Accepted Manuscript

Title: Determination of soluble sugar profile in rice

Authors: Xianqiao Hu, Changyun Fang, Lin Lu, Zhanqiang

PII: S1570-0232(17)30074-0
DOI: http://dx.doi.org/doi:10.1016/j.jchromb.2017.05.001
Reference: CHROMB 20590

To appear in: Journal of Chromatography B

Received date: 17-1-2017

Please cite this article as: Xianqiao Hu, Changyun Fang, Lin Lu, Zhanqiang Hu,
Yafang Shao, Zhiwei Zhu, Determination of soluble sugar profile in rice, Journal of
Chromatography Bhttp://dx.doi.org/10.1016/j.jchromb.2017.05.001

This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.
 Determination of soluble sugar profile in rice

Xianqiao Hu1,2, Changyun Fang1,2, Lin Lu1,2, Zhanqiang Hu1,2, Yafang

Shao1,2, Zhiwei Zhu1,2*

Ministry of Agriculture, Hangzhou 310006, China

*
Corresponding author. Tel.: +86 571 63370275; fax: +86 571 63370380.

E-mail: [email protected].

Highlights

 A method for analysis of soluble sugar profile in rice by using

IC-PAD was presented.

 Soluble glucose, fructose, sucrose, raffinose and maltose in

rice were determined.

 High performance was obtained for the method.

Abstract

Soluble sugars in rice are the main components affecting sweetness

taste of rice. In this paper, an accurate, precise and rapid method for

simultaneous determination of multi soluble sugars in rice by using ion

chromatography equipped with pulsed amperometric detector was

1
presented. Pretreatment and parameters of ion chromatography and

pulsed amperometric detector were optimized. Regression coefficients (R)

of 0.9998, 1.0000, 0.9979, 0.9998 and 0.9998 were obtained for glucose,

fructose, sucrose, raffinose and maltose, respectively. The recovery

ranges of five sugars were 92.9-112.0 % for milled rice matrix.

Repeatability and reproducibility of the method were 0.8-9.7 % and

1.9-7.6 %, respectively. Method LODs of 3.1-34.6 μg∙g-1 were obtained

for soluble sugars in milled rice matrix.

Keywords: soluble sugar, rice, ion chromatography, pulsed amperometric

detector

1. Introduction

Rice is one of the most important staple foods for world

population[1]. The taste of rice is primarily associated with soluble

components such as soluble sugars and amino acids. It was reported that

soluble sugars in rice such as glucose and sucrose are the main

components affecting sweetness taste of rice[2,3]. Hence, soluble sugar

content and profile is an important index to rice quality especially taste

quality. Sucrose, glucose, fructose, maltose and raffinose are the major

soluble sugars in rice[4-6]. It’s necessary to assess the soluble sugar profile

of rice kernels for rice quality assessment.

2
 There are many methods reported to determine sugar content in

food[7,8]. Hydrometer[9], refractometer[10] were used to measure the total

sugar content. Electronic tongue[2,11] and near infrared spectroscopy

(NIR)[12] were also applied for assessing the sweetness of samples.

However, the results were strongly influenced by the prediction training.

High pressure liquid chromatography(HPLC)[9,13], gas chromatography

(GC)[14] and capillary electrophoresis were also used to determine sugar

content in samples. As sugars do not absorb ultraviolet or visible

wavelengths, it is not possible to determine sugars using HPLC with

ultraviolet-visible detector unless proper derivatization was implemented.

