sign in sign up
<<
Home , Dietary fiber

Food Sci. Technol. Res., 8 (1), 80–84, 2002

Food Components in Fractions of

Hitomi ANDO,1 Yi-Chun CHEN,2 Hanjun TANG,2 Mayumi SHIMIZU,2 Katsumi WATANABE2* and Toshio MITSUNAGA2

1Kyoto Bunkyo Junior College, Kyoto 611-0041 Japan 2Department of Food and , Faculty of Agriculture, Kinki University, Nara 631-0052 Japan

Received September 21, 2001; Accepted December 13, 2001

Whole quinoa grain was separated into and milled grain, and the milled grain into perisperm and embryo. The proximate composition of the milled grain was similar to that of . The protein and lipid content of the embryo was 57% of total protein and 49% of total lipid, respectively. analysis showed that the quinoa grain was rich in K, Mg, Ca, P and Fe. The perisperm contained large oval aggregates 20–30 m in diameter and polygonal granules around 1 m in diameter. Differential scanning calorimetry data indicated a gelatinization tem- perature of 54.0 to 71.0˚C and enthalpy of 11.0 J/g starch. The water-soluble protein and NaCl-soluble protein frac- tions composed 28.7–36.2% and 28.9–32.9% of total protein in each fraction. Unsaturated accounted for 87.2–87.8% of total fatty acid. Phytate, a trypsin inhibitor activity and lipoxygenase activity in the embryo were high- est. The content of the bran was 86% of total saponin.

Keywords: quinoa seed, proximate composition, mineral, starch, protein, lipid,

Quinoa (Chenopodium quinoa Willd) is a traditional food crop mercial suppliers. in several South American countries which has been attracting Chemical analysis Moisture, ash, protein and lipid con- attention because of the nutritional value of its protein (Ranhotra tents were determined by the AOAC method (Horwiz,1970). et al., 1993). Quinoa grain is a disc-shape, about 2 mm in diame- Dietary fiber was measured by the enzyme-weight method ter and 0.5 mm in thickness. The major anatomical parts of the (Prosky et al., 1988). content (%) was calculated by sub- grain are the pericarp, the perisperm and the embryo. The peri- tracting moisture, ash, protein, lipid and dietary fiber contents carp (bran) usually contains , which are bitter antinutri- from 100%. Each mineral, with the exception of , in tional compounds (Mastebroek et al., 2000). Therefore, milled the sample solutions prepared by the HCl-extract method (Yasui grain with the pericarp removed is used as food. The embryo of et al., 1985) was measured with an atomic absorption spectro- the milled grain wraps around the perisperm like a headband. photometer (HITACHI Z-6100). Phosphorus was determined The high nutritional value of quinoa grain is due mainly to its colorimetrically with ammonium molybdate and ammonium high content of good quality protein (Mahoney et al., 1975, vanadate (Yasui, 1996). Phytate phosphorus and total saponin Gross et al., 1989, Ranhotra et al., 1993). Studies of quinoa lipid were determined by the colorimetric method (Huang & Lantz- (Koziol, 1992), starch (Atwell et al., 1983, Lorenz, 1990), miner- sch, 1983) and the gravimetic method (Lalitha et al., 1987), al (Bruin, 1964) and antinutritional compounds (Chauhuhan et respectively. Trypsin inhibitor activity was assayed (Mitsunaga, al., 1992, Mastebroek et al., 2000) have been carried out. How- 1979) using -N-benzoyl–D, L--p-nitroanilide as sub- ever, the food components of quinoa grain fractions have not strate. Lipoxygenase activity was measured by the conjugated been investigated. diene method (Takamura & Matoba, 1992). The objectives of this study were to characterize the distribu- Preparation and properties of starch granules Starch tion of food components in quinoa grain fractions and evaluate granules were prepared from the perisperm by a modified alkali the food value of this grain. method (Tang et al., 2000). The distribution of granule size was examined with a particle size analyzer (HORIBA Ltd, LA-700 Materials and Methods type). The shape and size of the granules was observed by scan- Materials Quinoa grain (Chenopodium quinoa Willd cul- ning electron microscopy (NIPPON DATAM, JSM-5400 LV). tivar Real TKW 2.8 g) grown in Bolivia in 1998 was used. X-ray diffraction was performed on the starch (10% moisture) Whole grain (Fig. 1A) was polished to separate the milled grain with an X-ray diffractometer (RIGAKU, Ltd, Rint-2000 type). (Fig.1B) and the bran (pericarp) with a milling machine Differential scanning calorimetry (DSC) was performed with a (SHIN-NAKANO KOGYO Ltd. MINI RICE POLISHING RP- starch to water ratio of 5 mg to 20 l (RIGAKU Ltd., DSC- 5). The milled grain was then divided into the perisperm (Fig. 8240D). absorption spectra and -amylolysis limit were 1C) and the embryo with this machine and sieves. Whole grain, measured following the methods of Takeda et al. (1983). milled grain, bran, perisperm and embryo were ground with a Determination of the amount of protein in each fraction mill for use as samples. All chemicals were purchased from com- Protein was extracted with H2O, 0.5 M NaCl, 60% n-propanol and 1% lactic acid solutions, successively, by the modified Maes method (Mitsunaga & Mitsuda, 1975). Insoluble protein was cal- *To whom correspondence should be addressed. E-mail: [email protected] culated by subtracting soluble protein from total protein. Food Components in Quinoa Seed 81

