2013 Volume 61 Issue 1 Pages 15-20
Cassava starch granules were treated with three types of amylase and the resulting insoluble starch granules collected and their gelling properties investigated. Gel produced from starch granules treated with either α-amylase from Bacillus amyloliquefaciens or β-amylase from barley was slightly weaker than that from native cassava starch granules. On the other hand, starch granules treated with α-amylase from Aspergillus niger produced starch gel with significantly enhanced hardness and elasticity. This observation is surprising because very limited hydrolysis (2.8%) of starch produces such a great impact on gelling properties of starch and because it contradicts the generally held view that partially hydrolyzed starch shows decreased gel-forming ability. In order to understand the mechanism behind this, we analyzed the starch component leached from starch granules into hot water using gel permeation chromatography. The results suggest that enzymatic treatment of starch granules significantly changes the properties of starch by altering the composition of leached material (matrix) from swelled starch granules (filler).
α-AMA, á-amylase from Aspergillus niger; α-AMB, α-amylase from Bacillus amyloliquefaciens: β-AMB, β-amylase from Barley.
Starch is the major storage polysaccharide in plants where it occurs as insoluble and semi-crystalline granules made of two types of glucose polymer, amylose, a linear α-1,4-linked glucan, and amylopectin, a highly branched glucan of short amylose chains branched with α-1,6 glycosidic linkages. Starch granules consist of alternating layers of amorphous and crystalline domains.1) The organization of amylose and amylopectin in the starch granule is not completely understood, but amylopectin side chains make up the framework of the crystalline lamellae, with branching points located in the amorphous domains,2) 3) 4) where amylose is also mainly localized.5) When starch granules are heated in water to above a certain temperature (the gelatinization temperature), the granules will irreversibly swell and some of the amylose fraction will be leached out. Upon cooling, the amylose in solution undergoes a process called retrogradation. If the concentration is high enough, this process results in the formation of a network which renders the solution into a gel. Starch gels are believed to be a composite of the swollen starch granules embedded in a leached amylose matrix.6) 7) Starch retrogradation may be divided into two steps, short- and long-term. Short-term retrogradation seems to occur with gelation and crystallization of amylose,8) 9) whereas long-term retrogradation is attributed to amylopectin.10)
Starch is an important renewable raw material used in various industries including foods, textiles, cosmetics, adhesives, paper and pharmaceuticals. It is used either as extracted from plants (native starch), or after chemical or enzymatic modification. Chemical modification of starch is widespread in both the food and non-food sectors since it can alter the characteristics of starch for diverse applications. The most common chemical modification processes are acid treatment, cross-linking, oxidation and substitution, including esterification and etherification. In general, these chemical modifications are carried out on insoluble starch granules. Enzymatic treatment of starch is also common, usually not to alter its properties but to convert it into low molecular weight derivatives, eg., glucose, fructose, maltose, maltodextrins and cyclodextrins. Enzymes used in starch processing are either hydrolytic enzymes (α-amylase, β-amylase, glucoamylase, α-glucosidase, isoamylase, pululanase) or transferases (transglucosidase, cyclodextrin glucanotransferase, branching enzyme, amylomaltase). Enzymatic treatment is usually carried out on the gelatinized form of starch since these enzymes hardly affect insoluble starch granules.
Enzymatic treatment of insoluble starch granules at sub-gelatinization temperature is now received increasing attention. It is mainly used to convert starch into fermentable sugars. The process requires less energy than the conventional process requiring complete gelatinization, so it is especially important for low cost production of ethanol. Another application of the enzymatic treatment of insoluble starch granules is to produce granules with altered surface structure, especially microporus granules. The porosity and surface area of starch granules are potentially important characteristics for food texture.11) 12)
Enzymatic treatment of insoluble starch granules to alter the functionality of starch is an attractive area of research but there are only a limited number of studies available. In one relatively comprehensive study, four starch granules (corn, cassava, mung bean and sago) were treated with a commercial enzyme preparation containing α-amylase from Aspergillus kawachi and glucoamylase from Aspergillus niger, and subjected to analysis for amylose content, X-ray diffraction pattern and pasting properties.13) The amylose content of corn, cassava and mung bean starch decreased significantly but that of sago starch was unchanged. None of the starches tested changed their X-ray diffraction type, but corn, mung bean and sago increased in relative crystallinity value (cassava remained the same). The peak viscosity (measured by a Rapid Visco Analyzer) of corn and cassava starch decreased but that of sago and mung bean starch increased. A similar study was carried out on cassava, sweet potato, Peruvian carrot and potato starch granules treated with α-amylase from Bacillus sp.14) The amylose content of cassava, sweet potato and Peruvian carrot starches decreased but that of potato starch was unchanged. Intrinsic viscosity and pasting properties decreased in all the hydrolyzed starches. As observed in these two studies, the effects of enzyme treatment on starch granules seems to depend on the botanical source of the starch, extent of hydrolysis and enzyme used. Retrogradation and gel formation are the most important properties of starch in the food sector, but enzyme-treated starch granules have not been subjected to gel formation.
