1. food and drink
2. food
3. grains and pasta

Structural and Functional Characteristics of African Rice
(Oryza glaberrima) Flour and Starch
by
Joseph Kwesi Gayin
A Thesis
presented to
The University of Guelph
In partial fulfillment of requirements
for the degree of
Doctor of Philosophy
in
Food Science
Guelph, Ontario, Canada
© Joseph Kwesi Gayin, April, 2015
ABSTRACT
STRUCTURAL AND FUNCTIONAL CHARACTERISTICS OF AFRICAN RICE
(Oryza glaberrima) FLOUR AND STARCH
Joseph Kwesi Gayin
Advisors:
University of Guelph, 2015
Dr. Massimo Marcone
Dr. ElSayed Abdelaal
African rice (Oryza glaberrima) and Asian rice (Oryza sativa) are two distinct types of
domesticated rice with the former being less extensively studied. With the expansion of
rice-based food products in Sub-Saharan Africa, characterization of African rice and their
starches is essential for product development. Seven African Rice Accessions (ARAs)
were evaluated for grain quality attributes, thermal and in vitro digestive properties.
Additionally, the physical, molecular and structural properties of starches isolated from the
ARAs were investigated. They were compared with two Asian Rice Varieties (ARVs) and
NERICA 4 (cross between sativa x glaberrima). The physical grain quality characteristics
of the intermediate-high amylose ARAs were comparable to the low amylose ARVs
however, the ARAs had significantly (p<0.05) lower cooking solids loss and harder texture
while the ARVs were softer and more adhesive. The glycemic indices of both rice types
were > 70. The ARV flours showed higher breakdown tendency than the ARAs. All the
starches displayed 'A' type X-ray diffraction pattern and micrographs showed tightly
packed polyhedral granules. The gelatinization transition temperatures: onset (To), peak
(Tp) and conclusion (Tc) were significantly (p<0.05) higher in ARA flours, and melting
transition temperatures of retrograded flour gels showed significant differences (p<0.05)
among samples. The energy (ΔH) required to gelatinize the starches or melt retrograded
starch gels was greater in the ARAs. The relative molar concentration of B-chains in the
amylopectins was approximately 50% in all samples, consequently, the ratio of A:B-chains
was 1.0. The ARAs and NERICA 4 had greater amount of “fingerprint” Bfp-chains than
the ARVs but the major group of short B-chains was similar in all samples. Positive
correlations (p<0.05) were observed between external chain length (ECL) and To, Tp, Tc
and ΔH. The average degree of polymerization (DP) of the ARA clusters were larger and
the ARAs and NERICA 4 had significantly higher average DP of the branched building
blocks (Bbl) than the ARVs due to abundance of large Bbl of group 6. This also resulted in
higher average number of chains (NC), longer average chain length (CL) and higher degree
of branching (DB) of the Bbl in the ARAs.
DEDICATION
This thesis is dedicated to JEHOVAH GOD in whom I live, move and have my being; my
wife Susuana Gayin and kids- Christopher, Jason, Jeffrey and Amanda.
iv
ACKNOWLEDGMENTS
My profound appreciation goes to the Almighty God for his immeasurable Grace, favor
and protection for my family and I throughout my stay in Canada for my PhD programme
in Food Science. He comforted me and showed me compassion; His steadfast love never
ceased, it was new every morning!
I am grateful to my sponsors; GLOBAL RICE SCIENCE PARTNERSHIP (GRiSP) for the
award of scholarship to pursue this PhD programme in Food Science tenable at the
University of Guelph. I wish to also show gratitude to the AFRICA RICE CENTER
(AfricaRice) for their immeasurable financial support and facilitating role in making this
dream a reality. My special thanks go to the Management, the Training Coordinator and
the Head of Grain Quality Laboratory Dr. John Manful who was so instrumental in many
ways.
It is imperative to mention my Advisor Dr. Koushik Seetharaman and family (Debrah and
Samuel) who gave me an opportunity to work with him. My sincere appreciation to him
for the priceless guidance and advice throughout the time I worked with him. It is difficult
to find words to describe his personality, kindness, selflessness, benevolence and unique
human relations that has arguably made him one of the cherished people in my life.
Unfortunately he did not live to see the completing of the great work he began. May his
soul rest in peace! I wish to express many thanks to Dr. Rickey Yada who took over as
Adviser when Koushik left for the University of Minnesota.
encouragement is worth mentioning.
His inspiration and
My heart goes for Dr. Massimo Marcone who
succeeded Dr. Rickey Yada as Adviser when the latter left for the University of British
v
Columbia. His kindness and willingness to help at all times is amazing to say the least. It
is noteworthy to mention Dr. ElSayed Abdelaal for his co-advisory role. Your valuable
suggestions and encouragement right from the proposal stage through till the end of my
programme is very much appreciated. Dr. John Manful did not allow distance to hinder his
role as a Committee member. He ensured that AfricaRice met her financial obligations and
was very instrumental in getting funds transferred timeously. I owe a great debt to Dr. Eric
Bertoft for his kind advice and valuable suggestions throughout the course of this research.
Dr. Sanaa Ragaee perfectly played the role of a mum to my family and I. She ensured our
well-being and my kids never cease mentioning her name.
The office staff at the Department of Food Science - Dr. Art Hill, Anne Ingram, Leona
Varga and Tricia Anderson deserve recognition. I would like to thank you for all your
administrative support and providing me with laboratory supplies I needed for my research.
I would also like to acknowledge Dr. Lisa Duizer, Dr. Bruce Holbein and Dr. Kakuda.
In no particular order, I would like to acknowledge and appreciate the post doctorial
fellows in the cereal and bake laboratories and my laboratory mates (both seniors and
contemporaries) for making life in Guelph worthwhile and enjoyable: Gupreet Kaur, Jayne
Bock, Zhu Fan and Zhan-Hui Lu, Renuka Waduge, Komeine Nantanga, Shane Walker,
Sahar, Varatharajan Vamadevan, Simarata Dhillon, Danusha Kalinga, George and Lizzy
Annor, Anna Källman, Ronny, Bruce Manion, Matt MacSweeney, Avi Goldstein, Azadeh,
Allison, Ryan West, Hong Kai Ma, Pragyani Bora, Igor Guzar, Saba, Magid, Hanadi,
Seham, Ghazal Peyampour, Vivian Adams, and Enoch Quayson. My thanks also go to the
Anim-Somuah family for their help and support in many ways.
vi
Additionally, I would also like to thank my family Susan, Christopher, Jason, Jeffrey,
Amanda, for their support in many ways. I am forever grateful to my mom Christiana
Sam, all my siblings- Gifty, Sarah and Emma
Finally I would like to again acknowledge the immense contributions of Dr. Koushik
Seetharaman, Dr. John Manful, Dr. Eric Bertoft, Dr. Elsayed Abdelaal and Dr. Massimo
Marcone. To God be the Glory!
vii
Table of Contents
DEDICATION…………………………………………………………………………
iv
ACKNOWLEDGMENTS………………………………………………………................ v
Table of Contents…………………………………………………………………………
viii
List of Tables……………………………………………………………………................ xv
List of Figures……………………………………………………………………………..
xvii
List of Abbreviations……………………………………………………………………… xxi
CHAPTER 1: INTRODUCTION…………………………………………………………
1
CHAPTER 2: LITERATURE REVIEW………………………………………………….