Fig. 1. Scanning electron microscopy of quinoa grains and . A, whole grain; B, milled grain; C, perisperm; D, cross section of perisperm; E, starch aggregate; F, starch granule.

Lipid analysis The lipid in the quinoa fraction was ex- with 10% methanolic HCl and subjected to gas-liquid chroma- tracted with chloroform/methanol/water according to the proce- tography (Shirasaka et al., 1998). dure of Bligh and Dyer (1959). The lipid was transmethylated 82 H. ANDO et al.

Results and Discussion no difference among fractions. The proximate composition of Chemical components in each fraction During the milling quinoa grain fractions on the basis of dry weight is shown in of whole quinoa grains, the bran and embryo were separated Table 1. The protein, lipid, sugar, dietary fiber and ash content of from the perisperm portion, and 8.2% bran, 30.1% embryo and whole quinoa grain were 12.9, 6.5, 63.7, 13.9 and 3.0%, respec- 58.8% perisperm were obtained on a grain-weight basis with tively. The milling of the grain had little effect on the content of losses of 2.9%. each component. The protein, lipid, sugar, dietary fiber and ash The moisture content in each fraction was 10.0–11.9% with content of the embryo, however, were 23.5, 10.2, 43.1, 18.9 and 4.3%, respectively. The protein and lipid content were 57% of total protein and 49% of total lipid in whole grain, compared to 39% and 46% in the perisperm. The levels of protein (7.2%), lip- id (5.0%), dietary fiber (8.5%) and ash (1.1%) in the perisperm were considerably lower than in the corresponding whole grain. The sugar content of the perisperm was 78.2%, compared to 63.7% for whole grain.

Table 3. Characterization of quinoa starch. Analysis Range X-ray diffraction pattern A-type Gelatinizationa) (˚C) To 54.0±0.6 Tp 62.2±0.3 Tc 71.0±0.7 Enthalpy(H, J/g) 11.0±0.5 Iodine absorption spectrab) max (nm) 602±2.1 Blue value 0.287±0.012 Apparent contentc) (%) 23.9±1.0 -Amylolysis limit (%) 59±1.5 All values are the mean±SD of three separate measuresments. a)To,Tp,and Tc are onset, peak and conclusion temperature, respectively. b)max=peak absorbance value over the range of wavelengths examined. Fig. 2. Distribution of particle size of quinoa starch granules. Y axis blue value(BV)=absorbance at 680 nm. shows volume % of starch particles. X axis shows particle size (m) with c)Apparent content(%)=[BV(starch)/BV(amylase)]100, assuming logarithmic scale. the amylose BV to be 1.2 (Takeda et al., 1983).

Table 1. Proximate composition of quinoa grain fractions. Whole grain Milled grain Bran Perisperm Embryo Proteina) 12.9±0.1 (100) 13.3±0.1 (96) 6.1±0.1 (4) 7.2±0.2 (39) 23.5±0.1 (57) Lipid 6.5±0.2 (100) 6.7±0.2 (95) 3.9±0.1 (5) 5.0±0.2 (46) 10.2±0.2 (49) Sugarb) 63.7 (100) 64.6 (93) 54.2 (7) 78.2 (73) 43.1 (20) Dietary fiber 13.9±0.2 (100) 12.7±0.2 (84) 26.6±0.2 (16) 8.5±0.3 (39) 18.9±0.1 (45) SDFc) 4.3±0.2 (100) 4.4±0.1 (96) 2.1±0.1 (4) 2.2±0.1 (35) 7.5±0.2 (61) IDFd) 9.6±0.3 (100) 8.3±0.2 (79) 24.5±0.2 (21) 6.3±0.2 (41) 11.4±0.1 (38) Ash 3.0±0.1 (100) 2.7±0.1 (76) 9.2±0.2 (24) 1.1±0.1 (25) 4.3±0.2 (51) All values are on a dry basis (%) and are the mean±SD of three separate measurements. The number in parentheses represents the proportion (%) of the content in each fraction against the total content in whole grain. a) N¥6.25. b)Calculated from the difference. c)SDF, soluble dietary fiber. d)IDF, insoluble dietary fiber.