In the present study, we treated insoluble cassava starch granules with 3 enzymes, α-amylase from A. niger (α-AMA), α-amylase from Bacillus amyloliquefaciens (α-AMB) and β-amylase from barley (β-AMB), and examined the gelling properties of the residual starch granules. Surprisingly, starch granules treated with α-AMA produced starch gel with significantly enhanced hardness and elasticity. The mechanism behind this observation is investigated and discussed based on the filler in matrix model6) 7) of starch gel.
Materials. Native cassava starch was purchased from Vedan Vietnam Enterprise Co., Ltd. (Dong Nai Province, Vietnam). α-amylase (4,940 U/g) from A. niger was provided by Shinnihon Chemicals Co., Ltd. (Anjo, Japan). α-amylase (10 U/g) from B. amyloliquefaciens was provided by Novozymes Japan Ltd. (Chiba, Japan). The α-amylase activity was measured by the α-amylase Measuring Kit (Kikko-man Biochemifa Co., Ltd., Tokyo, Japan) using N3-G5-β-CNP as substrate. One unit of α-amylase activity was defined as the amount of enzyme producing 1 μmol of CNP for one minute under the assay conditions.15) β-Amylase from barley (OPTIMALT BBA, 1,230 DP°/g) was purchased from Danisco Japan Ltd. Genencor Division (Tokyo, Japan) and used without any further treatment.
Production of enzyme-treated starch granules. Native cassava starch granules (300 g) were suspended in distilled water (700 g), regulated pH (pH 5.0) with 1N HCl and incubated with or without enzyme (3 g) for 23 h at 50°C with continuous agitation. To inactivate enzyme activity, 5 g of sodium hypochlorite solution (the effective chlorine concentration: 10% (w/w)) was added to the starch suspension and stirred. After 10 min, disodium pyrosulfite (0.25 g) was added for neutralization and stirred for 10 min. The enzyme-treated starch granules were precipitated by centrifugation (3,000 rpm for 5 min). The amount of solubilized carbohydrates from the starch granules was determined by measuring the sugar content of the supernatant fraction by the phenol-sulfuric acid method.16) The precipitate was re-suspended with distilled water then precipitated again by centrifugation. This washing process was repeated three times and the enzyme-treated starch granules were obtained by drying the washed pellets with a blow dryer.
Rheometric analysis of starch gel. The physical properties of the starch gel were analyzed using a rheometer (RT-2010J-CW, Rheotech K.K., Tokyo, Japan). A starch suspension containing 20% (w/w) starch granules in distilled water was prepared, incubated at 65°C for 10 min and placed in a polyvinylidene chloride tube. The tube was heated to 90°C at 1°C/min, held at 90°C for 30 min, and kept for 1 h at 25°C followed by 19 h at 5°C. The starch gel formed in the casing tube was incubated at 25°C for 4 h, sliced into gel disks 25 mm thick and loaded onto the stage of the rheometer. The rupture stress, compression distance and Young’s modulus were measured by the rheometer with an adapter (diameter 5 mm, area 19.635 mm2) using a movement speed of 6 cm/min.
Scanning electron microscope (SEM). α-AMA-treated-starch granules were coated with Pt-Au using a Model E-1010 ion sputter coater (Hitachi Co., Ltd., Tokyo, Japan). The coated samples were then analyzed using a SU1510 SEM (Hitachi Co., Ltd., Tokyo, Japan) at an operating voltage of 10 kV.
Amylose content. The amylose content of the starch was determined by the methods of Hovenkamp-Hermelink et al.17) 100 μL of starch solution in 90% (v/v) DMSO were mixed with 900 μL of distilled water and vortexed, then mixed with 20 mL of iodine reagent. The absorbances at 618 and 550 nm were measured using a spectrophotometer (U-3900H, Hitachi Co., Ltd., Tokyo, Japan). Amylose content was calculated using the following equation.
