December 6, 2006

Relationships Between the Microstructure, Physical Features, and Chemical Composition of Different Maize Accessions From Latin America

By Narvez-Gonzlez, Ernesto David; de Dios Figueroa-Crdenas, Juan; Taba, Suketoshi; Tostado, Eduardo Castao; Et al


Chemical composition (moisture, total lipids, protein, and apparent amylose) and some physical features (1,000 kernel weight, hardness, and anatomical composition) were determined in 71 accessions representing races of maize from Latin America. Their microstructural characteristics (size and compaction of endosperm cell bodies, pericarp thickness, hornyfloury endosperm ratio, and morphology and size of starch granules) were also evaluated using environmental scanning electron microscopy (ESEM). Compaction was the most important microstructural feature of the maize kernels, representing kernel hardness. Highly compact kernels tended to be hard, with high protein, pericarp, and hard-endosperm content and high pericarp thickness, but with low moisture, amylose content, and kernel weight and size. The opposite was observed in the least compact kernels. Highly compact kernels tended to have small, polygonal starch granules (10 m). These results suggest that microstructure is responsible for the physical features of maize kernels and that microstructure is related to chemical composition.

Cereal Chem. 83(6):595-604

The maize (Zea mays L.) kernel is composed of three main structural parts: pericarp, germ, and endosperm (Wolf et al 1952a). The endosperm is divided into two zones. The first zone is the hard or vitreous starch; the second zone is the soft or floury starch (Hoseney 1998). The properties of the starch granules in the two types of endosperm are quite different, and hard endosperm is believed to contain compounds with a higher molecular weight than those contained in soft endosperm (Cagampang and Kirleis 1985). Knowledge of the anatomical proportions is often used to determine the end use of the kernel. The dry-milling industry removes pericarp and germ from endosperm because those fractions are undesirable for their purposes (Lin et al 2002). To determine the kernel's end use, the industry considers color, size, and hardness as quality attributes of kernels; the last two properties are related to endosperm type (Figueroa et al 2005). Corn kernel hardness has been studied intensely because of its importance for industrial grain processes. Although there are some conflicting results, it is generally believed that chemical composition can influence hardness, which in turn affects functionality. Researchers have studied the relationship between hardness and chemical characteristics for amylose (Dombrink-Kurtzman and Knutson 1997; Robutti et al 2000); type and amount of zein (Lopes and Larkins 1991; Dombrink-Kurtzman and Bietz 1993; Moro et al 1995; Robutti et al 1997); and protein (Robutti et al 2000). However, hardness may be related not only to the chemical characteristics of the grain but also to its physical characteristics. Both the chemical makeup of the corn kernel as well as the way in which the starch granules are assembled in specific microstructure should be considered.

Currently, scanning electron microscopy (SEM) and image analysis techniques are commonly used to evaluate and characterize the internal structure of kernels. Many authors have used SEM to describe starch granule morphology and the structural arrangement of the endosperm and to analyze the effect of certain treatments on starch granules (Fannon et al 1992; McDonough et al 1997; Raeker et al 1998; Sarpistein and Kohler 1999; Lauro et al 2000; McPherson and Jane 2000; Perera et al 2001; Naito et al 2004; Wilson and Betchel 2004). The aim of this study has been to evaluate the variability of microstructure among native maize races from Latin America and to relate this variation to physical features and chemical composition that may affect end use quality.



Samples of 71 accessions representing races of maize were provided by the International Maize and Wheat Improvement Center (CIMMYT, Mexico) for this study. Those selected races represent worldwide diversity from 27,000 accessions of maize from Mexico, the Caribbean, Central, and South America (Table I).

Chemical Composition

Protein (Approved Method 46-13, AACC 2000), moisture (Approved Method 44-19, AACC 2000), total fat (Method 7.062, AOAC 1984), and apparent amylose (Morrison and Laignelet 1983) content were determined.

