., & Huston, G. E. (1986). A method for isolation of milk fat

globules. Lipids, 21, 170–174.

Pippen, E. L., McCready, R. M., & Owens, H. S. (1950). Gelation

properties of partially acetylated pectins. Journal of American

Chemical Society, 72, 813–816.

Puppo, M. C., Speroni, F., Chapleau, N., de Lamballerie, M., Anon, M.

C., & Anton, M. (2005). Effect of high-pressure treatment on

emulsifying properties of soybean protein. Food Hydrocolloids, 19,

289–296.

Ralet, M.-C., Thibault, J.-F., Faulds, C. B., & Williamson, G.

(1994). Feruloylated oligosaccharides from cell-wall polysaccharides,

Part I. Isolation and purification of feruloyrated oligosaccharides

from cell walls of sugar-beet pulp. Carbohydrate Research, 263,

227–241.

Randall, R. C., Phillips, G. O., & Williams, P. A. (1988). The role of the

proteinaceous component on the emulsifying properties of gum arabic.


Ray, A. K., Bird, P. B., Iacobucci, G. A., & Clark, C. B. (1995).

Functionality of gum arabic. Fractionation, characterization and

evaluation of gum fractions in citrus oil emulsions and model

beverages. Food Hydoroclloids, 9, 123–131.

Rombouts, F. M., & Thibault, J.-F. (1986). Feruloyated pectic substances

from sugar beet pulp. Carbohydrate Research, 154, 177–188.

Voragen, A. G. J., Pilnik, W., Thibault, J.-F., Axelos, M. A. V., &

Renard, C. M. G. C. (1995). Chapter 10: Pectins. In A. M. Stephen

(Ed.), Food polysaccharides and their applications (pp. 287–339). New

York: Marcel Dekker, Inc.

Williams, P. A., & Phillips, G. O. (2000). Chapter 9: Gum arabic. In G. O.

Phillips, & P. A. Williams (Eds.), Handbook of hydrocolloids (pp.

155–168). Cambridge: Woodhead Publishing Ltd.

Williams, P. A., Sayers, C., Viebke, C., Senan, C., Mazoyer, J., &

Boulenguer, P. (2005). Elucidation of the emulsification properties of

sugar beet pectin. Journal of Agricultural and Food Chemistry, 53,

3592–3597.

Documents

Effects of the proteinaceous moiety on the emulsifying properties of sugar beet pectin