C:Fo

odCh

emist

ry

JFS C: Food Chemistry

Amylolytic Hydrolysis of Native Starch GranulesAffected by Granule Surface AreaJ.C. KIM, B.W. KONG, M.J. KIM, AND S.H. LEE

ABSTRACT: Initial stage of hydrolysis of native starch granules with various amylolytic enzymes, α-amylase fromBacillus subtilis, glucoamylase I (GA-I) and II (GA-II) from Aspergillus niger, and β-amylase from sweet potatoshowed that the reaction was apparently affected by a specific surface area of the starch granules. The ratios of thereciprocal of initial velocity of each amylolytic hydrolysis for native potato and maize starch to that for rice with theamylolytic enzymes were nearly equivalent to the ratio of surface area per mass of the 2 starch granules to that ofrice, that is, 6.94 and 2.25, respectively. Thus, the reciprocal of initial velocity of each enzymatic hydrolysis as ex-pressed in a Lineweaver–Burk plot was a linear function of the reciprocal of surface area for each starch granule. Asa result, it is concluded that amylolytic hydrolysis of native starch granules is governed by the specific surface area,not by the mass concentration, of each granule.

Keywords: amylase, amylolysis, granule, starch, surface area

Introduction

Among many factors influencing enzymatic hydrolysis of na-tive starch, effect of granule particle size of various botani-

cal origins has well been addressed (Ring and others 1988; Francoand others 1992; Guraya and others 2001; Yook and Robyt 2002;Kong and others 2003; Tester and others 2004). Tester and others(2006) reviewed extensively the hydrolysis of native starches withamylase. Lineweaver–Burk plots of porcine pancreatic α-amylaseactivity on native starch from potato, maize, and rice clearly de-scribed activity as a function of surface area of the granules ratherthan of substrate concentration (Kong and others 2003). The re-ciprocal of initial velocity was a linear function of the reciprocalof surface area for each starch granule. Specificity of various amy-lolytic enzymes, that is, α-amylase, β-amylase, and glucoamylase,may affect initial stage of hydrolysis. The differences in the actionpatterns of porcine pancreatic α-amylase and Bacillus amyloliq-uefaciens α-amylase with starch granules were due to differencesin the number of glucose binding sites at the active site of the2 enzymes (Yook and Robyt 2002). Two main forms of glucoamy-lase have been known, glucoamylase I (GA-I) and glucoamylase II(GA-II). GA-I is larger than GA-II, and possesses a specific domain,separate from the active site, that allows reversible binding of theenzyme to raw starch (Hayashida and others 1989). Adsorption ofamylolytic enzyme onto the native starch granule is an initial stagefor enzymatic hydrolysis of starch granule (Leloup and others 1991;Kong and others 2003). Porcine pancreatic α-amylase adsorption toboth native and gelatinized starches is of importance in the kinet-ics of the reaction (Slaughter and others 2001). As a consequence,initial rates of hydrolysis of native starch granules with various amy-lases including porcine pancreatic α-amylase may also be affectedby specific surface area rather than substrate concentration.

MS 20080275 Submitted 4/13/2008, Accepted 8/19/2008. Authors Kim, Kong,and Lee are with School of Food and Life Science and Food Science Inst. andauthor Kim is with School of Food and Life Science Food Science Inst. andBiohealth Products Research Center, Inje Univ., 607 Obangdong, Gimhae,Gyeongnam 621-749, Republic of Korea. Direct inquiries to author Kim(E-mail: jckim@inje.ac.kr).

The objective of this study was to elucidate hydrolysis of nativestarch with Bacillus subtilis α-amylase, sweet potato β-amylase,and Aspergillus niger glucoamylases as a function of particle sizeof maize, rice, and potato starch. Initial rates of hydrolysis of nativestarch granules by each enzyme were analyzed to determine the ef-fect of surface area of substrate on the amylolysis.

Materials and Methods

MaterialsNative potato, maize, and rice starches, sweet potato β-amylase

(A-7005; type I-B; 986 units/mg protein), thymerosal, and maltosewere purchased from Sigma (St. Louis, Mo., U.S.A.). B. subtilis α-amylase (Cat. Nr 101329) and A. niger glucoamylase (AMG 300 L)were obtained from Merck (Darmstadt, Germany) and Novozymes(Bagsvaerd, Denmark), respectively. To remove soluble carbohy-drates and impurities on the surface of starch granules, a properamount of starch granules was dispersed in a sufficient amount ofdistilled and deionized water and filtered using Whatman filter pa-per Nr 40 (Kent, U.K.) with suction before initiating enzyme reac-tion. All other chemicals were of analytical grade. Physicochemicalproperties of the native starches for analysis of the various amylol-ysis were utilized from the previous study (Kong and others 2003).

