Extraction and Characterization of Starch from Alkaline Cooked Corn Masa

Cereal Chern. 84(4):415-422 Starch granules undergo structural and morphological changes during food processing unit operations as they interact with other food ingredients. This study was conducted to isolate and characterize starch granules from com masa. A proteolytic enzyme, thermolysin, was effective in separating and isolating starch granules from endosperm proteins present in masa. The efficiency of starch extraction using thermolysin was 74% (w/w), and subsequent analyses showed that the isolated granules were Physiochemical properties of food starches have been studied extensively (Zobel 1984; Biliaderis 1998; Jacobs and Delcour 1998; Jenkins and Donald 1998; Sahai and Jackson 1999; Liu et al 2002). Starch in foods, exposed to specific hydrothermal conditions, would not necessarily undergo the same structural and morphological changes as would isolated starch (starch relatively free of proteins, lipids, etc.) exposed to the same treatment. Starch properties and food functionality depend on the type (or source) of starch, and food composition. Food components such as protein, fatty acids, and various solutes affect the thermal behavior of starch; this has been documented by various studies conducted using predetermined mixtures of starch and other ingredients (Larsson 1980; Lund and Lorenz 1984; Eliasson 1985; Lim et a12000; Bogracheva et al 2002; Gonera and Cornillon 2002; Gibbon et al 2003; Tolstoguzov 2003; Zhang and Hamaker 2003; Zhang et al 2003; Lindeboom et al 2004; Mondragon et al 2004; Kar et al 2005; Rumpold and Knorr 2005; Sayer et al 2005). Because of inherent difficulties associated with starch isolation from actual processed food products, knowledge of the structural changes that occur during hydrothermal treatments and starch-food component interactions is poorly understood. Alkaline cooking, which is referred to as nixtamalization, is an important process used in the preparation of tortillas, com chips, taco shells, tamales. and other Mexican-style foods. During nixtamalization, com is first cooked in the presence of lime, steeped, and then washed to produce nix tarnal. Nixtamal is stone-ground to form a soft, moist dough that is called mas a (Gomez et al 1987; Serna-Saldivar et al 1990). Although nixtamalization is widely used in the food industry, a comprehensive, fundamental understanding of starch functionality and its thermal behavior in masa is still lacking. The starch gelatinization process during mas a preparation has not been studied in detail, and only a few reports are available on the effect of masa components on starch functionality (Bryant and Hamaker 1997; Campus-Baypoli et al 1999). Starch in mas a has been characterized without extracting starch in purified form (Gomez et al 1991, 1992). Although these studies are helpful in identifying changes in starch granules during mas a 1 A contribution of the University of Nebraska Agricultural Research Division (Journal Series Number 15264), supported in part by funds provided through the Hatch Act. Names are necessary to report factually on available data; however, the USDA neither guarantees nor warrants the standard of the product, and the use of the name by the USDA implies no approval of the product to the exclusion of others that may also be suitable. 2 Department of Food Science and Technology, University of Nebraska-Lincoln, Lincoln, NE 68583-0919. 3 Corresponding author. Phone: 402-472-2814. Fax: 402-472-1693. E-mail address: [email protected] doi:10.1094/CCHEM-84-4-0415 This article is in the public domain and not copyrightable. It may be freely reprinted with customary crediting of the source. AACC International, Inc., 2007. free of contaminants. Starch samples were characterized using light microscopy, SEM, DSC, and XRD. Starch granules isolated from masa had undergone internal structural changes and some granules (",,40%) lost birefringence during nixtamalization. These internal changes occurred, in most cases, without visible alterations in general granular morphology. Nixtamalized granules underwent changes mostly consistent with a "heatmoisture treatment" process. production, the influence of the other components in mas a during starch analyses cannot