Physical Features

The weight of 1,000 kernels was determined for each sample (Mauricio et al 2004). Flotation index is determined by placing 100 kernels in a beaker containing 300 mL of NaNO^sub 3^ solution (1.250 g/mL). They were stirred to separate the kernels and left standing for 1 min. The number of floating kernels indicated the flotation index (Norma Oficial Mexicana 2002). The test was performed in duplicate.

Kernel hardness (N) was measured using a TA-XT2 texture analyzer (Texture Technologies Corp., Stable Micro Systems; Surrey, England) equipped with a 30 conical probe. Each one of 10 kernels was punctured at the opposite side of germ in the central part of the endosperm at velocity of 2 mm/sec and penetration depth of 2 mm (Mauricio et al 2004). The kernel composition was determined by hand- dissection. Five kernels were soaked in 10 mL of distilled water at 90C for 1 hr, and tip cap, pericarp, and finally the germ were removed from the kernel endosperm. Kernel fractions were dried for 12 hr in an oven at 100C. For determination of endosperm type, three corn kernels of each race were cut at the median longitudinal section of the kernel perpendicular to the face (Wolf et al 1952a) using a precision table cutting machine (Accutom-5 Struers; Metalinspec, S.A. de C.V. Parque Industrial Jurica, Quertaro, Mexico) with a diamond blade disk for precise and deformation-free cutting. The proportion of hard and soft endosperm in the total area was determined from pictures taken with a camera installed on an optical stereo microscope (Olympus SZ61 ; Olympus Imaging Europe GmbH) and processed using IMAGE-J software.


Passport Data of Races of Maize from Mexico and Latin Americaa

To study the starch granule morphology, an environmental scanning electron microscope (ESEM; Philips model XL30) with a beam of 15-20 kV (50 A) and GSE detector was used. The images were taken at 200, 500, and 3,500, 1.1 mBar, and a spot size of 4.5 to measure pericarp, packed cell, and starch granule size, respectively. Values from 1 to 4 were assigned to describe the degree of compaction (nondimensional) of the cell structures of the endosperm according to the characteristics observed at 500 (Fig. 1).

Three corn kernels of each race were split at the median longitudinal section of the kernel perpendicular to the face (Wolf et al 1952a) using a chisel mounted on the TA-XT2 texture analyzer and allowed to penetrate only 2 mm to break the kernel into two parts. One of the parts was selected and cut again with a diamond blade disk of a deformation-free cutting machine (Acuttom-5 Struers) at the opposite side of the investigating zone to get 4 mm thickness in order to mount the sample on glue tabs attached directly to aluminum stubs for observation.

The length, width, and thickness of endosperm cell bodies, pericarp thickness, and starch granule size were also measured in at least four photographs each for different areas of the broken kernel. The pictures were digitalized and processed with an image analysis software (IMAGE-J) program (Figs. 2 and 3).

Fig. 1. Cell bodies according to degree of compaction. A, Spherical starch, high intergranular spaces and nonvisible protein matrix; B, spheroid starch, intermediate intergranular spaces and nonvisible protein matrix; C, polygonal starch, without intergranular spaces and thin protein matrix; D, polygonal starch surrounded by a highly dense protein matrix.

Statistical Analysis

Simple Pearson correlation matrix and principal component analysis (PCA) (Manly 2000) were performed using PROC CORR and PROC PRINCOMP (SAS Institute, Gary, NC).


Physical Features

Larger and heavier kernels such as those of Hualtaco, Chillo, Huillcaparu, and San Jernimo tended to have low hardness values (r = -0.23, P ≤ 0.05) and low pericarp content (r = -0.51, P

Chemical Composition

Protein content ranged from 6.8% (San Jernimo and Chillo) to 14.2% (Popcorn and Curagua). Protein content was correlated with kernel hardness, measured by the flotation index (r = -0.30, P ≤ 0.05). Total lipid content range was 3.8-8.4% db and correlated with germ percentage (r = 0.43, P ≤ 0.0001), which coincides with Finnie and Svensson (2003). Also, some accessions such as Arrocillo, Palomero, Reventador, and Olotillo contained [asymptotically =]8% moisture, while others such as Cuzco Cristalino, Cacahuacintle, Avat Morot, and Diente de Caballo had [asymptotically =]12%. Moisture content correlated to hardness (r = - 0.42, P ≤ 0.0001), endosperm percentage (r = 0.25, P ≤ 0.05), the 1,000 kernel weight (r = 0.25, P ≤ 0.05), and soft endosperm percen\tage (r = 0.35, P ≤ 0.005).