Amylose (%) = (3.5 － 5.1 × R)/(10.4 × R － 19.9)
where R = A618 nm/A550 nm
Iodine reagent was made daily from 0.5 mL of iodine stock solution (0.26 g of I2 and 2.6 g of KI in 10 mL of water) mixed with 0.5 mL of 1 N HCl and diluted to 130 mL with distilled water.
Analysis of the unit chain length distribution of starch. A starch sample (10 mg) was dissolved in 1,250 μL of 1 N NaOH and cooled to 5°C for 20 h. After cooling, 5 μL of 5 N HCl, 5 μL of 1 M sodium acetate (pH 5.5) and 65 μL of distilled water were added to 25 μL of sample solution, and 2 μL of 0.11 mg/mL Pseudomonas isoamylase in 50 mM sodium acetate (pH 5.5) was added. The samples were incubated at 37°C for 20 h, filtered through a 0.45 μm nylon filter and subjected to high-performance anion-exchange chromatography with pulsed amperometric detection (HPAEC-PAD) using a Dionex ICS-3000 (Dionex Corp., Sunnyvale, USA). A sample eluted with a gradient of sodium acetate (0–2 min, 50 mM; 2–37 min, increasing from 50 mM to 350 mM with the installed gradient program 3; 37–45 min, increasing from 350 to 850 mM with the installed gradient program 7; 45–47 min, 850 mM) in 150 mM NaOH with a flow rate of 1 mL min－1.
Analysis of hot water extracts. Each starch sample (200 mg) was dispersed in 20 mL of 0.1% AgNO3 and then incubated at 80°C for 30 min. AgNO3 was used to inhibit α-amylase activity.18) The starch suspension was then centrifuged at 85 × G for 3 min and the supernatant recovered for further analysis. A Two millilitter aliquot of supernatant was loaded onto a Sepharose CL-2B (GE Healthcare UK Ltd., Buckinghamshire, UK) column (diameter 2.5 cm, length 45 cm) that had been equilibrated with 50 mmol/L sodium hydroxide containing 0.02% sodium azide and eluted with the same eluent at a flow rate of about 120 mL h－1 at room temperature. Fractions were taken at 100 drop intervals and the total carbohydrate content in each fraction was measured by the phenolic sulfuric method. To obtain the structural information of the leached starch, the eluate in each fraction was mixed with iodine solution (6.0 mM KI, 0.4 mM I2) and subjected to absorption spectral analysis.
Enzyme treatment of cassava starch granules.
Native cassava starch granules suspended in distilled water were incubated with three enzymes (α-AMA, α-AMB and β-AMB), and the amount of carbohydrates solubilized into water fraction was monitored along with the reaction time (Fig. 1). α-AMB most efficiently hydrolyzed insoluble starch granules, liberating 8.2% (w/w) of soluble sugars in the initial 15 min and more than 30% (w/w) in the subsequent 23 h. α-AMA and β-AMB liberated only 2.8% (w/w) and 0.5% (w/w) of soluble sugars, respectively, in 23 h. This result clearly indicates that the extent of hydrolysis of insoluble cassava starch granules depends on the enzyme used. Insoluble starch granules treated with α-AMA or β-AMB for 23 h or with α-AMB for 15 min were prepared as described in Materials and Methods, and were subjected to further studies. The extent of hydrolysis of α-AMA-, α-AMB- and β-AMB-treated starch granules thus produced were 2.8, 8.2 and 0.5% (w/w), respectively.
Time course of the hydrolysis rate by the enzymatic treatement.
Treatment of cassava starch granules with α-amylase from A. niger (α-AMA, ●), α-amylase from B. amyloliquefaciens (α-AMB, ▲) and β-amylase from barley (β-AMB, □). Extent of hydrolysis was defined as percentage of solubilized carbohydrate against total carbohydrate in the reaction mixture.
Rheometric analysis of starch gel.