Evaporative light-scattering detector (ELSD) and differential refractive

index detector (RID) are universal detectors, hence they were widely used

for sugar analysis, needing no derivatization.[9-12] ELSD is based on the

ability of particles to cause photon scattering, and hence, it can detect

most compounds less volatile than mobile phase[9]. However, high purity

quality of mobile solution was required for stable baseline. ELSD

calibration response curve is non-linear which might cause errors in

quantification[7]. Furthermore, it possesses relatively low sensitivity, and

consequently is not suitable for trace analysis. RID works as a differential

refractometer that measures the difference in refraction index of eluent

induced by solute[15]. The signal is highly dependent on wavelength,

temperature and density of solute, hence gradient elution is not applicable

3
to RID. Pulsed amperometric detector (PAD) is also used for detecting

sugars and fructan in fruit and vegetable extracts. Compared to ELSD and

RID, PAD is seen to more sensitive and reliable as a detector for

carbohydrates. Besides, PAD is not sensitive to the changes of mobile

solution. Hence, PAD is a more suitable detector for sugar analysis. As

aqueous solution is much more appropriate for electron conductivity

during redox at the electrodes than organic solution, PAD is usually

applied for sugar analysis in inulin[17,18], olive plant[19], rice wine[20], raw

sugar[21]. However, it has not been applied for sugar analysis in rice.

Soluble sugars in rice were extracted and concentrated before introduced

into HPLC with RID[2,6], making the determination tedious.

This work reported a sensitive and reliable method for determining

multi soluble sugars in rice by using IC-PAD. The pretreatment for rice

sample and the condition for separation and detection were optimized.

The performance of the IC-PAD method was investigated.

2. Material and Methods

2.1 Chemicals and materials

High-purity (≥99%) glucose, fructose and sucrose were obtained

from National Institute of metrology, China. High-purity (≥99%) maltose

and raffinose were purchased from Dr. Ehrenstorfer GmbH, Germany. 30 %

4
sodium hydroxide aqueous (NaOH) were purchased from Merck KGaA,

Germany. All solutions were prepared with deionized water using a

Millipore advantage 10 system (Millipore Co. USA).

2.2 Rice samples and pretreatment

Five rice samples were provided by Rice Product Quality

Supervision and Inspection Center, Ministry of Agriculture. Before

testing, rice samples were husked. Then half of husked rice sample were

milled until most of bran and part of embryo had been removed. After

that, both husked rice samples and milled rice samples were ground into

flour by using a Cyclotec 1093 sample mill(Foss Tecator, Sweden). The

rice flour samples were stored in -20 ºC before use.

During extraction, an aliquot of flour sample (0.25 g for husked rice

flour while 0.5 g for milled rice flour) was weighed and extracted with 10

mL of 50 % ethanol aqueous solution. The mixture was kept on a

mechanical shaker for 30 min at the room temperature. After centrifuged

at 3000 rpm for 10 min, the supernatant was collected. The process was

repeated twice and supernatants collected were evenly mixed. Finally, the

mixed supernatant was filtered through a 0.45 μm filter and ready for the

following measurement.

2.3 Determination of soluble sugars

Soluble sugars were determined by 850 Professional IC equipped

with 871 Advanced Bioscan (Metrohm, Shanghai, China). IC separation

5
was performed on a Metrosep Carb 1 column (5.0 μm，150 mm × 4.0 mm)

with a Metrosep guard column. The isocratic mobile phase was consisted

of solvent A (250 mmol·L-1 NaOH) and solvent B (H2O) at a ratio of

40:60. The flow rate for mobile phase solution was 1 mL·min-1. The

eluted analytes were detected and quantified by an 871 Advanced Bioscan

with PAD mode (Metrohm, Shanghai, China). Gold electrode and

Ag/AgCl electrode were used as working electrode and reference

electrode, respectively. The detail parameters for PAD was as following:

E1=0.1 V，t1=0.4 s，E2=0.7 V，t2=0.2 s，E3=-0.1 V，t3=0.4 s，tsample=40

ms. Here, E1, E2, and E3 were defined as the detection potential, oxidation

cleaning potential and reduction cleaning potential, respectively, and t1, t2

and t3 represented time duration to apply E1, E2 and E3, respectively.

Tsample was defined as detection time at the end of E1.