5
2.1.
Rice……………………………………………………………………………...
5
2.2
Proximate Composition of Starch………………………………………………. 7
2.3
Rice Starch………………………………………………………………………
2.3.1
Morphology of Rice Starch Granule……………………………………………. 9
2.3.2
Starch Structures………………………………………………………………...
10
2.3.3
Cooking Characteristics…………………………………………………………
14
2.3.4
Thermal Properties……………………………………………………………… 15
2.3.5
Pasting Properties……………………………………………………………….
18
2.3.6
Swelling and Solubility Characteristics…………………………………………
20
2.4
In vitro Digestibility and Glycemic Attributes of Rice…………………………. 21
2.5
Conclusions …………………………………………………………………….. 23
2.6
Significance of project ………………………………………………………….
24
2.7
References (Chapters 1 and 2)…………………………………………………..
25
viii
9
CHAPTER 3: CLASSIFICATION OF RICE BASED ON STATISTICAL ANALYSIS
OF PASTING PROPERTIES AND APPARENT AMYLOSE CONTENT: THE CASE
OF Oryza glaberrima ACCESSIONS FROM AFRICA………………………………….
42
3.1
Introduction……………………………………………………………………... 44
3.2
Materials and Methods………………………………………………………….. 45
3.3
Materials………………………………………………………………………...
45
3.3.1
Measurement of Pasting Properties ………..…………………………………...
46
3.3.2
Measurement of Apparent Amylose Content…………………………………...
46
3.4
Statistical Analysis……………………………………………………………… 47
3.5
Results and Discussion………………………………………………………….
47
3.5.1
Pasting Properties of Rice Flours ……………………………………................
47
3.5.2
Apparent Amylose Content of Rice Flours …………………………………….. 48
3.6
Classification of Rice Accessions………………………………………………. 50
3.7
Comparison of Clustering Pattern………………………………………………. 50
3.7.1
Cluster I…………………………………………………………………………. 50
3.7.2
Cluster II………………………………………………………………………...
51
3.7.3
Cluster III……………………………………………………………………….
51
3.7.4
Cluster IV……………………………………………………………………….. 52
3.7.5
Cluster V………………………………………………………………………...
3.8
Correlations……………………………………………………………………... 53
3.9
Conclusion………………………………………………………………………
3.10
Acknowledgements……………………………………………………………... 55
3.11
Tables and Figures………………………………………………………………
ix
53
54
56
3.12
References………………………………………………………………………. 68
CHAPTER 4: PHYSICOCHEMICAL PROPERTIES, COOKING QUALITY AND
72
GLYCEMIC ATTRIBUTES OF AFRICAN RICE (Oryza glaberrima)
4.1
Introduction……………………………………………………………………... 74
4.2
Materials and Methods………………………………………………………….. 76
4.3
Materials………………………………………………………………………...
76
4.4
Methods…………………………………………………………………………
77
4.4.1
Physico-Chemical Properties……………………………………………………
77
4.4.2
1000-Kernel Weight…………………………………………………………….
77
4.4.3
Bulk Density…………………………………………………………………….
77
4.4.4
Length-Width Ratio……………………………………………………………..
78
4.4.5
Amylose Content………………………………………………………………..
78
4.4.6
Nitrogen…………………………………………………………………………
78
4.4.7
Total Starch……………………………………………………………………... 78
4.4.8
Colour Measurement of Rice Kernels ………………………………………….. 79
4.4.9
Cooking Characteristics…………………………………………………………
79
4.4.10
Optimum Cooking Time………………………………………………………...
79
4.4.11
Textural Profile Analysis (TPA)………………………………………………... 80
4.4.12
Pasting Properties……………………………………………………………….
81
4.4.13
Differential Scanning Calorimetry (DSC)………………………………………
82
4.4.14
In vitro Digestibility and Expected Glycemic Index of Optimally Cooked Rice
82
4.5
Statistical Analysis……………………………………………………………… 84
4.6
Results and Discussion………………………………………………………….
x
84
4.7
Physicochemical Properties……………………………………………………..
84
4.8
Cooking and Textural Properties………………………………………………..
85
4.9
Pasting and Thermal Characteristics……………………………………………. 85
4.10
In vitro Digestibility of Cooked Rice Samples………………………………….
87
4.11
Discussion……………………………………………………………………….
87
4.12
Conclusion……………………………………………………………………....
91
4.13
Acknowledgements……………………………………………………………... 92
4.14
Tables and Figures………………………………………………………………
4.15
References………………………………………………………………………. 106
93
CHAPTER 5: PHYSICAL AND MOLECULAR CHARACTERIZATION OF
STARCHES ISOLATED FROM Oryza glaberrima ACCESSIONS………………….....
113
5.1
Introduction……………………………………………………………………... 115
5.2
Materials and Methods………………………………………………………….. 117
5.2.1
Samples………………………………………………………………………….
117
5.2.2
Methods…………………………………………………………………………
117
5.2.3
Starch Isolation………………………………………………………………….
117
5.2.4
Physical Characteristics…………………………………………………………
118
5.2.5
Starch Damage…………………………………………………………………..
118
5.2.6
Nitrogen…………………………………………………………………………
118
5.2.7
Particle Size Determination……………………………………………………..
119
5.2.8
Morphology …………………………………………………………………….
119
5.2.9
Scanning Electron Microscopy………………………………………………….
119
5.2.10
Wide Angle X-ray Diffraction Scattering ………..…………………………….
120
xi
5.2.11
Reflectance Spectra after Exposure of Starch to Iodine Vapor…………………. 121
5.2.12
Thermal Properties……………………………………………………………… 121
5.2.13
Rapid Visco Analyzer…………………………………………………………...
5.2.14
Differential Scanning Calorimetry ……...……………………………………… 122
5.2.15
Molecular Composition of Rice Starches……………………………………….
5.2.16
Chain Length and β-Limit Dextrins…………………………………………….. 123
5.2.17
Chain Length of Debranched Starches …………………………………………
124
5.2.18
Amylose ………………………………………………………………………...