Purification of glucoamylase I (GA-I) andglucoamylase II (GA-II) from A. niger

A. niger GA-I and GA-II were purified by the procedure describedby Dalmia and Nikolov (1991) with a slight modification. Two hun-dred milliliters of AMG 300, a commercial product of A. niger glu-coamylase, were centrifuged (Supra 22k, Hanil Sci. Ind. Co., Korea)for 15 min at 10000 × g. The supernatant was saturated to 85%with ammonium sulfate and centrifuged for 30 min at 10000 × g.The precipitate was dissolved in 50 mL of 50 mM citrate-phosphatebuffer (pH 7.5) and desalted using HitrapTM Desalting Column (GEHealthcare Bio-Sciences, Uppsala, Sweden) against the same bufferwith a flow rate of 5 mL/min. The desalted preparation was concen-trated by ultrafiltration (Amicon, Millipore, Mass., U.S.A.) with anYM-10 membrane (Millipore). The concentrated preparation wasloaded on a HiprepTM 16/10 DEAE FF column (GE Healthcare

C© 2008 Institute of Food Technologists R© Vol. 73, Nr. 9, 2008—JOURNAL OF FOOD SCIENCE C621doi: 10.1111/j.1750-3841.2008.00944.xFurther reproduction without permission is prohibited

C:FoodChemistry

Amylolysis of native starch granules . . .

Bio-Sciences) equilibrated with 50 mM citrate-phosphate buffer(pH 7.5). GA-I and GA-II were separated and eluted with a lineargradient (from 0 to 0.3 M) of NaCl in the same buffer at a flow rateof 5 mL/min by using AKTAexplorer (GE Healthcare Bio-Sciences).The enzyme fractions were concentrated and desalted with ultrafil-tration and identified with sodium dodecyl sulfate-polyacrylamidegel electrophoresis (SDS-PAGE). Enzyme activity of purified amy-lase was measured using dinitrosalysilic acid (Somogyi 1945).

Enzyme hydrolysisEach diluted enzyme solution was added to buffer solution to

make final volume of 50 mL. Concentrations of substrates (nativestarch granules) were 5, 10, 20, 40, 100, and 200 mg/mL. A com-bination of substrate concentration for each enzyme reaction wasmade depending on the hydrolysis rate. One unit of enzyme is de-fined as the amount of enzyme liberating 1.0 μmol of maltose fromstarch in 1 min at given reaction conditions. Reaction solution forBacillus subtillis α-amylase (4.8 unit) was 0.1 M phosphate buffer(pH 6.9) and for sweet potato β-amylase (9.6 unit) and Aspergillusniger GA-I (4.8 unit) and GA-II (9.6 unit) was 0.05 M acetate buffer(pH 4.5). Thymerosal at a dilution of 1:10000 (w/v) and a few dropsof toluene were added to the suspension to avoid microbial growth(Franco and others 1992). After enzyme addition, the solution wasimmediately stirred in a shaking water bath at 100 rpm for given

Figure 1 --- Double reciprocal plots ofthe hydrolysis of native starch byvarious amylolytic enzymesexpressed as a function of substrateconcentration: (A) Bacillus subtilisα-amylase, (B) sweet potatoβ-amylase, (C) Aspergillus nigerGA-I, (D) Aspergillus niger GA-II, �:potato starch, �: corn starch, �: ricestarch.

reaction time (10, 30, 60, 120, 180 s). One milliliter of each samplesolution was mixed with 4 mL or 9 mL of 50% ethanol followed byfiltration (Whatman Nr 40) with suction. The extent of enzymatichydrolysis was determined by measuring reducing sugar using mal-tose as a standard and presented as the amount of maltose equiv-alent in solution (Kong and others 2003). Reaction temperature forBacillus subtillis α-amylase, sweet potato β-amylase, and A. nigerglucoamylase was 37, 25, and 40 ◦C, respectively.

Results and Discussion

In the hydrolysis of native starch granules by various amylases,the action pattern of the specific amylase and the botanical ori-

gin of starch govern the extent of amylolysis (Planchot and oth-ers 1995). In view of the formation of enzyme–substrate complex,it can be argued that the number of contact points between sub-strate and enzyme molecule dominates the initial stage of amyloly-sis. The position specificity of porcine pancreatic α-amylase actionon maltotriose (G3) through maltooctaose (G8) revealed that thesize of the maltodextrin affects the point of maximum frequencyattack (Robyt and French 1970). Structural studies show that theoriginal pseudo-tetrasaccharide structure of acarbose is modifiedupon binding, presumably through a series of hydrolysis and trans-glycosylation reactions (Brayer and others 2000). For solid sub-strate, the exterior or surface of the particle reacts with the enzymemolecule at the beginning of amylolysis.