3. Results and Discussion

3.1 Sample pretreatment

Three extraction methods were compared by repeating extraction for

twice. With shake treatment method, the mixture was kept on a

mechanical shaker for 30 min. The mixture was treated with ultrasonic

for 30 min in ultrasonic treatment method. However, it was found that

sample flour was deposited at the bottom of centrifuge tube due to gravity

during ultrasonic treatment. Hence, in ultrasonic and shake treatment

6
method, the mixture was kept in an ultrasonic instrument for 30 min

during which the mixture was shook with a vortex shaker for 5 times to

avoid sample flour being deposited at the bottom. The results were

showed in Fig.1a. It was found that ultrasonic treatment method resulted

in the lowest sugar content. It was mainly because of incomplete

extraction caused by the deposition of sample flour at the bottom. Shake

treatment method resulted in highest contents of all sugars except sucrose

content. The highest sucrose content was obtained by using ultrasonic and

shake treatment method. However, the operation of ultrasonic and shake

treatment method is quite cumbersome. Hence, shake treatment method

was chosen taking account of extraction efficiency and ease of operation.

The number of extraction times was also investigated. With

extraction times increased from 1 to 2, glucose, fructose, sucrose and

maltose content increased while raffinose content decreased (Fig.1b).

However, the increase flatted when extraction times increased from 2 to 4.

Hence, the extraction was repeated twice in the following work.

3.2 Detection potential

Detection potential had great effect on the signals of object sugars

(Fig.2a). The peak areas of glucose and fructose increased rapidly with

the detection potential increased from -0.05 V to 0.1 V, and then leveled

off. Similar trends were observed for other three sugars, except for the

turning points of 0.2 V for sucrose and raffinose, and 0.05 V for maltose.

7
Meanwhile, detection potential affect the baseline and baseline noise as

well. As seen from Fig.2b, sharp decline of baseline drift in 30 min was

observed with detection potential increased from -0.05 to 0.05 V, and then

it leveled off at the detection potential of 0.05 to 0.20V, and finally the

baseline drift increased with the increase of detection potential up to test

voltage. A lowest baseline drift was obtained at the detection potential of

0.1 V. The baseline noise sharply decreased with the increase of detection

potential up to 0.1 V, and then sharply increased again with the increased

of detection potential up to test voltage. A lowest of baseline noise was

obtained at detection potential of 0.1 V. Hence, a detection potential of

0.1 V was employed to balance signal sensitivities of all sugars and

baseline condition.

3.3 Mobile phase solution

A solution containing five sugars was applied to investigate the

retention behaviors. The mobile phase was consisted of solvent A (250

mmol·L-1 NaOH) and solvent B (H2O). The proportion of solvent A

ranged from 10 % to 60 %. Hence, the concentration of NaOH in mobile

phase solution was from 25-150 mmol·L-1. The retention time of five

sugars were showed in Fig.S1. There might be some difficulty in

separations among fructose, sucrose and raffinose. Hence, the retention

behaviors of fructose, sucrose and raffinose were investigated as

representative compounds. As showed in Fig.3, the resolution of sucrose

8
and raffinose decreased with the increase of NaOH concentration.

However, opposite trend was observed with the resolution of fructose and

sucrose. A NaOH concentration of 100 mmol·L-1 (40 % solvent A) was

chosen due to the most time-saving condition with a resolution higher

than 1.5.

3.4 Analytical performance of developed method

A series of standard solutions containing glucose, fructose, sucrose,

raffinose and maltose (Table S1) were selected to investigate the

analytical performance of developed method according to each soluble

sugar content in rice. As seen from Table 1, regression coefficients (R) of

0.9998, 1.0000, 0.9979, 0.9998 and 0.9998 were obtained for glucose,

fructose, sucrose, raffinose and maltose, respectively. Method limits of

determination (LOD) of 18.8, 6.3, 34.6, 3.1 and 3.1 μg∙g-1 were obtained

for glucose, fructose, sucrose, raffinose and maltose, respectively, by

determining one milled rice sample for 7 repeats. Meanwhile, method

LODs of sugars in husked rice matrix were 28.3, 3.1, 141.4, 22.0 and 3.1

μg∙g-1.