124
5.2.19
Statistical Analysis……………………………………………………………… 125
5.3
Results ………………………………………………………………………….. 125
5.3.1
Starch Yield and Composition …………………………………………………
5.3.2
Morphology and Structure of Rice Starch Granules…………………………..... 125
5.3.3
Thermal Properties of Starch Granules…………………………………………. 127
5.3.4
Molecular Structure of Rice Starches…………………………………………...
128
5.4
Discussion……………………………………………………………………….
130
5.5
Conclusion………………………………………………………………………
136
5.6
Acknowledgements……………………………………………………………... 137
5.7
Tables and Figures………………………………………………………………
5.8
References………………………………………………………………………. 155
121
123
125
138
CHAPTER 6: UNIT AND INTERNAL CHAIN PROFILE OF AMYLOPECTIN
ISOLATED FROM Oryza glaberrima ACCESSIONS…………………………………... 162
6.1
Introduction……………………………………………………………………... 164
6.2
Materials and Methods………………………………………………………….. 167
xii
6.2.1
Materials………………………………………………………………………...
167
6.2.2
Methods…………………………………………………………………………
167
6.2.3
Starch isolation………………………………………………………………….
167
6.2.4
Isolation of Amylopectin from Rice Starch……………………………………..
167
6.2.5
Determination of Purity f Amylopectin by Gel Permeation Chromatography….
169
6.2.6
Production of φ,β-Limit Dextrin of Amylopectin………………………………
170
6.2.7
Analysis of Unit Chain Distribution ……………………………………………
171
6.2.8
Statistical Analysis ……………………………………………………………... 172
6.3
Results and Discussion………………………………………………………….
172
6.3.1
Purity of Amylopectin ………………………………………………………….
172
6.3.2
Unit Chain Length Distributions of Amylopectins……………………………... 173
6.3.3
Unit Chain Length Distributions of φ,β-Limit Dextrins………………………...
6.4
Correlations between Chain Length Classifications and Some Functional
173
Properties………………………………………………………………………
177
6.5
Conclusion………………………………………………………………………
179
6.6
Acknowledgements……………………………………………………………... 179
6.7
Tables and Figures………………………………………………………………
6.8
References………………………………………………………………………. 188
180
CHAPTER 7: STRUCTURE OF CLUSTERS AND BUILDING BLOCKS IN
AMYLOPECTIN FROM AFRICAN RICE ACCESSIONS……………………………... 193
7.1
Introduction……………………………………………………………………... 196
7.2
Materials and Methods………………………………………………………….. 198
7.2.1
Materials………………………………………………………………………...
xiii
198
7.2.2
Enzymes………………………………………………………………………… 198
7.2.3
Methods…………………………………………………………………………
199
7.2.4
Starch Isolation………………………………………………………………….
199
7.2.5
Isolation of Amylopectin from Rice Starch and Determination of Purity………
199
7.2.6
Cluster preparation - Time Curse for α-Amylolysis of Amylopectin ………...
199
7.2.7
Production of Clusters and Their Corresponding φ,β-Limit Dextrins…………..
200
7.2.8
Chain Length Distribution of Clusters………………………………………….. 201
7.2.9
Preparation of Building Blocks and Determination of Structure……………….. 202
7.2.10
Statistical Analysis……………………………………………………………… 202
7.3
Results and Discussion………………………………………………………….
203
7.3.1
Time Course Analysis of α-Amylosis of Amylopectin from Rice Starch………
203
7.3.2
Characterization of φ,β-Limit Dextrins Produced from Clusters……………….
204
7.3.3
Unit Chains in Clusters………………………………………………………….
205
7.3.4
Characterization of Building Blocks in Clusters………………………………..
208
7.3.5
Correlations……………………………………………………………………... 211
7.3.6
Conclusions……………………………………………………………………... 211
7.3.7
Acknowledgements……………………………………………………………... 212
7.3.8
Tables and Figures………………………………………………………………
7.3.9
References………………………………………………………………………. 225
CHAPTER 8: CONCLUSIONS AND FUTURE WORK………………………………...
xiv
213
229
List of Tables
Table 2.1
Nutritional Composition of Milled Rice……………………………………… 8
Table 3.1
Descriptive Statistics of Entire Population of 1,020 O. glaberrima
Accessions…………………………………………………………………….
Table 3.2
Distribution of 1020 O. glaberrima Rice Accessions Obtained through
Cluster Analysis Using SPSS………………………………………………
Table 3.3
57
Correlations between Pasting Parameters and Apparent Amylose Content
Within the 5 Clusters of 1,020 O. glaberrima Accessions…………………...
Table 4.1
56
59
Physicochemical Characteristics of African and Asian Rice Kernels and
flours………………………………………………………………………….. 93
Table 4.2
Cooking and Textural Properties of African and Asian Rice Kernels………... 94
Table 4.3
Pasting Properties of Flours from African and Asian Rice Kernels………….. 95
Table 4.4
DSC Gelatinization Temperatures and Enthalpy of Flours from African and
Asian Rice Kernels………….………………………………………………... 96
Table 4.5
Transition Parameters of Stored African and Asian Rice Flour Gels………..
Table 4.6
Nutritional Starch Fractions During Hydrolysis of Cooked African and Asian
Rice…..……………………………………………………………………….
Table 4.7
97
98
Kinetic Constant, Hydrolysis Index and Expected Glycemic Index of Cooked
African and Asian Rice……………………………………………………….. 99
Table 4.8
Correlations between Amylose; Protein and Cooking and Textural Properties 100
Table 4.9
Correlations between Amylose and Thermal and Pasting Properties………… 101
Table 5.1
Characteristics and Composition of Starches Isolated from African Rice
Accession and Asian Rice Varieties………………………………………….. 138
xv
Table 5.2
Pasting Properties of Native Starches Isolated from African Rice Accessions
and Asian Rice Varieties……………………………………………………..
Table 5.3
139
Gelatinization Parameters of Native Starches Isolated from African Rice
Accessions and Asian Rice Varieties…………………………………………
140
Table 5.4
Melting Parameters of Retrograded Starch Gels of African Rice Accessions... 141
Table 5.5
Correlations between Amylose Concentration and Some Measured
Parameters…………………………………………………………………….. 142
Table 6.1
Chain Length Classifications and φ, β-Limit Values of Rice Amylopectin as
obtained by HPAEC Analyses………………………………………………
Table 6.2
Relative Molar Amounts (%) of Chain Groupings in Amylopectins Isolated
from Rice Starches…………………………………………………………...
Table 6.3
181
Molar Ratios of Different Chain Groupings of Rice Amylopectins and Their
φ,β-Limit Dextrins……………………………………………………………
Table 6.4
180
182
Correlation between Chain Groupings of Rice Amylopectins and Some
Function Properties of Starch..………………………………………………..
