:: Korean Journal of Breeding Science ::
Journal Search Engine
Search Advanced Search Adode Reader(link)
Download PDF Export Citaion korean bibliography PMC previewer open access
ISSN : 0250-3360(Print)
ISSN : 2287-5174(Online)
Korean Journal of Breeding Science Vol.44 No.4 pp.444-449
DOI : https://doi.org/10.9787/KJBS.2012.44.4.444

Determination of Cyanidin-3-Glucoside Content using Visible/Near Infrared Reflectance Spectroscopy (VIS/NIRS) in Black Rice

Myoung-Gun Choung3*, Jeom-Sig Lee1, Do-Yeon Kwak2, No-Bong Park2, You-Chun Song2, Sea-Kwan Oh1, Jeong-Heui Lee1, Mi-Ra Yoon1, Hye-Won Kim1, Jieun Kwak1, Chung-Kon Kim1, Young-Sun Hwang3
3Dept. of Herbal Medicine Resource, Kangwon National University
1Rice Research Division, NICS, RDA, 2Dept. of Functional Crop, NICS, RDA
Received on July 20, 2012. Revised on November 2, 2012. Accepted on November 8, 2012

Abstract

Brown rice grain pigments of black rice have a higher content of bioactive substances such as anti-mutagenic substancethan the non-pigmented rice grain. The major anthocyanin pigment contained in black rice was cyanidin-3-glucoside. This studywas conducted to establish a rapid analysis method for determining cyanidin-3-glucoside contents in flour and whole rice seedsof black rice using VIS/NIRS technique. A total of 60 black rice samples were used for VIS/NIRS equation model developmentand validation. The value of coefficient of determination of external validation (r²) and standard error of performance (SEP)in whole rice seed sample were 0.653 and 97.2, respectively. Therefore, the value of it seemed to be difficult to analyzecyanidin-3-glucoside content in whole rice seed samples using VIS/NIRS. However, in rice flour sample, the best accurateequation model was obtained from the partial least square regression (PLS) method. The value of r², SEP and bias were 22.5,0.922 and -1.45 in the calibration transformed to the N-point smooth of log 1/R signal, 5 factors, respectively. Therefore, theresults of our study clearly demonstrate that the VIS/NIRS method would be applicable only for rapid determination ofcyanidin-3-glucose content in black rice flour samples.

44(4)444(2012.12).pdf709.1KB

INTRODUCTION

Rice (Oryza sativa L.) has variation seed coat colors, such as white, brown, red, and blackish purple. The pigments of colored rice are found in the bran or hull in grain (Shibuya 1990). The edible natural pigments extracted from plant organs become steadily popular to consumer because of those physiological functions desirable for food preservation and human health. So, the natural pigments from colored rice is widely used as food colorants in the processing of wine, bread, ice creams, cakes, and so on (Araceli 2009; Cho et al. 1996; Yoshinaga 1986). 

The pigments extracted from colored rice can be largely divided into two types such as anthocyanins or tannins. Anthocyanin pigments are mainly contained in black rice while tannin pigments are mainly contained in red rice (Choi & Oh 1996; Reddy 1995). The anthocyanin pigments contained in black rice was over 80% of cyaniding-3-glucoside. The other minor pigments were peonidin-3-glucoside, malvidin -3-glucoside, and cyaniding-3-ramnoglucoside (Lee 2010; Nagai et al. 1960; Ryu et al. 1998). Anthocyanin in 10 Korean black rice varieties showed significant differences and Cyaniding-3-glucoside content exhibited 52~1,601 ㎍/g (Lee 2010). Cyaniding-3-glucoside is associated with healthy effects such as anti-oxidant, anti-allergy, anti-inflammatory and anti-cancer function (Lim et al. 2011; Min et al. 2010; Park et al. 2008; Tsude et al. 1994). 

The VIS/NIRS is a multi-trait technique that fulfills most of the requirements for rapid, accurate, and cost-effective mass screening technique for high quality breeding in many crops. In the case of milled rice, the NIRS technique has been shown to produce accurate and reliable results to determine the contents of amylose, protein, amino acid, lipid, moisture and the degree of starch gelatinization (Hwang et al. 1994; Kawamura et al. 1997; Barton et al. 1998; Delwiche et al. 1996; Windharm et al. 1997; Zhang et al. 2011). Also, Delwiche et al. (1996) reported the application of NIRS reflectance analysis of whole-grain rice to predict amylose, protein, whiteness, transparency, milling degree, and paste viscosity characteristics. 