Table 2. Mineral content of quinoa grain fractions. Whole grain Milled grain Bran Perisperm Embryo K 825.7±12.3 (100) 639.3±9.4 (71) 2908.5±38.2 (29) 387.9±4.6 (28) 1125.4±18.3 (43) Mg 452.6±13.8 (100) 415.2±12.1 (83) 958.3±17.2 (17) 215.2±10.2 (29) 750.2±11.0 (54) Ca 121.3±5.2 (100) 91.8±3.8 (70) 481.3±11.1 (30) 71.8±2.4 (34) 139.7±7.2 (36) P 359.5±11.3 (100) 360.2±14.2 (92) 350.8±12.2 (8) 286.6±10.5 (50) 482.6±12.8 (42) Fe 9.5±0.2 (100) 9.2±0.1 (87) 14.3±0.2 (13) 7.2±0.1 (48) 11.3±0.3 (39) Mn 3.7±0.5 (100) 3.4±0.2 (81) 10.8±0.6 (19) 2.4±0.3 (38) 5.1±0.7 (43) Cu 0.7±0.1 (100) 0.6±0.1 (81) 1.4±0.2 (19) 0.5±0.1 (44) 0.8±0.1 (37) Zn 0.8±0.1 (100) 0.8±0.1 (91) 0.7±0.1 (9) 0.6±0.1 (48) 1.1±0.2 (43) Na 1.3±0.3 (100) 1.2±0.2 (79) 3.2±0.1 (21) 0.5±0.2 (31) 1.5±0.3 (48) All values (mg/mg%) are the mean±SD of three separate measurements. The number in parentheses represents the proportion (%) of the content in each fraction against the total content in whole grain. Food Components in Quinoa Seed 83

The mineral content of the various quinoa fractions is summa- the starch granules is summarized in Table 3. Quinoa starch ex- rized in Table 2. The proportions of each mineral content in hibited the A X-ray diffaction pattern (Fig. 3) which is common perisperm and embryo against that in whole grain were approxi- to most starches (Atwell et al., 1983, Tang et al., 2000, mately the same. In general, quinoa appeared to be a good source Zheng et al., 1998). DSC analysis (Fig. 4) indicated a gelatiniza- of and to have a good composition; for example, the tion temperature range of 54.0–71.0˚C and enthalpy of 11.0 J/g -phosphorus ratio (1 : 0.7–3.9) was better than that of starch. These values were similar to those of several cereal (1 : 7.8–54.0) (Kagawa, 2001). starches obtained using a similar DSC method (Tang et al., Characterization of starch Figure 1D shows a cross sec- 1998). tion of the quinoa perisperm, in which starch granules are packed Distribution of each protein Table 4 shows the distribu- densely among starch aggregates (20–30 m in diameter) (Fig. tion of each protein by in the fraction. The water-solu- 1E). All quinoa starch granules prepared by the alkali method ble protein and NaCl-soluble protein fractions composed 28.7– were polygonal in shape (Fig. 1F) with a diameter of 0.08–2.0 36.2% and 28.9–32.9% of total protein in each fraction. The dis- m. The shape was similar to that of rice starch, but the particle tribution of each protein showed the same tendency in each frac- size was less than that of rice (0.5–3.9 m), (0.7–39.2 tion. The distribution of quinoa fractions differs from that of m), (1.0–39.2 m) and corn (1.0–7.7 m) (Tang et al., wheat (Ando et al., 2002) and barley (Ando et al., 1996) grains. 1998). Their small size suggests that the starch granules could Fatty acid composition of lipid The fatty acid composition have applications in other industries (Fig. 2). Characterization of of the lipid in each fraction is shown in Table 5. The main satu- rated fatty acid was palmitic acid, which accounted for around 10% of total fatty acids in each fraction. Unsaturated fatty acids were oleic acid (19.7–29.5%), linoleic acid (49.0–56.4%) and linolenic acid (8.7–11.7%), which constituted 87.2–87.8% of total fatty acids. The composition of quinoa oil is similar to that of oil. The embryo is potentially a valuable source of oil. Other components The levels of antinutrients such as phytate, saponin and trypsin inhibitor activity, and lipoxygenase activity are listed in Table 6. Phytate and trypsin inhibitor activity in the embryo were highest. The embryo contained 60% of the total phytate and 89% of the total trypsin inhibitor activity in whole grain. The embryo also contained 62% of total lipoxygen- ase activity. Quinoa flour has a smell similar to soybean, and which limits the appeal of quinoa as a foodstuff. The flour from the milled grain has a stronger smell than the flour from the perisperm. Due to the distribution of lipoxygenase activity, the smell is likely attributable to oxidize and decompose unsaturated fatty acids. Saponin content in the bran was highest, accounting for 86% of the total saponin in whole grain. Fig. 3. X-ray diffraction pattern of quinoa starch. In quinoa grain fractions, the pericarp (bran) contained lots of