183
Table 7.1
Characterization of φ,β Limit Dextrins of Clusters from Amylopectin ……… 213
Table 7.2
Relative Molar Distribution (%) of Different Chain Categories in Clusters
Isolated from Amylopectin………………………………………………….
Table 7.3
Structure of Branched Building Blocks in Clusters of Amylopectin from
African Rice and Asian Rice…………………………………………………
Table 7.4
214
215
Building Block Structures of Clusters of Amylopectin from African Rice
Accessions and Asian Rice Varieties…………………………………………
xvi
216
List of Figures
Figure 2.1
Structure of amylose and amylopectin…………………………………….. 11
Figure 3.1
Graphical representation of distribution of 1,020 African rice accessions
into 5 clusters and 23 sub clusters…………………………………………
Figure 3.2
61
Distribution of entire population of 1, 020 O. glaberrima accessions based
on apparent amylose content and pasting properties………………………. 62
Figure 3.3
Distribution of O. glaberrima accessions in cluster I population based on
apparent amylose content and pasting properties………………………….. 63
Figure 3.4
Distribution of O. glaberrima accessions in cluster II population based on
apparent amylose content and pasting properties………………………….. 64
Figure 3.5
Distribution of O. glaberrima accessions in cluster III population based
on apparent amylose content and pasting properties………………………. 65
Figure 3.6
Distribution of O. glaberrima accessions in cluster IV population based
on apparent amylose content and pasting properties………………………. 66
Figure 3.7
Distribution of O. glaberrima accessions in cluster V population based on
apparent amylose content and pasting properties………………………….. 67
Figure.4.1
Pasting profiles of African and Asian rice flours………………………….. 102
Figure.4.2
DSC thermograms of African and Asian rice flours………………………. 103
Figure.4.3
Hydrolysis kinetics of cooked and freeze-dried African and Asian rice
kernels……………………………………………………………………... 104
Figure 4.4
Multiple linear regression models to describe the relationship between
amylose, OCT and cooking and textural parameters……………………… 105
Figure 5.1
SEM micrographs of isolated starch from African rice accessions and
xvii
Asian rice varieties………………………………………………………… 143
Figure 5.2
Particle size distribution of starches isolated from African rice accessions
and Asian rice varieties……………………………………………………. 144
Figure 5.3
Wide angle x-ray diffractograms of starches isolated from African rice
accessions and Asian rice varieties………………………………………... 145
Figure 5.4
Wide angle x-ray diffractograms of starch and iodinated starches from
African and Asian rice…………………………………………………….. 146
Figure 5.5
K/S Spectra of starches isolated from African rice accessions and Asian
rice varieties……………………………………………………………….. 147
Figure 5.6
RVA pasting profiles of starches isolated from African rice accessions
and Asian rice……………………………………………………………… 148
Figure 5.7
Thermal properties of starches isolated from African rice accessions and
Asian rice varieties………………………………………………………… 149
Figure 5.8
Thermal properties of retrograded starch gels fromAfrican rice accessions
and Asian rice varieties……………………………………………………. 150
Figure 5.9
Gel-Permeation Chromatograms of debranched rice starches fractionated
on Sepharose CL-6B………………………………………………………. 151
Figure 5.10
Sepharose CL-2B Gel-Permeation Chromatograms of whole rice starches
and their corresponding β-LDs…………………………………………….. 152
Figure 5.11
Correlation between amylose concentration and pasting parameters……... 153
Figure 5.12
Multiple linear regression models to describe the relationship between
amylose and thermal parameters of isolated starches……………………... 154
Figure 6.1
Gel permeation chromatograms (Sepharose CL-6B) of debranched whole
xviii
starch and purified amylopectin from rice starch (Koshihikari)…………... 184
Figure 6.2
Unit chain outline of debranched rice amylopectins obtained by High
Performance Anion-Exchange Chromatography (HPAEC)………………. 185
Figure 6.3
Unit chain outline of debranched φ,β-limit dextrins of rice amylopectins
obtained by High-Performance Anion-Exchange Chromatography
(HPAEC)…………………………………………………………………... 186
Figure 6.4
Multiple linear regression models to describe the relationship between
ECL, Enthalpy, Final and Setback viscosities……………………………..
Figure 7.1
187
Time course for α-amylolysis of rice amylopectin by gel-permeation
chromatography on Sepharose CL-6B…………………………………….. 217
Figure 7.2
Development of relative molar amounts of branched dextrins in the time
course for α-amylolysis of rice amylopectin for the production of clusters
Figure 7.3
218
Size distribution of φ,β-limit dextrins of clusters isolated from rice
amylopectin by GPC on Sepharose CL-6B………………………………... 219
Figure 7.4
HPAEC chromatogram of unit chain distribution of cluster from TOG
12440 (African rice)……………………………………………………….. 220
Figure 7.5
Internal chain length distribution of cluster from TOG 12440 (African
rice)………………………………………………………………………... 221
Figure 7.6
Size distribution obtained by GPC on Superdex 30 of building blocks in
clusters of rice amylopectin ………………………………………………. 222
Figure 7.7a
HPAEC chromatogram of unit chain distribution of building blocks from
TOG 12440 (African rice). ……………………….……………………….. 223
Figure 7.7b
HPAEC chromatogram of unit chain distribution of building blocks from
xix
TOG 12440 (African rice) ……………………….………………………... 223
Figure 7.7c
Example of unit chain compositions in building blocks of TOG 12440
(African rice) on weight basis……………………………………………... 224
xx
List of Abbreviations
AAC
Apparent Amylose Content
AACC
The American Association of Cereal Chemists
Acrystal
Chains Calculated as all A-chains less Afp
Afp
Fingerprint A-Chains (DP 2)
ARA
African Rice accession
Bfp
Fingerprint B-Chains (DP 3 - DP 7)
BDV
Breakdown Viscosity
BL
Long B-Chains (DP >26)
BL-CLLD
Chain Length of Long B-chains in Limit Dextrins
BS-CLLD
Chain Length of Short B-chains in Limit Dextrins
BS
Short B-Chains
BSmajor
Major Group of Short B-Chains (DP 8-25)
C
Cluster
CL
Average Chain Length
CLLD
Average Chain Length of φ, β-Limit Dextrin
xxi
DB
Degree of Branching
DBbl
Density of Building Blocks
DMSO
Dimethyl Sulfoxide
DP
Degree of Polymerization
DSC
Differential Scanning Calorimetry
ECL
External Chain Length
eGI
Expected Glycemic Index
FV
Final Viscosity
GI
Glycemic Index
GL
Glycemic Load
GPC
Gel-Permeation Chromatography
HPAEC-
High Performance Anion-Exchange Chromatography with Pulsed
PAD
Amperometric Detection
IB-CL
Inter-Block Chain Length
ICL
Internal Chain Length
L
Long Chains in Whole Amylopectin (DP ≥ 38)
xxii
LCAM
Long Chain Amylose
LCL
Chain Length of Long Chains
LER
Length Expansion Ratio
MR
Melting Range
NBbl
Number of Branched Building Blocks
NC
Number of Chains
NERICA
New rice for Africa
OCT
Optimum Cooking Time
PKT
Peak Time
PT