The cyanidin-3-glucoside, which was major anthocyanin in black rice, analysis using HPLC is requiring high cost, more time and labor than NIRS method. Therefore, this study was conducted to establish a rapid analysis method for determining cyanidin-3-glucoside contents of black rice using VIS/NIRS technique and to provide the mass screening technique for high quality blackish rice breeding. 

MATERIALS AND METHODS

Rice samples

A total number of 60 black rice samples were grown at the experimental field of National Institute of Crop Science at Milyang South Korea. The rice samples were harvested at the stage of 20~25% moisture contents and dried immediately to 11~13% constant moisture contents by air-drying at ambient temperatures. The uniform ranged rice samples were used for further analysis. The whole rice samples were measured VIS/NIRS method, and then rice samples (20 g) were ground in a Heiko Sample Mill with 1.0 mm screen for 2 min. The ground rice samples were well-mixed and used for the analysis of cyanidin-3-glucoside content by HPLC and VIS/NIRS methods. 

Analysis of cyanidin-3-glucoside contents

Isolation and purification of cyanidin-3-glucoside in rice sample was performed following the Choi et al. (1994), and the chemical structure of isolated cyanidin-3-glucoside was confirmed by Choung et al. (2001). The relative purity of isolated cyanidin-3-glucoside was 99.5% by RP-HPLC with UV-VIS detector (Fig. 1). 

Fig. 1. Chromatogram of cyanidin-3-glucoside standard (left) and black rice sample (right) using RP-HPLC.

To determine the cyanidin-3-glucoside contents, 0.4 g rice sample was weighted and extracted four times with 10 mL of 1% HCl in 80% methanol (v/v) at 4℃ for 24 hrs, and each extract was centrifuged at 10,000 rpm for 10 min. The combined extract was made up in a total volume 100 mL volumetric flask using distilled water. Finally, the extracted solution was filtered with a 0.45 ㎛ membrane filter to analyze by reverse-phase HPLC(RP-HPLC). The cyanidin-3-glucoside content of crude black rice extract was determination by RP-HPLC. The HPLC system was composed of an L-6200 intelligent pump, an L-4250 UVvisible variable wavelength detector, and a D-2500 integrator (Hitachi Co., Japan). Injections were carried out with a Rheodyne 7725i injector equipped with a 20 μL sample loop. The column was a Tosoh ODS-120T (150×4.6 mm i.d., Japan), and the flow rate was set at 0.7 mL/min by isocratic elution, using a Water: MeOH: Formic acid (70:25:5, v/v/v) with monitoring at 530 nm, and the column temperature was set at 30℃. For protection of the analytical column, a Novapak-pak C18 guard insert column (Waters, Milford, MA) was used. The cyanidin-3-glucoside contents were calculated by HPLC peak area compared with external standard calibration curve. The linear standard calibration curve (r≥ 0.999) was generated by injecting 0.05 μg to 1 μg of purified cyanidin-3-glucoside in 20 μL of 1% HCl in 80% methanol (v/v). 

Visible/near-infrared spectroscopy measurement

The whole and grind samples were scanned on a monochromator NIRS systems model 6500 (Silver Spring, MD, USA). For each sample, approximately 5 g was poured into a small cell cup (standard cell cup) with flour or whole seed, presented to the instrument in reflectance mode, and scanned from 400 to 2,500 nm at 2 nm intervals. Each sample was scanned 2 times, and the average reflectance spectrum was stored for calibration (40 samples) and validation (20 samples). The validation sample set was not used for the calibration. 

The instrument was operated by the software package NIRS 3 (version 3. 11), which includes module for acquisition and processing of spectra. Accuracy of calibration equation was determined as standard error of prediction (SEP) value and coefficient of determination (r2) value of the external validation sample sets. 

RESULTS AND DISCUSSION

The log 1/R spectra of the rice flour with different cyanidin-3-glucoside contents are shown in Fig. 2. The changes on cyanidin-3-glucoside contents of rice flour samples showed vast transformation of absorbance between 400 and 1,200 nm. These spectra indicate that the functional groups of the rice flour colorant were closely connected with this wavelength range. 

Fig. 2. Raw spectra of NIRS with different Cyanidin-3-glucoside (C3G) contents in rice flour samples.

The laboratory reference statistics values for cyanidin-3- glucoside content based on rice flour samples are showed in Table 1. The cyanidin-3-glucoside content of calibration sample set was ranged from 102 to 1,250 mg/100 g and external validation sample set were ranged from 218 to 809 mg/100 g. The range of cyanidin-3-glucoside content in external validation sample set fell within the calibration sample set. 

Table 1. Cyanidin-3-glucoside (C3G) contents of black rice for the NIRS calibration and external validation set.