Fig. 2. Pericarp micrographs at 200 for popcorn (A) and Cacahuacintle (B).

Fig. 3. Endosperm cell body micrographs at 500 for Elotes Occidentales-N (A) and Cristalino de Chihuahua (B).


Correlation Coefficients Among Physical, Chemical, and Microstructural Parameters of Maize Races of Latin America (n = 71)


Physical and Chemical Features of Races of Maize from Mexico and Latin America


Physical and Chemical Features of Races of Maize from Mexico and Latin America


Microstructural Features of Races of Maize from Mexico and Latin America


Microstructural Features of Races of Maize from Mexico and Latin America

These correlations suggest that water is mainly stored in the endosperm, which greatly influences kernel hardness. It is also possible that water activity in the endosperm is related to starch granule crystallinity (Tang et al 2000). In addition, apparent amylose ranged from [asymptotically =]1% in Reventador-S, Arrocillo Amarillo, and Cuban Flint with horny or hard endosperm to [asymptotically =]12.5% in Avat Morot, San Jernimo Huancavelinano, and Hualtaco with floury or soft endosperm (Table III). Amylose influences pasting and thermal behavior and, thus, cereal functionality (Sasaki et al 2000; Lin et al 2002).

Apparent amylose correlated to soft endosperm percentage (r = 0.38, P ≤, 0.0005), which explains its relationship to hardness (r = -0.30, P ≤ 0.05). Peng et al (1999) observed two types of starch granules in wheat: A-type (>10 m) and small B-type (

Pericarp Thickness

Pericarp thickness ranged from 90 m (Popcorn, Cnico, Dulce, Arrocillo, and Nal-Tel) to [asymptotically =]30 m (Cacahuacintle and Shima) (Fig. 2). Pericarp thickness and pericarp content were correlated (r = 0.39, P ≤ 0.0005). There was no evidence of any correlation between pericarp thickness and hardness but it was noted that kernels with high endosperm content tended to have thin pericarp (r = -0.31, P ≤ 0.05) (Tables II and III).

Endosperm Cell Bodies

Endosperm cell bodies ranged from 2,600 m^sup 2^ (Cristalino de Chihuahua) to 17,700 m^sup 2^ in terms of surface area (Elotes Occidentales) (Fig. 3, Table IV). Kernels with high endosperm content tended to have large cell bodies (r = 0.27, P ≤ 0.05). Cell body size decreased when pericarp content increased (r = - 0.23, P ≤ 0.05) and protein content increased (r = -0.24, P ≤ 0.05). The length of endosperm cells range is 70-260 m and the width of endosperm cells range is 99-25 m (Table IV). Similar ranges have been reported in corn by Wolf et al (1952b).


Compaction correlated to hardness value (r = 0.73, P ≤ 0.0001) (Fig. 4), the 1,000 kernel weight (r = -0.43, P ≤ 0.0005), pericarp (r = 0.23, P ≤ 0.05), soft-endosperm (r = - 0.69, P ≤ 0.0001), moisture (r = -0.34, P ≤ 0.005), protein (r = 0.25, P ≤ 0.05), and amylose (r = -0.33, P ≤ 0.005) content (Table IV). Empty spaces in cell bodies permit the conservation of the spherical form of starch granules.

Furthermore, low empty spaces in cell bodies promote the deformation of starch granules until they adopt a polygonal form (Hoseney 1998). The protein matrix is located precisely in the hard endosperm (Tester et al 2004). The protein matrix serves as supporting material between starch granules, giving more rigidity to the endosperm. Similar results were found in rice where the chalky portion of the kernels consists of layers of loosely packed, spherical starch granules, with air spaces between them, as opposed to the translucent portion that consists of polygonal, densely packed cells (Lisle et al 2000).