C622 JOURNAL OF FOOD SCIENCE—Vol. 73, Nr. 9, 2008

C:Fo

odCh

emist

ry

Amylolysis of native starch granules . . .

Lineweaver–Burk plots of amylolysis at the initial stage as a func-tion of substrate concentration are presented in Figure 1. The re-sults agree well with many other studies (Ring and others 1988;Franco and others 1992; Guraya and others 2001; Kong and others2003) indicating that rates of amylolysis of native starches were in-versely proportional to particle size of the starch granule. Assumingall particles were spherical, the reciprocal of relative ratio of spe-cific surface area for potato to rice is 6.94, and that for maize to

Table 1 --- Comparison of slopes of Lineweaver–Burk plotof native starch hydrolysis and specific surface areaswith a reference to rice starch.

Reciprocal ofRatio of slopea relative ratio

of specificSweetB. subtilis potato A. niger A. niger surface areab

α-amylase β-amylase GA-I GA-II [1/(A/Ar)]

Potato 6.94 6.94 6.94 6.94 6.94Maize 2.23 2.24 2.24 2.24 2.25Rice 1.00 1.00 1.00 1.00 1.00aObtained from the slope of each curve in Figure 2 with a reference to that ofrice starch.bCalculation was made according to inverse relation of specific surface area tothe particle diameter and density where particle diameter of potato, maize, andrice were 43.79, 16.63, and 5.99 μm, respectively, and density of potato, maize,and rice were 1.52, 1.58, and 1.60 g/cm3, respectively, with a reference to ricestarch (Ar), and all particles are assumed to be spherical (Kong and others2003).

Figure 2 --- Double reciprocal plots ofthe hydrolysis of native starch byvarious amylolytic enzymesexpressed as a function of substrateconcentration normalized to rice: (A)Bacillus subtilis α-amylase, (B)sweet potato β-amylase, (C)Aspergillus niger GA-I, (D)Aspergillus niger GA-II, �: potatostarch, �: corn starch, �: rice starch.

rice is 2.25 (Kong and others 2003). When normalized to rice, theratios of slope for the Lineweaver–Burk plots of the hydrolysis ofpotato and maize were almost equivalent to the reciprocal of thenormalized ratio of specific surface area (Table 1). Measuring hy-drolyzed product at 10 s after initiation of amylolysis sufficientlyrepresented initial behavior of hydrolysis. As a consequence, theMichaelis–Menten equation applies, and the plot of the reciprocalof initial rates of all the examined amylolytic enzymes against thereciprocal of surface area of native starch granules is expressed as:

1V0

=(

Km

Vmax

)(ASa

)+ 1

Vmax

with Sa = AS0 where A(cm2/g) is specific surface area of starch gran-ule, Sa is surface area occupied by initial substrate concentration,S0(mg/mL) is initial concentration of substrate, V 0 is initial veloc-ity, K m is Michaelis–Menten constant, and V max is maximum ve-locity. Specifically, the abscissa in Figure 2 represents the reciprocalof substrate concentration normalized to rice, which can be seenfrom the relation that (1/S0) (1/(A/Ar)) is equal to (Ar/Sa) where Ar

is specific surface area of rice starch granule. The linear relationshiphad an R2 = 0.99 (Figure 2).

Many factors, such as amylose–amylopectin ratio, the type ofamylodextrin, the type and arrangement of crystal structures, theaverage molecular weights of the components, and the existenceof other material, make it difficult to define if granule hydrolysis is

Vol. 73, Nr. 9, 2008—JOURNAL OF FOOD SCIENCE C623

C:FoodChemistry

Amylolysis of native starch granules . . .

due to enzyme specificity, binding affinity, molecular and physicalstructure of the granule, or to a combination of factors (Williamsonand others 1992). However, it is clear that specific surface area foreach granule most significantly affects the initial stage as well asthe whole period (Ring and others 1988; Planchot and others 1995;Kong and others 2003) of amylolysis of the native starch granule.The fact that the ratio of the specific surface area of maize to potatostarch granules measured by photomicrography is 3.2 (Hellmanand Melvin 1950), which is favorably comparable to 3.08 by assum-ing that the particles are spherical (Kong and others 2003), demon-strates the assumption of sphericity for native starch granules is notnecessarily insignificant.