The accuracy of developed method was evaluated by recovery assay

in both milled rice matrix and husked rice matrix. Milled rice samples

and husked rice samples were spiked with standard solutions containing

five sugars at low, intermediate and high concentration levels,

respectively, and then determined with developed method for six repeats.

9
The recovery ranges of five sugars were 92.9-112.0 % for milled rice

matrix and 90.7-107.9 % for husked rice matrix (Table 1), indicating the

accuracy of developed method .

The precision of developed method was also evaluated. As seen

from Table 2, the precision of developed method was very good in

determining sugar content in rice. Repeatability (n=6) for milled rice

samples were 0.8-3.0 %, 2.9-7.5 %, 1.0-3.3 %, 1.8-4.7 % and 2.9-8.6 %,

while reproducibility for three different persons were 2.2-5.7 %,

2.8-5.3 %, 1.9-3.3 %, 2.8-3.2 % and 4.7-7.5 % for glucose, fructose,

sucrose, raffinose and maltose, respectively. Repeatability (n=6) for

husked rice were 2.8-8.8 %, 2.1-6.3 %, 1.0-7.1 %, 2.0-9.7 % and

2.4-8.9 %, while reproducibility (n=3) were 3.2-5.9 %, 3.5-6.7 %,

1.7-4.1 %, 2.0-5.9 % and 3.0-7.6 % for glucose, fructose, sucrose,

raffinose and maltose, respectively. Both repeatability and reproducibility

were all lower than 10 %, indicating the acceptable precision of the

method for soluble sugar content analysis.

4. Conclusion

Soluble sugars are the main components for sweetness taste of rice.

In this paper, a sensitive and reliable method for simultaneous

determination of soluble sugars in rice by using IC-PAD was presented.

The method was accurate, precise and rapid. The major soluble sugars

10
including glucose, fructose, sucrose, raffinose and maltose were

determined in both husked and milled rice samples. Regression

coefficients of 0.9979-1.0000 were obtained for five sugars. The recovery

ranges of five sugars were 92.9-112.0 % for milled rice matrix and

90.7-107.9 % for husked rice matrix. Repeatability and reproducibility of

the method were 0.8-9.7 % and 1.9-7.6 %, respectively. Method LODs of

18.8, 6.3, 34.6, 3.1 and 3.1 μg∙g-1 were obtained for glucose, fructose,

sucrose, raffinose and maltose in milled rice matrix, and 28.3, 3.1, 141.4,

22.0 and 3.1 μg∙g-1 in husked rice matrix. The developed method was

applicable to sugar analysis for rice, and useful for sweetness analysis for

rice quality in future.

Acknowledgment

This work was funded by Zhejiang Provincial Natural Science

Foundation of China (grant No. LQ15C200007), Special Fund of Chinese

Central Government for Basic Scientific Research Operations in

Commonweal Research Institutes (project No. 2014RG006-4 and

2014RG006-1), and Science and Technology Planning Project of

Zhejiang Province, China ( project No. 2016C37039), National Natural

Science Foundation of China (project No. 31500246), and the National

Key Research and Development Program of China(2016YFF0201803).

11
References

[1] X.Q. Hu, L. Lu, C.Y. Fang, B.W. Duan, Z.W. Zhu, Determination of

apparent amylose content in rice by using paper-based microfluidic chips,

J. Agric. Food Chem. 63 (2015) 9863-9868.

[2] T.U. Tran, K. Suzuki, H. Okadome, S. Homma, K. Ohtsubo, Analysis

of different the tastes of brown rice and milled rice with milling yields

using a taste sensing system, Food Chem. 88 (2004) 557-566.

[3] T.U. Tran, K. Suzuki, H. Okadome, H. Ikezaki, S. Homma, K.

Ohtsubo, Detection of changes in taste of japonica and indica brown and

milled rice (Oryza sativa L.) during storage using physicochemical

analyses and a taste sensing system, J. Agric. Food Chem. 53 (2005)

1108-1118.

[4] Y.H. Cao, Y. Wang, X.L. Chen, J.N. Ye, Study on sugar profile of rice

during ageing by capillary electrophoresis with electrochemical detection,

Food Chem. 86 (2004) 131-136.