Calibration was obtained by partial least square regression (PLS) or automatic regression method as described by Williams et al. (1991). Table 2 and 3 showed the values of multiple correlation (MR) and standard error of calibration (SEC) of the cyanidin-3-glucoside content using PLS method and automatic regression method for calibration sample set, respectively. The MR and SEC values of cyanidin-3-glucoside content in calibration sample set by PLS regression method were higher than those by the automatic regression method in both rice flour and whole rice seed sample. Compare to rice flour sample and whole rice seed sample, the MR and SEC values of rice flour sample were higher than those in whole rice seed sample. 

Table 2. Comparisons on the effects of different sample type for calibration equation by partial least square regression (PLS).

Table 3. Comparisons on the effects of different sample type for calibration equation by automatic regression.

Low coefficient of determination (r2 ) of whole rice seed sample set was considered due to the gaps between rice seed in small sample cup, and the degree of uniformity for external shape of rice seed. Norris and Williams (1984) and Windham et al. (1997) reported that the difference of particle size in sample was one of the most important factors in NIRS analysis. Because the difference of particle size causes a change in the amount of radiation scattered by samples. 

In the case of automatic regression method (Table 3), the best accurate results of calibration equation and prediction were obtained by combining the VIS/NIRS wavelength range (400~700 nm and 2,230 nm). Chen et al. (1997) reported that the best calibration and prediction results of quantifying surface lipid content of milled rice were VIS/ NIRS wavelength range (400~700 nm, and 1,500~2,500 nm). However, extending the combined wavelength range to the entire VIS/NIRS spectrum (400~2,500 nm) did not improve the SEC and SEP value. 

External validation sample set allows NIRS equation to be validated for prediction accuracy based on random samples not used in calibration sample sets. The value of coefficient of determination (r2), SEP and bias of cyanidin- 3-glucoside content by the PLS method were higher than those by the automatic regression method in both rice flour and whole seed rice sample. In comparison with different sample types, flour samples were higher than whole rice seed samples (Table 2 and 3). 

The value of coefficient of determination (r2), SEP and bias were represented as accuracy of NIRS equation (Windham et al. 1997). Based on the coefficient of determination (r2), SEP and bias, the optimal equation using partial least square regression method was accurately predicting the cyanidin-3-glucoside contents of external validation sample set. 

The quality of a NIRS equation has been judged by the ability to predict accurately from independent spectra. In this study, the well predicting NIRS equation of cyanidin 3-glucoside content in rice flour sample was obtained from the PLS method. The coefficient of determination (r2), SEP and bias of NIRS equation for cyanidin-3-glucoside in rice flour sample were 0.922, 22.5 and -1.45 in the calibration transformed to the N-point smooth of log 1/R signal, respectively (Fig. 3). However, in the case of whole rice seed samples, coefficient of determination (r2) and SEP values were 0.653 and 97.2 by PLS method (Table 2). The coefficient of determination (r2) and SEP values of whole rice seed samples were lower than those of rice flour samples (Table 2 and 3). Chen et al. (1997) reported that in developing calibration equation by means of PLS techniques, the number of principal components is an important factor affecting SEP. Therefore, SEP tended to increase again as the number of principal components was increased due to data overfitting. The optimum number of PLS factors has been determined for each spectral range by identifying the minimum SEP. In the reported PLS calibration, five principal components were included in the calibration equation (Table 2). 

Fig. 3. Scatter plots of Cyanidin-3-glucoside (C3G) content in brown rice flour samples by HPLC vs. NIRS for the calibration (upper) and external validation (lower) samples set by PLS.

Based on our results, it seemed to be difficult to analyze cyanidin-3-glucoside content in whole rice seed samples using VIS/NIRS method, due to gap between grain samples (different particle size of intact seed), uniformity of colored area in external rice shape and amount of small cell. Accordingly, the VIS/NIRS analysis for cyanidin-3-glucoside content in whole rice seed samples require further examination to be acceptation rapid and accurate scan method.

In rice flour sample, the best accurate equation model was obtained from the PLS method. The value of coefficient of determination (r2) and SEP were 0.922 and 22.5, in the calibration transformed to the N-point smooth of log 1/R signal, 5 factors, respectively. Therefore, the results of our study clearly demonstrate that the VIS/NIRS method would be applicable only for rapid determination of cyanidin -3-glucose content of blackish rice flour samples. 

ACKNOWLEDGMENT

This work was carried out with the support of Cooperative Research Program for Agriculture Science & Technology Development (Project No. PJ9070482012) Rural Development Administration and fund by National Institute Crop Science, Republic of Korea. 