be completely eliminated. There are no highly effective methods available to isolate starch from masa. Starch isolation is an important first analytical step in evaluating the structural changes starch granules undergo during nixtamalization. The objectives of this study were to isolate starch granules from freshly prepared mas a and to investigate morphological and structural changes of these extracted starch granules. MATERIALS AND METHODS A white food com (Hybrid Zimmerman 1851W, 2000 crop year) was obtained from Wilson Genetics LLC. (Harlan, IA) [Now available from Garst Seed Company, Slater, IAJ. The grain was stored at -16°C before use. Grain Composition Analysis Proximate composition of whole com was analyzed using the Official Methods 920.39, 935.29, and 942.05 (AOAC International 2004). Starch Extraction from Corn Kernels Starch from the food com was extracted using a laboratory wetmilling procedure: com was steeped at 51 ± 1°C for 48 hr with 0.47% (w/v) lactic acid and 0.15% (w/v) sodium bisulfite. This procedure is more fully outlined by Wehling et al (1993). In addition, a small amount of endosperm starch (raw starch) used for selected DSC experiments was physically scraped from com kernels that had been cracked open. Masa Preparation Nixtamal and mas a were produced using the method developed by Yglesias et al (2005), which mimics the industrial-scale mas a production process. Com was cooked for 30 min at 90°C and steeped for 9.5 hr. Starch in mas a was stabilized and preserved using a liquid nitrogen freeze-drying method previously documented to create minimal changes from the fresh state (Yglesias and Jackson 2005). Samples were stored in sealed polypropylene containers at -18°C until further analyses. Starch Extraction from Masa Starch from mas a was extracted using two different methods. For Method 1, freeze-dried masa was ground into flour using a cyclone sample mill (model 3010-030, Udy Co., Fort Collins, CO). Masa (10 g) was dispersed in 500 mL of distilled water by stirring using a magnetic stirrer for 6 hr. The mixture was passed through a loose cotton wool plug in a funnel and then it was filtered using a 60-f.!m polyester mesh screen (Spectrum Laboratories, Rancho Dominguez, CA) under suction to remove coarse particles. The filtrate containing starch was collected and washed with Vol. 84, No.4, 2007 415 distilled water containing 1 % (w/v) sodium hydroxide four times and then washed with distilled water several times. The recovered starch was dispersed in SOO mL of distilled water and filtered using IO-Ilm nylon mesh (Spectrum Laboratories) under suction to remove fine contaminants. Starch was recovered and freezedried (Virtis Sentry 8L freeze dryer, SP Industries, Gardiner, NY) at -SsoC and 200 mTorr pressure for 36 hr. Samples were stored in sealed polypropylene containers until used. For Method 2, dried masa was ground into flour using a cyclone sample mill (Udy model 3010-030). Starch in masa was extracted by the protease digestion method described by MuForster and Wasserman (1998), with slight modifications. Dry mas a flour (S g) was mixed with 330 units of thermolysin (from Bacillus thermoproteolyticus Rokko [EC 3.4.24.27], SigmaAldrich, Steinheim) in the presence of S mM calcium chloride in 100 mL of aqueous solution in a 12S-mL screw-cap Erlemmeyer flask. The mixture was kept in a 60°C water bath for 4 hr and gently hand-mixed at 30-min intervals. The enzyme action was terminated by adding 0.03 g of EDTA to the reaction mixture. After cooling to room temperature, the mixture was filtered through a loose cotton wool plug in a funnel to remove coarse particles and washed five times with distilled water to remove residual proteins. Starch was recovered by centrifuging the suspension at 1,300 x g for 7 min (model RC-3, Ivan Sorvall, Norwalk, CT) after each washing step. Recovered starch was redispersed in SOO mL of distilled water and filtered using a 60-llm polyester mesh screen (Spectrum Laboratories) under suction to remove coarse particles. The filtrate containing starch was collected and washed with distilled water containing 1 % (w/v) sodium hydroxide four times and then washed with distilled water several times. Recovered starch was dispersed in SOO mL of distilled water and filtered