Pasting Temperature
PV
Peak Viscosity
RDS
Rapidly Digestible Starch
RS
Residual Starch
RT%
Percentage retrogradation
RVA
Rapid Visco Analyzer
RVU
Rapid Visco Units
xxiii
S
Short Chains in Whole Amylopectin (DP ≤ 37)
SB
Setback Viscosity
SC
Sub Cluster
SDS
Sodium Dodecyl Sulfate
SEM
Scanning Electron Microscopy
SCAM
Short Chain Amylose
SCL
Chain Length of Short Chains
SL
Solids Loss
Tc
Conclusion of Gelatinisation Temperature
To
Onset of Gelatinisation Temperature
Tp
Peak of Gelatinisation Temperature
TICL
Total Internal Chain Length
TPA
Texture Profile Analysis
TV
Trough Viscosity
VER
Volume Expansion Ratio
WAXS
Wide Angle X-ray Diffraction Scattering
xxiv
WUR
Water Uptake Ratio
β-LDs
Beta Limit Dextrins
ΔH
Enthalpy of Gelatinization
φ,β-LD
Phi Beta Limit Dextrins
λmax
Wavelength at the Absorbance Maxima
xxv
CHAPTER 1: INTRODUCTION
Rice is one of the major cereal crops cultivated worldwide and its starch is one of the key
ingredients of various food products (Lawal et al. 2011). The Food and Agricultural
Organization (FAO) estimates that approximately 496.6 million tonnes of milled rice
(paddy that has its husk, germ and bran layers removed) was produced in 2013 while the
global demand for rice as a food is about 410.4 million tonnes (FAO 2014).
The
production of milled rice in Africa in 2013 was 17.5 million tonnes and the current estimate
for rice imports stands at 13.9 million tonnes (FAO 2014). The two widely cultivated rice
species are Oryza sativa L. (Asian rice) and Oryza glaberrima Steud. (African rice) and
both species can grow in a wide range of water-soil regimes (Luh 1991). Rice is consumed
largely in the cooked form, however, products such as snacks, cakes and noodles can be
made from rice (Marshall et al. 1990). Starch is the predominant component (90% dry
basis) of milled rice and its content of amylose and physico-chemical properties (i.e.
physical and chemical), including gelatinization temperature, gel consistency and pasting
viscosity, determine the cooking and eating qualities of rice. Consequently these properties
are important since there is increased consumer demand for the improvement of grain
quality to meet diverse preferences (Bao et al. 2008).
Some unique properties of rice starch are flavor carrying capability, hypoallergenicity,
bland taste, smoothness, creaminess and spreadability. These properties permit rice starch
and flour to be applied in a number of value added products like gluten free bread,
beverages, puddings and low fat sauces (Lawal et al. 2011; Bryant et al. 2001).
Additionally, the functional properties of rice have significant effects on the quality of
1
starch-based products (Bhattacharyya et al. 2004). Physico-chemical properties of rice
starch depend largely on the rice variety, isolation procedure, climatic and soil conditions
during development of the rice grain (Bao et al. 2004; Singh et al. 2005).
The functional properties of starch such as solubility, swelling, water absorption capacity,
gelatinization and pasting provide useful insights into diverse prospective applications of
starches for domestic and industrial purposes. The gelatinization and pasting properties of
rice starch give indications of the ease of its cooking while rheological characteristics offer
information about the viscoelastic properties of the starches.
The two main components of starch are amylose and amylopectin and the proportions and
structures of these two components are the main factors that affect the cooking quality of
rice. Previous studies have shown that amylose has a large impact on the cooking quality
of a rice variety (Cuevas and Fitzgerald 2007) and its content in rice has been reported to
vary from 0 to 33%. Milled rice cultivars may be generally classified on the basis of their
amylose into waxy (1–2%), very low (2–12%), low (12–20%), intermediate (20–25%), or
high (>25%) (Bao 2012). Rice grains with low amylose levels are associated with tender,
cohesive, glossy grains when cooked, while grains with higher amylose levels appear dry,
fluffy and separated as cooked rice (Preecha 2008).
The textural properties of rice flours are mainly contributed by starch due to its
gelatinization, retrogradation and rheological behavior. Starch gelatinization involves the
disruption of molecular/crystalline order within the starch granule and the process includes
changes in the physical, chemical and nutritional properties of starch (Tsai et al. 1997).
Additional changes include water and heat diffusivity, rheological behavior, swelling and
2
deformation of original shape of starchy products and susceptibility to enzymatic digestion
(Bhattacharyya et al. 2004).
Rice has become a popular food in Africa over the years because of urbanization,
population growth, taste, ease of cooking and changes in consumer habits. This rice diet
shift has been supported by income growth which has shifted urban consumers’ preferences
in favour of products that can be easily cooked, at the expense of other types of cereals and
tuber crops that require more time to prepare (Hirsch 1998; Diagana et al. 1999). However,
the volume of rice produced in West Africa is insufficient to meet the growing consumer
demand in the sub-region (FAO 2012).
Asian Rice (O. sativa) has become popular with farmers compared to African rice (O.
glaberrima) because of comparative advantages of higher grain yield, better grain qualities
and wide adaptability to rice production ecologies and agronomic practices. However, in
spite of its predisposition to lodging and shattering, O. glaberrima is generally adapted to a
number of rice biotic and abiotic stresses and has other interesting traits that include a
preferred taste locally, early maturity, better weed competitiveness and ceremonial and
cultural importance (Maji 2010; Falade et al. 2014). This has necessitated the systematic
characterization of a collection of more than 2,500 accessions of Oryza glaberrima by
Africa Rice Center (AfricaRice) because it is thought that there may be a valuable resource
of potentially very important gene pool that may be of benefit in rice breeding programs in
this region. A special effort is being made to screen for resistance against major diseases
and environmental stresses such as acidity, iron toxicity, cold and salinity. To contribute to
the broader objective of this characterization process, this study seeks to explore the
3
structural and functional characteristics of African rice flour and starches. Samples were
selected from a population of 1,020 African rice accessions from West Africa based on
data of their pasting characteristics and apparent amylose content using hierarchical cluster
analysis technique.