[5] J. Patindol, L. Wang, Y.J. Wang, Cellulase-assisted extraction of

oligosaccharides from defatted rice bran, J. Food Sci. 72 (2007)

C516-C521.

[6] C. Zhu, L.P. Li, X. Liu, Changes in sugars during rice seed

desiccation, Russ. J. Plant Physiol. 53 (2006) 198-204.

[7] L.S. Magwaza, U.L. Opara, Analytical methods for determination of

sugars and sweetness of horticultural products-A review, Sci. Hortic. 184

12
(2015) 179-192.

[8] H.R. Sharma, G.S. Chauhan, K. Agrawal, Physico-chemical

characteristics of rice bran processed by dry heating and extrusion

cooking, Int. J. Food Prop. 7 (2004) 603-614.

[9] C.M. Ma, Z. Sun, C.B. Chen, L.L. Zhang, S.H. Zhu, Simultaneous

separation and determination of fructose, sorbitol, glucose and sucrose in

fruits by HPLC-ELSD, Food Chem. 145 (2014) 784-788.

[10] P. Shanmugavelan, S.Y. Kim, J.B. Kim, H.W. Kim, S.M. Cho, S.N.

Kim, S.Y. Kim, Y.S. Cho, H.R. Kim, Evaluation of sugar content and

composition in commonly consumed Korean vegetables, fruits, cereals,

seed plants, and leaves by HPLC-ELSD, Carbohydr. Res. 380 (2013)

112-117.

[11] Y. Zhang, H.F. Li, Y. Ma, Y. Jin, G.H. Kong, J.M. Lin, Microwave

assisted extraction-solid phase extraction for high-efficient and rapid

analysis of monosaccharides in plants, Talanta 129 (2014) 404-410.

[12] K.L. Cheong, D.T. Wu, Y. Deng, F. Leong, J. Zhao, W.J. Zhang,

S.P. Li, Qualitation and quantification of specific polysaccharides from

Panax species using GC-MS, saccharide mapping and

HPSEC-RID-MALLS, Carbohydr. Polym. 153 (2016) 47-54.

[13] O. Pourali, F.S. Asghari, H. Yoshida, Sub-critical water treatment

of rice bran to produce valuable materials, Food Chem. 115 (2009) 1-7.

[14] S. Sun, H. Wang, J. Xie, Y. Su, Simultaneous determination of

13
rhamnose, xylitol, arabitol, fructose, glucose, inositol, sucrose, maltose in

jujube (Zizyphus jujube Mill.) extract: comparison of HPLC-ELSD,

LC-ESI-MS/MS and GC-MS, Chem. Cent. J. 10 (2016).

[15] M. Raessler, Sample preparation and current applications of liquid

chromatography for the determination of non-structural carbohydrates in

plants, Trac-Trends Anal. Chem. 30 (2011) 1833-1843.

[16] AOAC official method 2000.17 Determination of trace glucose

and fructosein raw cane sugar

[17] S.N. Ronkart, C.S. Blecker, H. Fourmanoir, C. Fougnies, C.

Deroanne, J.C. Van Herck, M. Paquot, Isolation and identification of

inulooligosaccharides resulting from inulin hydrolysis, Anal. Chim. Acta

604 (2007) 81-87.

[18] K. Schutz, E. Muks, R. Carle, A. Schieber, Separation and

quantification of inulin in selected artichoke (Cynara scolymus L.)

cultivars and dandelion (Taraxacum officinale WEB. ex WIGG.) roots by

high-performance anion exchange chromatography with pulsed

amperometric detection, Biomed. Chromatogr. 20 (2006) 1295-1303.

[19] T.R.I. Cataldi, G. Margiotta, L. Iasi, B. Di Chio, C. Xiloyannis,

S.A. Bufo, Determination of sugar compounds in olive plant extracts by

anion-exchange chromatography with pulsed amperometric detection,

Anal. Chem. 72 (2000) 3902-3907.