PCA Characteristic Vectors of Physical Features, Chemical Composition, and Compaction Grade of Samples

Fig. 4. Relationship between degree of compaction and kernel hardness (n = 71, R^sup 2^ = 0.54, P

Fig. 5. Starch granule micrographs at 3,500 for soft-endosperm (A) and hard-endosperm starch (B).

Starch granules in hard endosperm were polygonal, while in soft endosperm they were almost spherical (Fig. 5), which agrees with Hoseney (1998) and Fannon et al (1992). The adhesion force between protein and starch is strong enough to pull starch granules closer and closer together. They are tightly packed and become polygonal in shape (Robutti et al 1974; Hoseney 1998).

Starch Granule Size and Morphology

Starch granule size was 8-16 m; in diameter. Almost 69% of the samples showed starch granules of 8-12 m; 28% showed granules of 12- 16 m; and only 3% showed granules

Fig. 6. Grouping of races according to physical features, chemical composition, and degree of compaction. PC1 = Principal component 1, PC2 = principal component 2; 1, 2, 3 = Least low- intermediate, and highintermediate compact level, respectively; 4 = most compact level. Other abbreviations listed in Table II.

Compaction, Chemical Composition, and Physical Features

Principal component analysis (PCA) was performed to explore the variability of samples according to the degree of compaction, chemical composition, and physical features. The first two principal components explained 57% of the total variability accounted for by the original traits (Table V). The first principal component (PCl) on the positive side is associated with the apparent amylose content, soft-endosperm percentage, flotation index, and 1,000 kernel weight. Associated with the negative side is hardness, degree of compaction, and pericarp content. This component can be considered a dimension of hardness. At the same time, the second principal component (PC2) was characterized by a contrast of endosperm percentage and germ and lipid content. Kernels with the least degree of compaction, identified with the number 1, are mainly located on the positive side of PCl in Fig. 6. They were characterized by high moisture and apparent amylose content, heavy kernels, high flotation index, and high softendosperm percentage. On the other hand, highly compact kernels identified with the number 4 were located on the negative side of PCl with high hardness, pericarp, and hard endosperm percentages. Kernels with an intermediate degree of compaction, identified with the numbers 2 and 3, are dispersed in a middle-hardness zone, showing characteristics that are intermediate between the two above-mentioned groups but separated by their contrast in endosperm and germ contents.


Physical features of kernels are an expression of their microstructure, which is closely related to their chemical composition. Compaction of endosperm cell bodies was the most important microstructural feature related to the physical characteristics of the kernel. Highly compact kernels tended to be small with low weight and with low moisture and amylose content, but with high hard-endosperm percentage, protein content, and pericarp thickness. The opposite was found for the least compact kernels.


This work was supported by the grant QRO-2004-C01-38 of the CONACyT-Queretaro State Government. E. D. Narvez-Gonzlez thanks the secretaria de Relaciones Exteriores Mexico for the MC scholarship. We thank CIMMYT for supplying the samples. We thank Jos Eleazar Urbina, Marcela Gaytn, and Jos Juan Vles from Cinvestav-Quertaro for their technical support.


AACC International. 2000. Approved Methods of the American Association of Cereal Chemists, 10th Ed. Methods 44-19 and 46-13. The Association: St. Paul, MN.

AOAC. 1984. Official Methods of Analysis of the Association of Official Analytical Chemists. 14th Ed. Method 7.062. The Association: Washington, DC.

Cagampang, G. B., and Kirleis, A. W. 1985. Properties of starches isolated from sorghum floury and corneus endosperm. Starch 37:253- 257.

Campell, M. R., Pollak, L. M., and White, P. J. 1995. Genetic variation for starch thermal and functional properties among nonmutant maize inbreds. Cereal Chem. 72:281-286.

Dombrink-Kurtzman, M. A., and Bietz, J. A. 1993. Zein composition in hard and soft endosperm of maize. Cereal Chem. 70:105-108.