The goal of this research is to ultimately promote and add value to African rice and to
identify African rice accessions exhibiting promising traits for further improvement. The
study is aimed to investigate the following hypotheses:
a) Organization of glucan chains in African rice starches is related to their in vitro
digestibility.
b) Organization of glucan chains in African rice starches is related to their thermal and
pasting properties.
To investigate the above hypotheses the following objectives were studied:
a) Determine the cooking quality and glycemic attributes of selected African rice
accessions
b) Establish the physical, thermal and molecular characteristics of starches isolated from
African rice
c) Investigate the structure of African rice starches and relate it to their thermal and
pasting properties
4
CHAPTER 2: LITERATURE REVIEW
2.1. Rice
Rice is the seed of the monocot plant of the genus Oryza and of the grass family
Gramineae. Other crops that belong to this grass family include wheat, corn, rye, oats and
barley (Khush 1997). Rice cultivation has a long history and it is believed to have been
first domesticated nearly 9000 years ago by inhabitants in the region of the Yangtze River
valley in China (Fuller et al. 2010; Liu et al. 2007). Rice is now grown worldwide and it is
a major source of food for more than half of the world’s population, particularly those
living in some of the most populous countries such as China and India. Rice is cultivated
in three main ecologies namely; rain-fed lowland, rain-fed upland and irrigated plains
(Sarla and Swanmy 2005) and it ranks fifth in terms of area under cultivation after maize,
sorghum, millets and cassava (Nayar 2010).
There are two distinct types of domesticated rice; Oryza sativa (Asian rice) and Oryza
glaberrima (African rice) (Ma and Bennetzen 2004). O. glaberrima is reported to have
been domesticated more than 3,500 years ago from wild ancestors along the Niger River in
Mali, West Africa (Carney 2001). O. sativa is distributed globally but highly concentrated
in Asia making it more diverse than O. glaberrima. The two major cultivated subspecies
of O. sativa are the sticky, short grained japonica or sinica variety and the non-sticky,
long-grained indica variety. O. glaberrima is grown in West Africa which is the most
dominant rice growing region in Africa (Sweeney and McCouch 2007; Oko and Ogwu
2010; Nayar 2010). The growing areas span eighteen countries, namely; Benin, Burkina
Faso, Cameroon, Cape Verde, Chad, Cote d’Ivoire, Gambia, Ghana, Guinea, GuineaBissau, Liberia, Mali, Mauritania, Niger, Nigeria, Senegal, Sierra Leone and Togo (Vernet
5
2002). The two major types of this species reported are the photosensitive type that grows
in deep water and the early erect type that grows both in lowland and upland ecologies,
(Ghesquiere et al. 1997). African rice has some useful qualities, which include its ability to
compete with weeds for nutrients (Jones et al. 1997; ), tolerance to drought conditions,
acidity and iron toxicity (Maji et al. 2001; Sano et al. 1984) and the ability to respond to
low input conditions (Sarla and Swanmy 2005). Additionally, African rice is resistant to
the rice yellow mottle virus (Ndjiondjop et al. 1999), nematodes (PlowRight et al. 1999)
and stem borer (Sauphanor 1985).
Climatic and field conditions, cultural management methods and genetics of the crop
influence the physiological processes that occur within the plant. The above-mentioned
factors also affect the duration of the major developmental phases namely; germination,
vegetative, reproductive and grain filling stages. The composition and dimensions of the
grain that eventually develop are also affected (Champagne 2004). Rice cultivars are
typically different in many physical characteristics and these include kernel size,
length:width ratio, kernel shape and kernel hardness (Pomeranz and Webb 1985). The rice
grain (rough rice or paddy) consists of an outer protective covering (husk) and inner edible
part (brown, cargo, dehulled or dehusked rice) (Juliano and Bechtel 1985). Rice is marked
as paddy, brown rice or polished rice. The hull constitutes 16-28% of the rough rice, while
the pericarp constitutes 1-2%, aleurone, nucellus and seed-coat 4-6%, germ 1%, scutellum
2%, and endosperm 90 to 91% of the brown rice (Juliano 1972). The predominant type of
rice consumed worldwide is the polished or white rice (Nanri et al. 2010) and it is obtained
by subjecting paddy rice through a series of mechanized unit operations (rice milling)
during when the embryo, pericarp, seed-coat, testa and aleurone layer are removed.
6
Consequently, there is considerable loss of nutrients such as protein, fibre, ash, vitamins,
fat, and polyphenols, leaving the endosperm to be largely starch (Shobana et al. 2011).
White rice has lower total dietary fiber content (0.7-2%) than brown rice (3-4%) due to the
removal of the bran layer (Juliano 1985). Using apparent amylose content as a basis,
milled rice cultivars may be generally classified into waxy (0-2%), very low (2-12%), low
(12-20%), intermediate (20-25%) or high (25-33%) (Lawal et al. 2011). The International
Standards organization on the basis of length:width ratio, classifies milled rice kernels into
slender (>3.0), medium (2.1-3.0), bold (1.1-2.0) or round (≤1.0).
2.2 Proximate Composition of Rice
The crude protein content of milled rice has been reported to be approximately 7% and it is
considered to be a major factor in determining the texture, pasting capacity and sensory
characteristics of rice. Rice protein consists of glutelin (80%), globulin (12%), albumin
(5%) and prolamin (3%) (Juliano 1985). The lipid content of rice, which can be up to 15%,
is mainly in the bran layer, however, about 1.5 to 1.7 % is present in milled rice (Juliano
and Goddard 1986). The B vitamins, mainly thiamine, riboflavin, niacin, folate and B6 are
concentrated in the bran layers and no vitamins A, D or C are present in the rice grain
(FAO 1954). Minerals such as calcium, magnesium and phosphorus are present along with
some traces of iron, copper, zinc and manganese (Yousaf 1992). The nutrient data for
white rice (medium-grain, raw unenriched) and brown rice (medium-grain, raw) is
presented in Table 2.1.