Differences in kinetics, as expressed in the parameters K m andV max, of the examined enzymes may come from differences in op-timum conditions and enzyme specificity. The presence of starch-binding domain in GA-I increases the affinity to starch gran-ules and subsequently accelerates rate of amylolysis compared toGA-II (Hayashida and others 1989; Dalmia and Nikolov 1991),which is consistent with results in this study (Figure 2). Furtherstudies in kinetics of raw starch hydrolysis at consecutive stepsmay provide additional information regarding the mechanism ofenzyme action and crystal structure of starch granule. This studyconfirms that enzyme reaction rate is dependent on substrate con-centration in amount per unit volume of solution since the activesite in the enzyme molecule is limited to contacting the surface ofnative starch granule at the beginning of the enzyme reaction. Inother words, the interior of the granule is not accessible as a sub-strate in amylolysis, distinctly different than that in the case of sol-ubilized (gelatinized) starch. As a consequence, the abscissa in Fig-ure 2 shows the real substrate concentration involved in the initialstage of enzyme hydrolysis normalized to rice.

Conclusions

In accordance with the amylolytic hydrolysis of native starchgranules with porcine pancreatic α-amylase (Kong and others

2003), α-amylase from Bacillus subtillis, porcine pancreatic α-amylase, Glucoamylase I(GA-I) and II(GA-II) from A. niger, andβ-amylase from sweet potato showed that the hydrolysis wasdescribed as a function of surface area of the granules ratherthan of mass (substrate) concentration. Interpreted from the rate

difference of GA-I and GA-II, resistances in initial adsorption of en-zyme molecules on the surface of starch granules is considered tobe important in explaining the differences of amylolytic hydrolysisrate among the various enzymes.

AcknowledgmentThis study was supported by the Korea Research Foundation Grantfunded by the Korean government (F00060).

ReferencesBrayer GD, Sidhu G, Maurus R, Rydberg EH, Braun C, Wang Y, Nguyen NT, Over-

all CM, Withers SG. 2000. Subsite mapping of the human pancreatic α-amylaseactive site through structural, kinetic, and mutagenesis techniques. Biochemistry39:4778–91.

Dalmia BK, Nikolov ZL. 1991. Characterization of glucoamylase adsorption to rawstarch. Enzyme Microb Technol 13:982–90.

Franco CML, Preto SJR, Ciacco CF. 1992. Factors that affect the enzymatic degradationof natural starch granules-effect of the size of the granules. Starch/Starke 44:422–6.

Guraya HS, James C, Champagne ET. 2001. Effect of cooling, and freezing on the di-gestibility of debranched rice starch and physical properties of the resulting mate-rial. Starch/Starke 53:64–74.

Hayashida S, Nakahara K, Kanlayakrit W, Hara T, Teramoto Y. 1989. Characteristicsand function of raw-starch-affinity site on Aspergilus awamori var. kawachi glu-coamylase I molecule. Agric Biol Chem 53:143–9.

Hellman NN, Melvin EH. 1950. Surface area of starch and its role in water sorption.J Am Chem Soc 72:5186–8.

Kong BW, Kim JI, Kim MJ, Kim JC. 2003. Porcine pancreatic α-amylase hydrolysisof native starch granules as a function of granule surface area. Biotechnol Prog19:1162–6.

Leloup VM, Colonna P, Ring SG. 1991. α-amylase adsorption on starch crystallites.Biotechnol Bioeng 38:127–34.

Planchot V, Colonna P, Gallant DJ, Bouchet B. 1995. Extensive degradation of nativestarch granules by alpha-amylase from Aspergilus fumigatus. J Cereal Sci 21:163–71.

Ring SG, Gee JM, Wittam M, Orford P, Johnson IT. 1988. Resistant starch: its chemicalform in foodstuffs and effect on digestibility in vitro. Food Chem 28:97–109.

Robyt JF, French D. 1970. The action pattern of Porcine pancreatic α-amylase in rela-tionship to the substrate binding site of the enzyme. J Biol Chem 245:3917–27.

Slaughter SL, Ellis PR, Butterworth PJ. 2001. An investigation of the action of porcinepancreatic α-amylase on native and gelatinized starches. Biochem Biophys Acta1525:29–36.

Somogyi M. 1945. A new reagent for the determination of sugars. J Biol Chem 160:61–8.

Tester RF, Karkalas J, Qi X. 2004. Starch structure and digestibility enzyme-substraterelationship. World’s Poult Sci J 60:186–95.

Tester RF, Qi X, Karkalas J. 2006. Review: hydrolysis of native starches with amylases.Anim Feed Sci Technol 130:39–54.

Williamson G, Belshaw NJ, Sief DJ, Noel TR, Ring SG, Cairns P, Morris VJ, Clark SA,Parker ML. 1992. Hydrolysis of A and B type crystalline polymorphs of starch byα-amylase, β-amylase and glucoamylase 1. Carbohydr Polym 18:179–87.

Yook C, Robyt JF. 2002. Reactions of alpha amylases with starch granules in aqueoussuspension giving products in solution and in a minimum amount of water givingproducts inside the granule. Carbohydr Res 337:1113–7.

C624 JOURNAL OF FOOD SCIENCE—Vol. 73, Nr. 9, 2008