Dombrink-Kurtzman, M. A., and Knutson, C. A. 1997. A study of maize endosperm hardness in relati\on to amylose content and susceptibility to damage. Cereal Chem. 74:776-780.

Fannon, J. E., Hauber, R. J., and BeMiller, J. N. 1992. Surface pores of starch granules. Cereal Chem. 69:284-288.

Figueroa, C. J. D., Mauricio, A., Taba, S., Morales, E., Mendoza, A., Gaytn, M., Rincn-Snches, F., Reyes, M. L., and Vles, J. J. 2005. Kernel characteristics and tortilla making quality of maize accesions from Mexico, the Caribbean, and South and Central America. In: Latin American Maize Germplasm Conservation: Regeneration, In Situ Conservation, Core Subsets, and Prebreeding. S. Taba, ed. CIMMYT: Mexico, D.F.

Finnic, C., and Svensson, B. 2003. Feasibility study of a tissue- specific approach to barley proteome analysis: Aleurone layer, endosperm, embryo and single seeds. J. Cereal Sci. 38:217-227.

Hoseney, R. C. 1998. Principles of Cereal Science and Technology. 2nd Ed. AACC International: St. Paul, MN.

Lauro, M., Poutanen, K., and Forsell, P. 2000. Effect of partial gelatinization and lipid addition on α-amylolisis of barley starch granules. Cereal Chem. 77:595-601.

Lin, Y. P., Aboubacar, A., Zehr, B. E., and Hamaker, B. R. 2002. Corn dry-milled grit and flour fractions exhibit differences in amylopectin fine structure and gel texture. Cereal Chem. 79:354- 358.

Lisle, A. J., Martin, M., and Fitzgerald, M. A. 2000. Chalky and translucent rice grains differ in starch composition and structure, and cooking properties. Cereal Chem. 77:627-632.

Lopes, M. A., and Larkins, B. A. 1991 Gamma-zein content is related to endosperm modification in quality protein maize. Crop Sci. 31:16551662.

Manly, B. F. J. 2000. Multivariate Statistical Methods. A Primer. 2nd Ed. Chapman and Hall: New Zeland.

Mauricio, R. A. S., Figueroa, J. D. C., Taba, S., Reyes, M. L. V., Rincn, F. S., and Mendoza, A. G. 2004. Characterization of maize accessions by grain and tortilla quality traits. Rev. Fitotec. Mex. 27:213-222.

McDonough, C. M., Anderson, B. J., and Rooney, L. W. 1997. Structural characteristics of steam-flaked sorghum. Cereal Chem. 74:542-547.

McPherson, A. E., and Jane, J. 2000. Extrusion of cross-linked hydroxypropylated com starches II. Morphological and molecular characterization. Cereal Chem. 77:326-332.

Morrison, W., and Laignelet, B. 1983. An improved procedure for determining apparent and total amylose in cereal and other starches. J. Cereal Sci. 1:9-20.

Moro, G. L., Lopes, M. A., Harbben, J. E., Hamaker, B. R., and Larkins, B. A. 1995. Phenotypic effects of opaque2 modifier genes in normal maize endosperm. Cereal Chem. 72:94-99.

Norma Oficial Mexicana. 2002. NMX-FF034-2002-SCFI. Productos alimenticios no industrializados para consumo humano -cereales- Parte I. Maz blanco para proceso alcalino para tortillas de maz y productos de maz nixtamalizado. Especificaciones y mtodos de prueba. Diario Oficial de la Federacin, 22 de mayo de 2002.

Naito, S., Fukami, S., Mizokami, Y., Ishida, N., Takano, H., Koizumi, M., and Kano, H. 2004. Effect of freeze-thaw cycles on the gluten fibrils and crumb grain structures of breads made from frozen doughs. Cereal Chem. 81:80-86.

Peng, M., Gao, M., Abdel-Aal, E.-S. A., Hucl, P., and Chibbar, R. N. 1999. Separation and characterization of A- and B-type starch granules in wheat endosperm. Cereal Chem. 76:375-379.