When gelatinized starch paste is left at a low temperature, retrogradation occurs and the starch paste forms a gel. Gel formation is an important property of starch, especially in the food industry since the strength and hardness of starch gel affect the texture of foods. The three types of enzyme-treated starch granules described above were gelatinized with water and converted into gel disks as described in Materials and Methods. These gel disks were subjected to rheometric analysis to measure rupture stress, compression distance and Young’s modulus. As shown in Table 1, the Young’s modulus of starch gels prepared from α-AMB- or β-AMB-treated starch granules were slightly smaller than those prepared from native cassava starch granules. These results suggest that α-AMB and β-AMB treatments slightly weaken the gelling properties of cassava starch granules. This observation was not surprising because hydrolysis of starch should have a negative effect on its gelling properties. On the other hand, the Young’s modulus of the gel produced from α-AMA-treated cassava starch granules was higher than that of the gel produced from native cassava starch granules. α-AMA treatment appears to significantly affect the gelling properties of cassava starch granules, producing a hard and elastic starch gel. This observation is surprising since it contradicts the generally held view that partially hydrolyzed starch shows decreased gel-forming ability. These rheometric experiments indicate that limited hydrolysis of cassava starch granules to release only 2.8% (w/w) of solid matter into solution has a big impact on their gelling properties. It should also be noted that this is not true for all enzymes tested but only for one specific enzyme, α-AMA.
Rheometric properties of starch gel prepared from native and enzyme-treated cassava starches.
Extent of hydrolysis of α-AMA-, α-AMB- and β-AMB-treated starch granules was 2.8, 8.2 and 0.5% (w/w), respectively. The value is the mean ± SD of at least three measurements.
Analysis of α-AMA-treated starch granules.
In order to understand the mechanism underlying the unexpected gelling properties of α-AMA-treated cassava starch granules, several structural analyses of α-AMA-treated and native starch granules were carried out. However, we could not observe any difference between the two starches in scanning electron microscopy analysis (Fig. 2A), amylose content (Table 2), unit chain distribution of amylopectin by HPAEC-PAD analysis (Fig. 2B) or amylograph analysis (data not shown). All these results indicate that α-AMA- treated and native starch granules are very similar in granule morphology, composition and structure of starch components and pasting properties, but different in gelling properties.
Analysis of native and α-AMA-treated cassava starch granules.
(A) Morphology of native (a, b) and α-AMA-treated (c, d) cassava starch granules as observed by scanning electron microscopy. Left panels (a, c) are × 1,000 and right panels (b, d) are × 4,500. (B) Debranched unit chains from native (white) and α-AMA-treated (black) cassava starch are compared by displaying the percentage of each peak area against the sum of peak area from DP 6 to 30.
Properties of native and α-AMA treated cassava starches.
One hundred microlitters of starch solution in 90% (v/v) DMSO were mixed with 900 μL of distilled water and vortexed, then mixed with 20 mL of iodine reagent. The absorbances at 618 and 550 nm were measured using a spectrophotometer. Each value is the mean ± SD (n = 3).
Analysis of hot water extracts. The filler in the matrix model of starch gel6) 19) may be important to understand the mechanism underlying the gelling properties of α-AMA-treated cassava starch granules. Based on this model, not filler (swelled starch granules) but matrix (leached material from starch granules) should be an important factor determining the properties of the gel. So we suspended α-AMA-treated and native starch granules in hot water and collected the leached starch component for analysis. Starch granules were suspended in water containing 0.1% (w/v) AgNO3 as an anti-microbial agent, and incubated at 80°C for 30 min. The leached starch component was collected as described in Materials and Methods and subjected to GPC analysis. The amount of carbohydrate and the λmax of the starch-iodine complex of each fraction were measured and summarized in Fig. 3.
Gel permeation chromatography of starch component leached from starch granule.
A starch (○, native cassava starch; ●, α-AMA-treated cassava starch) suspension in water was heated at 80°C for 30 min and the leached starch component was fractionated using a Sepharose CL-2B column as described in “Analysis of hot water extracts” in the Experimental section. Amount of carbohydrate in each fraction (A) is measured and displayed as a percentage against total amount of carbohydrate eluted from the column. λmax of the polysaccharide-iodine complex (B) were analyzed. The value is the mean of three independent experiments.
The leached starch components collected from native and α-AMA-treated starch granules showed wide molecular weight distribution as shown in Fig. 3A, and their elution profiles were similar. However, differences were found in the λmax values of fractions 25 to 28, where very high molecular weight starch components are eluted (Fig. 3B). The λmax values of fractions 25 to 28 for α-AMA-treated starch granules were all higher than 600 nm but those for native cassava starch granules were around 600 nm or below.
The λmax value of the amylose-iodine complex is known to be around 650 nm, whereas that of the amylopectin-iodine complex is much lower (around 550 nm). From these studies, we conclude that the leached starch components (matrix) collected from native cassava starch granules contain high molecular weight amylopectin which is absent or much less in α-AMA-treated starch granules. From these studies, we conclude that the leached starch components (matrix) is differ between native cassava starch granules and α-AMA-treated starch granules, where the former composed by amylose with wide molecular weight distribution and high molecular weight amylopectin, but the latter composed by amylose with much less amount of amylopectin or amylopectin with long side chains.