Table 4. Distribution of each protein in quinoa grain fractions. Milled grain Perisperm Embryo Water-soluble 33.1±3.1 28.7±1.9 36.2±2.8 NaCl-soluble 28.9±2.8 31.0±2.3 32.9±1.5 n-Propanol-soluble 3.2±0.5 2.7±0.4 2.7±0.4 Lactic acid-soluble 3.2±0.4 3.2±0.3 2.0±0.3 Insoluble 31.6±2.9 34.4±2.0 26.2±2.2 All values are on a dry basis (%) and are the mean±SD of three separate measurements.

Table 5. Fatty acid composition of lipid of quinoa grain fractions. Fatty acid Milled grain Perisperm Embryo Miristic (C14:0) 0.2±0.1 0.1±0.0 0.2±0.1 Palmitic (C16:0) 10.3±0.2 10.8±0.3 9.5±0.2 Stearic (C18:0) 0.8±0.1 0.7±0.2 0.9±0.2 Oleic (C18:1) 25.6±0.3 29.5±0.2 19.7±0.3 Linoleic (C18:2) 52.0±0.4 49.0±0.3 56.4±0.3 Linolenic (C18:3) 9.8±0.1 8.7±0.2 11.7±0.2 Fig. 4. Differential scanning calorimetry thermal curve of quinoa starch. Others 1.3 1.2 1.6 To (onset temperature), Tp (peak temperature) and Tc (conclusion tempera- All values are on a dry basis (%) and are the mean±SD of three separate ture) are 54.0, 62.2 and 71.0˚C, respectively. measurements. 84 H. ANDO et al.

Table 6. Phytate phosphate and total saponin content, and trypsin inhibitor and lipoxygenase activities of quinoa grain fractions. Whole grain Milled grain Bran Perisperm Embryo Phytate phosphate (mg/mg%) 163.2±4.3 (100) 171.3±3.2 (95) 92.5±3.5 (5) 76.2±3.1 (35) 231.7±5.1 (60) Total saponin (mg/mg%) 263.2±9.0 (100) 39.5±1.8 (14) 2704.2±12.3 (86) 15.6±0.2 (3) 95.3±1.9 (11) Trypsin inhibitor activity (units/g) 68.9±1.9 (100) 72.5±1.5 (99) 5.7±0.7 (1) 7.9±0.5 (10) 162.5±3.2 (89) Lipoxygenase activity (units/g) 50.9±0.9 (100) 55.8±1.1 (100) 0.0 (0) 30.1±0.7 (38) 95.3±1.5 (62) All values are the mean±SD of three separate measurements. The number in parentheses represents the proportion (%) of the content in each fraction against the content in whole grain.