[20] H. Yu, Y.S. Ding, S.F. Mou, Direct and simultaneous

14
determination of amino acids and sugars in rice wine by

high-performance anion-exchange chromatography with integrated

pulsed amperometric detection, Chromatographia 57 (2003) 721-728.

[21] W. Suksom, W. Wannachai, S. Boonchiangma, S. Chanthai, S.

Srijaranai, Ion chromatographic analysis of monosaccharides and

disaccharides in raw sugar, Chromatographia 78 (2015) 873-879.

Figure Captions

Figure 1 Influence of pretreatment on the final result. (a) sugar contents

obtained with three sample extraction methods; (b) sugar contents

obtained with different times.

Figure 2 Effect of detection potential on signals of five sugars and

baseline condition. (a) effect of detection potential on signals of five

sugars; (b) effect of detection potential on baseline drift and baseline

noise.

Figure 3 Effect of NaOH concentration in mobile phase solution on the

resolutions.

15
Figure 1

(a)

(b)

16
Figure 2

(a)

(b)
17
Figure 3

18
 Table 1 Linear equation, correlation coefficient, method LOD and recovery of developed method
Linear equation Linear range(mg∙L-1) R2 Method LOD1 (μg∙g-1) Method LOD2 (μg∙g-1) Recovery1 (%) Recovery2 (%)

Glucose y=6.9000x+1.9595 1-50 0.9996 18.8 28.3 106.6-112.0 103.6-107.9

Fructose y=4.4973x+0.0181 0.2-10 1.0000 6.3 3.1 92.9-99.4 90.7-102.3

Sucrose y=4.1143x+8.3690 2-100 0.9959 34.6 141.4 95.4-99.4 93.8-101.5

Raffinose y=4.0642x+1.3252 0.5-25 0.9996 3.1 22.0 10.7-111.7 97.5-103.7

Maltose y=4.4988x+0.2098 0.5-25 0.9997 3.1 3.1 96.4-99.1 92.2-106.2

1
: obtained with milled rice matrix;
2
: obtained with husked rice matrix.

19
 Table 2 Repeatability and reproducibility of developed method

Milled rice samples Husked rice samples

Sample Mean Repeatability Reproducibility Mean Repeatability Reproducibility

no. (mg∙g-1) (%, n=6) (%, n=3) (mg∙g-1) (%, n=6) (%, n=3)

glucose 1 0.220 3.0 3.3 0.555 2.8 3.5

2 0.399 1.4 5.7 0.604 6.7 4.9

3 0.240 2.8 2.9 0.631 6.7 5.4

4 0.646 2.5 2.2 1.707 8.8 5.9

5 2.286 0.8 2.6 3.818 3.2 3.2

fructose 1 0.028 7.5 5.3 0.049 3.6 3.5

2 0.050 2.9 2.8 0.066 4.3 6.7

3 0.046 3.4 4.0 0.083 5.2 5.1

4 0.029 4.5 4.3 0.045 6.3 6.2

5 0.114 3.6 3.7 0.116 2.1 4.0

sucrose 1 1.462 1.4 3.3 5.864 1.8 2.2

2 1.740 1.3 2.7 5.238 2.2 1.8

3 1.382 1.3 3.3 6.105 1.8 1.7

4 2.063 3.3 3.3 6.301 7.1 4.1

5 5.524 1.0 1.9 10.301 1.0 2.5

raffinose 1 0.149 1.8 3.0 0.807 3.2 3.3

2 0.192 3.2 2.9 0.774 2.0 2.0

3 0.095 2.8 2.8 0.568 2.4 2.0

4 0.135 4.7 2.9 0.586 9.7 5.9

5 0.251 2.6 3.2 0.630 2.9 4.8

maltose 1 0.029 5.4 7.1 0.178 6.2 6.4

2 0.068 2.9 6.8 0.321 5.1 6.2

3 0.040 5.1 4.7 0.235 8.9 7.6

4 0.113 8.6 7.5 1.220 3.8 5.3

20
5 1.335 4.2 5.4 3.924 2.4 3.0

21