Enzymatic treatment of insoluble starch granules to alter the functionality of starch is an attractive area of research. In the present study, we find that α-AMA-treated cassava starch granules can produce gel with greater hardness and elasticity. This is surprising because α-AMA treatment liberates only 2.8% (w/w) of the insoluble starch component into the soluble fraction, but has a big impact on gelling properties. It should be emphasized that this finding contradicts the generally held view that partially hydrolyzed starch produces gel with decreased viscosity and elasticity.20) α-AMB- and β-AMB-treated starch granules obtained in this study both showed weakened gelling properties (Table 1), so the enhanced gelling properties of α-AMA-treated cassava starch granules may be specific to this particular enzyme. Further studies are necessary to address this issue.
In spite of many basic and applied studies of starch gel, our understanding of starch gel formation is still incomplete and may vary depending on the type of starch and the conditions of formation. It is known that starch gel is a composite of filler (swelled starch granules) in a matrix (leached starch component) as proposed by Rheol6) and subsequently confirmed by Goesaert.19) When starch granules suspended in water are heated to temperatures under 100°C, they swell and leach starch component (mainly amylose) resulting in the dispersion of swelled starch granules in water containing leached amylose. Subsequent cooling helps the leached amylose to form local intra-molecular cross-links and a continuous macro gel (matrix) which include swelled starch granules (fillers), as schematically shown in Fig. 4A.
Schematic illustrations of the gelling process of native (A) and α-AMA-treated (B) cassava starch granules.
If starch gel is like filler in a matrix model, the leached starch component should be an important factor determining the properties of the gel. We carefully collected the starch component leached from native and α-AMA-treated cassava starch granules and analyzed its amount and composition with GPC. Leached starch component from native cassava starch granules contained amylose with wide molecular weight distribution as major component but also contained high molecular weight amylopectin. However, leached starch component from α-AMA-treated starch granules contained amylose with much less amount of amylopectin or amylopectin with long side chains (Fig. 3B).
This suggests that the matrix of native starch gel is composed of amylose and amylopectin, while that of α-amylase-treated starch gel is composed of amylose with decreased level of amylopectin, as schematically summarized in Fig. 4B. We would like to conclude that the enhanced gelling properties of α-AMA-treated cassava starch are produced by the strong matrix structure made from amylose.
Amylose is the minor component of starch and its content in the cassava starch granule varies between 17.9-23.6%.21) Amylose is the predominant starch component leached from the starch granule.22) However, amylopectin is also leached as we observed in the native cassava starch granule (Fig. 3B). It is thus very interesting why amylopectin is not leached from α-AMA-treated cassava starch granules. We cannot find any significant difference in amylose content or amylopectin unit chain distribution profiles when whole starch is subjected to analysis (Fig. 2B, Table 2). This means that the difference produced by α-AMA treatment occurs in a limited part of the starch granule. In this particular case, the limited part should be the surface of the starch granule. The α-AMA may preferentially hydrolyze amylopectin molecules at the surface and increase the local (surface) amylose content. Alternatively, the α-AMA may alter the surface structure for selective leaching of amylose. However, our understanding of starch granules and the organization of amylose and amylopectin especially at the surface of the granule is not sufficient to address the mechanism of this interesting finding.
Our results suggest that enzymatic treatment of cassava starch granules can provide significant impact on the properties of the starch without changing the size or morphology of the starch granules. However, this may not the case in cereal starches. In the case of maize and rice, α-amylase treatment produces starch granules covered with holes.23) These holes go toward the center of the granule and gradually break it.24) Our α-AMA-treated cassava starch granules were smooth with no holes (Fig. 2A). This difference is of great interest and will be the subject of further studies that will provide a better understanding of the structure and organization of starch granules.
In conclusion, enzymatic treatment of insoluble starch granules at sub-gelatinization temperature is potentially a novel and powerful tool to alter the properties of starch. The efficacy and feasibility of this approach should be examined for different enzymes and sources of starch.
The authors thank Dr. Shiho Suzuki, Mr. Makoto Nakaya, Dr. Akiko Kubo and Mr. Kenichi Kurita for helpful guidance, and Mrs. Sanae Takahashi for technical assistance.