using a IO-Ilm nylon mesh screen (Spectrum Laboratories) under suction to remove fine contaminants and freeze-dried using a freeze dryer (Virtis Sentry 8L, SP industries, Gardiner, NY) at -SO°C and 100 mTorr vacuum pressure for 36 hr. Samples were stored in sealed polypropylene containers until used. Masa Protein Total protein contents (total nitrogen x 6.2S) of mas a and isolated starch samples were analyzed by the Dumas combustion procedure using a nitrogen deterrninator (FP-528, Leco Corporation, St. Joseph, MI). The equipment was calibrated using EDTA. Total Starch in Masa Masa samples were dispersed by adding 10 mL of 2N sodium hydroxide to 0.5 g of masa in 70-mL capacity test tubes and heated for 25 min at 94°C and adjusted to pH 4.S using 2N hydrochloric acid and 10 mL of acetate buffer (9.1 g of sodium acetate [Sigma Chemical Co., St. Louis, MO] and 44.6 mL of glacial acetic acid [Fisher Scientific, Pittsburg, PAl diluted to SOO mL). Amyloglucosidase [EC 3.2.1.3] (300 units, Sigma Chemical) was added and the slurry was incubated for 70 min at SO°C. The reaction was terminated by adding S mL of 25% (v/v) trichloroacetic acid (Fisher Scientific, Pittsburg, PA) and the volume was adjusted to 100 mL in a volumetric flask using phosphate buffer (40 g of anhydrous monobasic sodium phosphate, anhydrous [Fisher Scientific] and 10 g of anhydrous dibasic sodium phosphate [Sigma] diluted to 1 L). Glucose concentration was measured using a biochemistry analyzer (model 2700 Select, YSI, Yellow Springs, OH) fitted with an immobilized glucose oxidase membrane [EC 1.1.3.4] in the presence of phosphate buffer (YSI 2357 buffer concentrate kit). Percent total starch (dry basis) was calculated using this equation: Total starch (%) = {[Glucose in sample (gIL) Blank A Blank B] x 0.9 x 0.10 x 100 }/Sample weight (g), where Blank A is the amount of free glucose in sample (gIL) and Blank B is the enzyme blank (gIL). 416 CEREAL CHEMISTRY The blanks A and B were used to correct the sample reading for free glucose in the sample and the amyloglucosidase reagent, respectively. Conversion factors were used to convert glucose into starch (0.9) and convert units of volume (0.10). The analyzer was calibrated using a D-glucose standard (2.50 gIL, YSI 2776 solution). Starch Yield and Isolation Efficiency Starch yield was calculated as: Starch yield (%) = {[Weight of isolated starch (g) x 100% ]/Weight of the corn sample (g) } . The starch isolation efficiency was calculated as: Starch isolation efficiency = {[Weight of isolated starch (g) x 100% ]lTotal starch in corn sample (g) } Light Microscopy of Masa Masa samples (:::;0.10 g of sample in :::;15 mL of water) were heated to specific temperatures in a glass petri dish placed on a laboratory electric hot plate while monitoring the temperature with a noncontact infrared thermometer. The heated sample in the petri dish was promptly observed under a microscope (Provis AX70, Olympus America, Melville, NY) equipped with a 60x water immersion lens (600x magnification) and a camera (Optronics, Goleta, CA) attached to the eye piece. Digital images were captured using software (v.2.1C, MagnaFire, Optronics). Polarized Light Microscopy of Starch A small amount (:::;0.005 g) of starch was mounted on a glass slide with a few drops of water and covered with a cover slip. The mounted sample was then observed under polarized light (polarizers at 90° to light path) using a microscope (Optiphot, Nikon USA, Melville, NY) and a lens converter (model MD, Meiji Techno Co., Saitama, Japan) equipped with polarizing filters at the light source and lens piece (Fisher Scientific, Pittsburgh, PA). Images were captured using a Sony video imaging system (model VPC 920 adapter and PVM-1354Q monitor) and a Sony photo printer (Mavigraph UP-1200A). Scanning Electron Microscopy (SEM) Masa and isolated starch samples were mounted on metal stubs and coated with gold palladium (:::;20 nm thickness) using a Hummer sputter-coating system (Anatech Ltd., Union City, CA). Samples were then observed using a scanning electron microscope (S3000N, Hitachi Science Systems, Tokyo) at an acceleration potential of 15 kY. Pictures were captured using automated image capturing software (Hitachi High-Technologies, Pleasanton, CA). Differential Scanning