7
Table 2.1. Nutritional Composition of milled rice
Value/100g
White rice
Brown rice
(medium-grain, raw,
(medium-grain,
unenriched)
raw)
12.89
12.37
Nutrient
Water
Unit
g
Energy
kcal
360
362
Protein
g
6.61
7.5
Total lipid (fat)
g
0.58
2.68
Carbohydrate, by difference
g
79.34
76.17
Calcium, Ca
mg
9
33
Iron, Fe
mg
0.8
1.8
Magnesium, Mg
mg
35
143
Phosphorus, P
mg
108
264
Potassium, K
mg
86
268
Sodium, Na
mg
1
4
Zinc, Zn
mg
1.16
2.02
Thiamin
mg
0.07
0.413
Riboflavin
mg
0.048
0.043
Niacin
mg
1.6
4.308
Vitamin B-6
mg
0.145
0.509
Folate
µg
9
20
Fatty acids, total saturated
g
0.158
0.536
g
0.181
0.971
g
0.155
0.959
Fatty acids, total
monounsaturated
Fatty acids, total
polyunsaturated
Source: USDA-National Nutrient Database for Standard Reference Release 27 (2014)
8
2.3 Rice Starch
Normal rice starch contains about 70-80% amylopectin and 20-30% amylose (Pérez and
Bertoft 2010) and its crystalline structure exhibits A-type X-ray diffraction pattern. The
granule size of rice starch has been reported to be among the smallest in cereal grains
(Vandeputte and Delcour 2004). Rice starch possesses some unique characteristics such as
bland taste (flavor), smooth texture, creaminess, spreadability, acid resistance, freeze-thaw
stability of pastes, hypoallergenicity and a wide range of amylose:amylopectin ratios.
These qualities make it valuable in the formulation of food products (Mitchell 2009; Lawal
et al. 2011).
2.3.1
Morphology of Rice Starch Granule
Rice starch exists as compound polyhedral, oval, irregular, angular, or smooth shaped
granules and measures between 3 and 8 µm in size. Starch granules are tightly packed in
the endosperm cells within an amyloplast (Poochinya et al. 2008). The granule size of
starch can be measured by a variety of techniques, which include laser light scattering, field
flow fractionation and microscopy (light microscopy and scanning electron microscopy
(SEM) (Lindeboom et al. 2004), but SEM seems to be the most frequently used technique
because, in addition to size it also provides information on the morphology and the
characteristics of the surface of the granule (Chmelik 2001). Wang et al. (2012) studied the
granular characteristics of starches separated from ten Chinese rice cultivars using scanning
electron microscope and reported that the rice starch granules were angular and polygonal,
with characteristic dimensions in the range of 3-8 µm. Additionally the granule size
distribution of the different rice starches exhibited a unimodal shape with a mode diameter
9
of about 10 µm. Zhang et al. (2010) also used SEM to investigate the morphology and
physicochemical properties of mechanically activated rice starches from three non-waxy
rice cultivars (indica, japonica and glutinous rice from China) and reported an average
granule size of 4.16-5.00 µm. The granules had polyhedron shapes with pronounced edges
and their surfaces were smooth.
Singh et al. (2006) also reported rice starch granular size ranging between 1.5 and 5.8 mm
when the relationships between physicochemical, morphological, thermal and rheological
properties of rice starches starch separated from 19 different indica rice cultivars was
studied using SEM. Some rice starch samples from Indian rice cultivars studied by Sodhi
and Singh (2003) showed the presence of hexagonal granules in addition to the normal
pentagonal shaped granules having size in the range of 2.4-5.4 µm. Scanning electron
microscopy was used to determine the morphology of the starch granules isolated from five
improved rice varieties from West Africa. Generally, the starches were polyhedral in
appearance with irregular shapes and the granule sizes varied from 1.5 to 6.1 µm. These
values are similar to those reported by Singh et al. (2006) and Sodhi et al. (2003), but
slightly smaller than that reported by Li and Yeh (2001) for Taiwan rice starches where
lazer particle size analyzer was used to study the granule size distribution.
2.3.2
Starch Structures
Most starch granules are made up of alternating amorphous and semi-crystalline shells
which are between 100 and 400 nm thick at the lowest level of structure (Gallant et al.
1997). These structures are termed “growth rings” and are approximately 400 nm apart in
rice starch as measured by atomic force microscopy (Dang and Copeland 2003). X-ray
10
diffraction investigations have indicated a periodicity of about 9-10 nm within the granule
and the periodicity is interpreted as due to crystalline and amorphous lamellae found within
the semi-crystalline shells (Blanshard et al. 1984).
Most cereal starches give A-type
crystalline pattern, while some tubers, such as potato rhizomes and cereal starches rich in
amylose, yield the B-pattern while legumes give C-patterns (Pérez and Bertoft 2010).
Starch is made up of two types of macromolecules namely, amylose and amylopectin.
Figure 2.1. Structure of Amylose and Amylopectin
Amylose forms about 15-35% of the starch and it is a linear molecule composed of Dglucose units connected through α-(1,4)-linkages and the molecular size, shape and
structure vary with the botanical origin (Tester et al. 2004). Amylose crystallizes readily in
solution, in the form of left-handed double helices that are packed in a parallel fashion
forming either A- or B-type allomorphs (Imberty et al. 1991).
Amylose also forms
complexes with ligands such as iodine, fatty acids and alcohols (Bail et al. 1999). The
dimensions of the single stranded helices that are formed depend on the type of ligand
molecule and the crystalized single helices are known as V-types. Amylopectin on the
other hand is a highly branched molecule with α-(1,4)- linked D-glucose backbones and
about 5% α-(1,6)-linked branches (Pérez and Bertoft 2010; Hizukuri and Takagi 1984).
11
There are two main types of chains in amylopectin, namely short chains with CL 6~36 and
long chains with CL > 36 (Bertoft et al. 2008). Chains that do not carry other chains are
known as A-chains, whereas chains that carry other chains are B-chains. The C-chain
carries the sole reducing end-group of the macromolecule (Peat et al. 1952). All A-chains
are external chains (completely situated outside the outermost branches) while all B-chains
consist of an external chain segment and an internal part. On the basis of size-distribution
of the internal unit chains, (the internal part of the B-chains) Bertoft et al. (2008) divided
the structure of amylopectin was into four characteristic types. Type 1 amylopectin has
very little long B-chains and a broad size-distribution of short B-chains while type 2 has
considerably more long B-chains (especially B2-chains) and a narrower size distribution of
the short chains. Type 3 has a little more long chains compared to type 2, whereas type 4
has considerably more long B-chains (particularly B3-chains). Cereals characteristically
belong to type 1 (e.g., barley, oat, and rye) and type 2 (e.g., maize and rice), whereas
legumes, roots and tubers are of type 3 (e.g., cassava and mung bean) and type 4 (e.g.,
potato and canna) (Bertoft et al. 2008). These structural groups have been shown to
correlate with thermal properties of starches (Vamadevan et al. 2013). In addition to the
unit chain profile of amylopectin, its internal chain structure, which describes the
organization of chains in amylopectins, has also received considerable attention. Internal
chain structure is obtained from limit dextrins, mainly β-limit and φ,β-limit dextrins. βamylase cleaves the external chains from amylopectins resulting in very short A-chain
stubs in the form of maltosyl or maltotriosyl residues. Phosphorylase from rabbit liver
produces φ-limit dextrins that results in the production of glucose 1-phosphate. Perez and
Bertoft (2010) reported that in φ,β-limit dextrins, all A-chains are reduced into 4 or 2
12
residues irrespective of the length of the original chain length. In φ,β-limit dextrins, DP 37 was also classified as fingerprint B-chains based on the fact that they were unique to
certain samples.