Reference

1.Araceli CO. 2009. Chemical studies of anthocyanins: A review. Food. Chemistry. 113:859- 871.
2.Barton FE, Windham WR, Champagne ET, Lyon BG. 1998. Optimal geometries for the development of rice quality spectroscopic chemometric models. Cereal Chemistry 75(3): 315-319.
3.Chen H, Marks BP, and Siebenmorgen TJ. 1997. Qualifying surface lipid content of milled rice via visible/near-infrared spectroscopy. Cereal chemistry 74(6): 826-831.
4.Cho MH, Yoon HH, and Hahn TR. 1996. Thermal stability of the major color component cyanidin 3-glucoside from a Korean pigmented rice variety in aqueous solution. Agri. Chemistry and Biotech. 39(3):245-248.
5.Choi HC and Oh SK. 1996. Diversity and function of pigments in colored rice. Korea J. Crop Sci. 41: 1-9.
6.Choi SW, Kang WW, and Osawa T. 1994. Isolation and identification of anthocyanin pigments in black rice. Food and Biotechnology 3: 131-136.
7.Choung MG, Baek IY, Kang ST, Han WY, Shin DC, Moon HP, and Kang KH. 2001. Isolation and determination of anthocyanins in seed coats of black soybean (Glycine max (L.) Merr.). J. Agric. Food Chem. 49: 5845-5851.
8.Delwiche SR, Mckenzie KS, Webb BD. 1996. Quality characteristics in rice by near-infrared reflectance analysis of whole-grain milled samples. Cereal Chemistry 73(2) 257-263.
9.Hwang HG, Cho RK, Sohn JK and Lee SK. 1994. Rapid evaluation of chemical components of rice grain using near infrared spectroscopy. Korea J. Crop Sci. 39(1): 7-14.
10.Kawamura S. Natsuga M. Itoh K. 1997. Visual and nearinfrared reflectance spectroscopy for rice taste evaluation. ASAE. 40(6): 1755-1758.
11.Lee HL. 2010. Identification and quantification of anthocyanins from the grains of black rice(Oryza sativa L.) varieties. Food Sci. Biotechnol. 19(2): 391-397.
12.Lim TG, Kwon JY, Kim J, Song NR, Lee KM, Lee KM, Heo YS, Lee HJ, Lee KW. 2011. Cyanidin-3-glucoside suppresses B[a] PDE-induced cyclooxygenase-2 expression by directly inhibiting Fyn kinase activity. Biochem. Pharmacol. 82: 167-174.
13.Min SW, Ryu SN, Kim DH. 2010. Anti-inflammatory effects of black rice, cyanidin-3-O-β-glycoside, and its metabolites, cyanidin and protocatechuic acid. Int. Immunopharmacol. 10: 959–966.
14.Nagai IG. Suzushino G, and Tsuboki Y. 1960. Anthoxanthins and anthocyanins in Oryzaceae. Jpn. J. Breed 10(4): 47-56.
15.Norris KH and Willams PC. 1984. Optimization of mathematical treatments of raw near-infrared signal in the measurement of protein in hard red spring wheat. Influence of particle size. Cereal Chem. 61(2): 158-165.
16.Park YS, Kim SJ, and Chang HI. 2008. Isolation of anthocyanin from black rice (Heugjinjubyeo) and screening of its antioxidant activities. Korea Journal Microbiological Biotechnology 36(1): 55-60.
17.Reddy AR. 1995. Genetic and molecular analysis of the anthocyanin pigmentation pathway in rice. Rice Genetics III:341-352.
18.Ryu SN, Park SZ and Ho VT. 1998. High performance liquid chromatographic determination of anthocyanin pigments in some varieties of black rice. Journal of Food and Drug analysis 6(4): 729-736.
19.Shibuya N. 1990. Chemical structure of cell walls of rice grain and grain quality. Nippon Shokuhin kogyo Gakkaishi 37(9): 740-748.
20.Tsude T, Watanabe M, Ohshima K, Norinobu S, Choi SW, Kawakishi S and Osawa T. 1994. Antioxidative activity of the anthocyanin pigments cyanidin-3-O-β-D-glucoside and cyanidin. J. Agric. Food Chem. 42: 2407-2410.
21.Windham WR, Lyon BG, Champagne ET. 1997. Prediction of cooked rice texture quality using near-infrared reflectance analysis of whole-grain milled samples. Cereal Chemistry 74(5): 626-632.
22.Yoshinaga K. 1986. Liquor with pigments of red rice. J. Brew. Soci. Japan 81: 337-342.
23.Zhang B, Rong ZQ, Shi Y, Wub JG, Shi CH. 2011. Prediction of the amino acid composition in brown rice using different sample status by near-infrared reflectance spectroscopy. Food Chemistry 127: 275–281.