Calorimetry (DSC) Starch and mas a samples (:::; 1 0 mg, db) with :::;55 ilL of distilled water (i.e., starch in excess water) were hermetically sealed in DSC pans (Perkin Elmer Pan Sell Kit 0319-15251152611535) and kept at room temperature for 2-3 hr. Next, samples were scanned against a blank (empty pan) using a Perkin Elmer Pyris 1 differential scanning calorimeter from 2S to 90°C at a 10°C/min scanning rate. Pyris v.3.52 software (Perkin Elmer) was used to collect onset (To), peak (Tp), and end (Tc) temperatures, and the transition enthalpy (/1.H). Equipment was calibrated using indium as the reference material. X-ray Diffraction (XRD) X-ray diffraction data were obtained using a Bruker-AXS D8 Discover XRD system (Bruker AXS Microanalysis GmbH, Berlin) with a general area detector diffraction system (GADDS). The system was equipped with a copper target X-ray tube set to 40 kV and 30 mA, a Gobel mirror, a 0.5-mm pinhole collimator, and a Bruker-AXS HI-Star area detector. The samples were mounted on an aluminum sample plate with a few drops of ethanol and compressed using a glass slide to obtain a smooth surface. The mounted samples were allowed to dry at room temperature for ",,20 min before scanning. Sample surfaces were aligned to the center of the X-ray beam using a laser/video microscope system. Data were collected in reflection mode using the scan conditions of Omega = 4 degrees, detector swing angle = 18 degrees, sample to detector distance = 9.75 cm, exposure time = 180 sec. Area detector data were processed using Bruker-AXS GADDS system software by integrating over 29 = 7 to 35 degrees and X = -130 to -50 degrees. The integration was conducted along the Debye rings resulting in a diffractogram of intensity versus 29. Percent relative crystallinity was calculated according to the method outlined by Nara et al (1978) using quartz as 100% crystalline material (Elias son et al 1987) and an 85°C treated sample of each starch as 100% amorphous material. Peak fitting software (v.6.0/SR-4, Microcal Software, Northhampton, MA) was used to calculate absolute differences between XRD profiles. The percentage relative crystallinity was calculated as % Relative crystallinity = [(L: lIs Ial / L: lIe Ial) X 100%] where lIs Ial = absolute difference between the sample [Is] and amorphous [Ia] intensities, and lIe Ial = absolute difference between the crystalline (quartz) [Iel and amorphous [Ia] intensities. Statistical Analyses All results are averages of at least three independent replicates. Mean separations were performed using the Tukey-Kramer HSD test using JMP v.5 .0.1.2 software (SAS Institute, Cary, NC). RESULTS AND DISCUSSION The proximate composition of the com used in this study is given in Table I. Several previous studies have shown that the proteolytic enzyme thermolysin can be successfully used to remove proteins associated with starch granules (Belles et al 2000; MezoVillanueva and Serna-Saldivar 2004; Han et al 2005). Accordingly, thermolysin digestion was used to remove starch granules from protein matrices in masa. Starch isolated by Method 2 (enzyme digestion) gave starch with significantly (P < 0.05) less protein compared with the control (Method O. The total protein contents were 4.7 and 0.74% (w/w) in starches isolated by Methods 1 and 2, respectively. Total protein in masa was 11.13% (w/w), which was slightly higher than published values of 8.4-9.5% adjusted for the 6.25 conversion factor (Bello-Perez et al 2003) and 8.8% (Y glesias et al 2005). According to these results, ;:::57.7% of protein in masa can be removed by water washing (Method 1), and the rest of the starch granule bound proteins (;:::35.6%%) are removed by thermolysin action (Method 2). A very small amount (;:::6.5%) of total protein in masa is left with purified starch granules after thermolysin treatment. In SEM images (Fig. 1), protein and endosperm remnants were clearly visible among starch granules isolated by Method 1, whereas digesting protein with thermolysin (Method 2) resulted in relatively "uncontaminated" starch granules. The starch yield of thermolysin extraction (weight of granules extracted from mas a) was 60.0% (w/w, SD 2%). The starch extraction efficiency of the method was 74% (w/w, db) based on amount of starch in masa.