Perera, C., Lu, Z., Sell, J., and Jane, J. 2001. Comparison of physicochemical properties and structures of sugary-2 cornstarch with normal and waxy cultivars. Cereal Chem. 78:249-256.

Raeker, M. ., Gaines, C. S., Finney, P. L., and Donelson, T. 1998. Granule size distribution and chemical composition of starches from 12 soft wheat cultivars. Cereal Chem. 75:721-728.

Robutti, J. L., Hoseney, R. C., and Wassom, C. E. 1974. Modified opaque-2 corn endosperms. II. Structure viewed with scanning electron microscope. Cereal Chem. 51:173-180.

Robutti, J. L., Borras, F. S., and Eyherabide, G. H. 1997. Zein composition of mechanically separated coarse and fine portions of maize kernels. Cereal Chem. 74:75-78.

Robutti, J., Borras, F., Ferrer, M., Percibaldi, M., and Knutson, C. A. 2000. Evaluation of quality factors in Argentine maize races. Cereal Chem. 77:24-26.

Sarpistein, H. D., and Kohler, J. M. 1999. Effects of sampling and wheat grade on precision and accuracy of kernel features determined by digital image analysis. Cereal Chem. 76:110-115.

Sasaki, T., Yasui, T., and Matsuki, J. 2000. Effect of amylose content on gelatinization, retrogradation, and pasting properties of starches from waxy and nonwaxy wheat and their F^sub 1^ seeds. Cereal Chem. 77:58-63.

Seetharaman, K., Tziotis, A, Boiras, F., White, P. J., Ferrer, M., and Robutti, J. 2001. Thermal and functional characterization of starch from Argentinean corn. Cereal Chem. 78:379-386.

Tang, H., Ando, H., Watanabe, K., Takeda, Y., and Mitsunaga, T. 2000. Some physicochemical properties of small-, medium-, and large- granule starches in fractions of waxy barley grain. Cereal Chem. 77:27-31.

Tester, R. F., Karlakas, J., and Qi, X. 2004. Starch- Composition, fine structure and architecture. J. Cereal Sci. 39:151- 165.

Wilson, J. D., and Betchel, D. B. 2004. Measuring wheat starch size distribution using analysis and laser diffraction technology. Published online at abstracts/, 2004. AACC International: St. Paul, MN.

Wolf, M. J., Buzan, C. L, MacMasters, M. M., and Rist, C. E. 1952a. Structure of the mature corn kernel. I. Gross anatomy and structural relationships. Cereal Chem. 29:321-333.

Wolf, M. J., Buzan, C. L, MacMasters, M. M., and Rist, C. E. 1952b. Structure of the mature corn kernel. HI. Microscopic structure of the endosperm of dent corn. Cereal Chem. 29:349-361.

[Received January 9,2006. Accepted July 13, 2006.]

Ernesto David Narvez-Gonzlez,1 Juan de Dios Figueroa-Crdenas,2,3 Suketoshi Taba,4 Eduardo Castao Tostado,1 Ramn lvar Martnez Peniche,1 and Froyln Rincn Snchez5

1 Universidad Autnoma de Quertaro, Centra Universitario, Cerro de las Campanas S/N, Quertaro, Quertaro, Mxico, CP 76010.

2 Centra de Investigaciones y Estudios Avanzados del IPN, Unidad Quertaro, Libramiento Norponiente No. 2000 Fraccionamiento Real de Juriquilla, Quertaro, Mxico, CP 76230.

3 Corresponding author. E-mail: [email protected]

4 Centra Internacional de Mejoramiento de Maz y Trigo, Carretera Mxico-Veracruz km 45 El Batn, Texcoco, Estado de Mxico, Mxico, CP 56130.

5 Universidad Autnoma Agraria Antonio Narro, Buenavista, Saltillo, Coahuila, Mxico CP 25315.

DOI: 10.1094/CC-83-0595

2006 AACC International, Inc.

Copyright American Association of Cereal Chemists Nov/Dec 2006

(c) 2006 Cereal Chemistry. Provided by ProQuest Information and Learning. All rights Reserved.