dietary fiber, minerals and saponin, compared to the other frac- Lorenz, K. (1990). Quinoa (Chenopodium quinoa) starch—Physico- tions. In particular, bitter saponin accounted for the above-men- chemical properties and functional characteristics. Starch/Starke, 42, tioned 86% of the total saponin. The embryo was rich in protein 82–86. Mahoney, A.W., Lopez, J.G. and Handricks, D.G. (1975). An evalua- (57% of total protein) and in lipids (49% of total lipid). The tion of the protein quality of quinoa. J. Agric. Food Chem., 24, 190– nutritional composition is similar to that of soybean. Therefore, 196. the embryo may be an excellent source of protein and lipid. The Mastebroek H.D., Limburg, H., Gilles, I. and Maroim, H.J. (2000). main component of the perisperm was starch, which had smaller Occurrence of sapogenins in leaves and of quinoa (Chenopo- dium quinoa Willd). J. Sci. Food Agric., 80, 152–156. granules than several other cereals. The perisperm also contained Mitsunaga, T. and Mitsuda, H. (1975). Comprehensive examination of a lot of protein, lipid, dietary fiber, and minerals but few undesir- protein distribution profile in wheat grain. J. Nutr. Sci. Vitaminol., able components. Consequently, this may be a good foodstuff. 21, 363–372. Mitsunaga, T. (1979). Isolation and characterization of trypsin inhibi- References tors from wheat germ. J. Nutr. Sci. Vitaminol., 25, 43–52. Prosky, L., Asp, N.G., Schweizer, T.F., Devries, J.W. and Furda, I. Ando, H., Iizuka, H., Shimizu, M., Yoshida, T., Inaba, K. and Mitsu- (1988). Determination of insoluble, soluble, and total dietary fiber naga, T. (1996). Comprehensive examination of protein and amino in and food products: interlaboratory study. J. Assoc. Off. acid distribution profiles in barley grain. Nippon Eiyo Shokuryo Anal. Chem., 71, 1017–1023. Gakkaisi, 49, 113–118 (in Japanese). Ranhotra, G.S., Gelroth, J.A., Glaser, B.K., Larenz, K.J. and Johnson Ando, H., Sugi, K., Watanabe, K., Morita, N. and Mitsunaga, T. D. (1993). Composition and protein quality of quinoa. Cereal (2002). Distribution of food components in each fraction in wheat Chem., 70, 303–305. grain. Food Sci. Technol. Res., 8, 10–13. Shirasaka, N., Umehara, T., Murakami, T., Yoshizumi, H. and Shimizu, Atwell, W.A., Patrick, B.M., Johnson, L.A. and Glass, R.W. (1983). S. (1998). Microbial conversion of palmitoleic acid to 9, 12-hexade- Characterization of quinoa starch. Cereal Chem., 60, 9–11. canoic acid by Trichoderma SP. AM076. J. Am. Oil Chem. Soc., 75, Bligh, E.G. and Dyer, W.J. (1959). Rapid method of total lipid extrac- 717–720. tion and purification. Can. J. Biochem. Physiol., 37, 911–917. Takamura, H. and Matoba, T. (1992). Measurement of lipoxygenase Bruin, A. (1964). Investigation of the food value of quinoa and cani- and cycloxygenase. Kagaku To Seibutu, 30, 814–815 (in Japanese). hua seed. J. Food Sci., 29, 872–876. Takeda, C., Takeda, Y. and Hizukuri, S. (1983). Physicochemical prop- Chauhuhan, G.S., Eskin, N.A.M. and Tkachuk, R. (1992). Nutritents erties of lily starch. Cereal Chem., 60, 212–216. and antinutrients in quinoa seed. Cereal Chem., 69, 85–88. Tang, H., Yoshida, T., Watanabe, K. and Mitsunaga, T. (1998). Some Gross, R., Koch, F., Malaga, L., De Miranda, A.F., Schoenberger, H. properties of starch granules in various . Kinki Daigaku Nou- and Trugo, I.C. (1989). Chemical composition and protein quality of gaku Sougou Kenkyusho Houkoku, 6, 83–89 (in Japanese). some local Andean food sources. Food Chem., 34, 25–34. Tang, H., Ando, H., Watanabe, K., Takeda, Y. and Mitsunaga, T. Horwiz, W. ed (1970). Official method of analysis of the association of (2000). Some physicochemical properties of small-, medium- and analytical chemists, 11 th ed. AOAC (Washington D.C.), p. 12. large-granule starches in fractions of waxy barley grain. Cereal Huang, W. and Lantzsch, H. (1983). Sensitive method for the rapid de- Chem., 77, 27–31. termination of phytate in cereals and cereal products. J. Sci. Food Yasui, A., Koizumi, T. and Tsutsumi, H. (1985). Application of hydro- Agric., 34, 1423–1426. chloric acid extraction procedure to determination of metal elements Kagawa, Y. ed (2001). “Standard tables of food composition in Japan in cereal, leaves and marine alge by atomic absorption spectropho- (5 th Revision)” Kagawa Nutrition Univ., Tokyo, pp. 25–41 (in tometry. Nippon Shokuhin Kogyo Gakkaishi, 32, 226–233 (in Japa- Japanese). nese). Koziol, M.J. (1992). Chemical composition an nutritional evaluation Yasui, A. (1996). Phosphorus. In “New Food Analytical Method,” ed of quinoa (Chenopodium quinoa Willd) J. Food Comp. Anal., 5, 35– by R. Nakamura, Korin. Tokyo, pp. 258–259 (in Japanese). 68. Zheng, G.H., Han, H.L. and Bhatty, R.S. (1998). Physicochemical pro- Lalitha, T., Seshadri, R. and Venkatraman, L.V. (1987). Isolation and perties of zero amylose hull-less barley starch. Cereal Chem., 75, properties of saponins from Madhuca butyracea seeds. J. Agric. 520–524. Food Chem., 35, 744–749.