The internal structure of amylopectins from diverse botanical sources have been reported
on over the past years (Kalinga et al. 2013; Bertoft 2013; Zhu et al. 2011; Kong et al. 2009;
Laohaphatanaleart et al. 2009; Bertoft et al. 2008; Hanashiro et al. 2005; Klucinec and
Thompson 2002; Bertoft and Koch 2000; Takeda et al. 1987 and Hizukuri 1985).
However, so far no information on the structural characteristics of African rice starch has
been found in literature.
In the study of the cluster structure of amylopectin over time, different models have
emerged. The Hizukuri model (Hizukuri 1985), also known as the "classical" model is the
most referred one. In the cluster model, the clusters are interconnected by long chains
which participate in the formation of the crystalline and amorphous lamellae. In the twodirectional backbone model suggested by Bertoft et al. (2012a), the cluster chains are
anchored perpendicularly to the long chains which are mainly present in the amorphous
region and do not participate in the formation of the crystalline lamellae. A cluster refers
to a group of chains, in which the branches are separated by internal chain segments shorter
than nine glucosyl residues (Bertoft, 2007). In addition to the large units of clusters,
building blocks of clusters have also been investigated in recent times. Building blocks are
basically tightly branched units found in clusters. The branching density in building blocks
is known to be different for starches of different botanical origin (Bertoft, 2012b).
13
2.3.3 Cooking characteristics
Characterization of quality attributes of different types of rice varieties is important for the
improvement in rice quality and rice-based products. The eating and cooking qualities of
rice is very essential to consumers and breeders because most rice is cooked and consumed
directly. Rice eating and cooking qualities are highly related to some easily measurable
physico-chemical properties.
These include gelatinization temperature and pasting
viscosity which reflect the starch functionality of the grain and apparent amylose content
(Singh et al. 2005; Bao 2012). However, amylose content is widely recognized as the most
important determinants for different rice products (Juliano 1998).
Rice with similar
amylose content can have differences in their eating qualities but this can be explained by
differences in their amylopectin structures (Ong and Blanshard 1995). High-amylose rice
cultivars are common in indica rice cultivars and are usually dry, fluffy and separate when
cooked and become hard on cooling. Low-amylose cultivars which are common among
temperate japonica rice cultivars are soft and sticky whereas intermediate-amylose rice is
soft but not sticky (Bao 2012). Apart from the amylose content, the cooking quality of rice
can also be influenced by components such as: proteins, lipids or amylopectin (Cai et al.
2011). The texture and physical appearance of cooked rice may vary depending on the
method of cooking (Leelayuthsoontorn andThipayarat, 2006). The texture of cooked rice is
related to its amylose content and the fine structure of amylopectin. The intra- and/or intermolecular interactions of starch with other components in rice such as proteins, lipids and
non-starch polysaccharides affect the texture of rice (Prasert and Suwannaporn, 2009).
Mohapatra and Bal (2006) reported that the degree of milling, amylose content and grain
thickness affected the cooking qualities as well as textural attributes of three indica rice
14
cultivars. Additionally, the grain thickness of the rice cultivars was found to be negatively
correlated with optimum cooking time, adhesiveness and positively with water uptake
ratio, volume expansion ratio, length expansion ratio, hardness, and cohesiveness. Singh et
al. (2005) worked on the physico-chemical, cooking and textural properties of 23 milled
Indian rice cultivars and reported that cooking time showed a negative correlation with
amylose content and a positive correlation with bulk density of the milled rice. Gruel solid
loss showed a significant positive correlation with the amylose content and was negatively
correlated with cooking time. The rice cultivars with higher cooking time showed lower
gruel solid loss. Textural parameters had a positive correlation with amylose and negative
correlation with cooking time. After studying the effect of processing on the nutritionally
important starch fractions in six rice varieties, Rashmi and Urooj (2003) concluded that
cooking methods (pressure cooking, boiling, straining and steaming) influence the
nutritionally important starch fractions (rapidly digestible starch, slowly digestible starch,
resistant starch) in rice varieties.
2.3.4
Thermal Properties
Starch undergoes an irreversible transition termed gelatinization when it is heated in the
presence of excess water. The process is endothermic and it results in the disruption of
molecular order within the starch granules. Double helical and crystalline structures are
disrupted in starches during the process of gelatinization. Increase in temperature causes
the crystallites to break apart, and then to undergo hydration resulting in several changes
such as granular swelling, native crystalline melting, loss of birefringence, and starch
solubilization (Atwell et al. 1988). When cooled, the starch molecules tend to recrystallize
15
(retrogradation), and the process may be associated with water separating from the gel
(syneresis) (Yeh and Yeh 1993; Hoover and Manuel 1995).
Differential scanning
calorimetry (DSC), X-ray scattering, light scattering, optical microscopy, thermomechanical analysis (TMA) and NMR spectroscopy are some techniques that have been
employed to study structural changes underlying gelatinization (Jenkins and Donald 1998).
The use of the DSC to measure melting transition temperatures and enthalpy of
gelatinization involves the heating of starch (or flour) in excess water. The temperature
range and heating rate are programmed and thermograms are recorded with an empty pan
as a reference.
Thermal properties typically reported using DSC include onset
gelatinization (To), peak (Tp), and conclusion (Tc) temperatures, peak height index (PHI),
gelatinization range (R), and enthalpy (ΔH). Crystallite quality and the overall crystallinity
of the starch are measured by the peak temperature (Tp) and the enthalpy of gelatinization
(ΔHg) respectively (Tester and Morrison 1990). Onset temperature (To) and conclusion
temperature, (Tc) determine the boundaries of the different phases in a semi-crystalline
material like starch (Biliaderis et al. 1986).
The enthalpy of gelatinization (ΔHg) is
estimated by integrating the area under the endothermic peak produced as a result of the
melting of amylopectin crystallites in the starches. ΔHg is expressed as J/g and it gives an
indication of the rupturing of the links between and within the amylopectin double helices
(Cooke and Gidley 1992).
Double helical and crystalline structures are disrupted in
starches during gelatinization. This phase transition involves melting of crystals which is
illustrated by DSC endotherms for various native starches (Jacobs et al. 1995). Li and Yeh
(2001) studied the relationships between thermal, rheological characteristics and swelling
power of starches from different botanical sources that included starch isolated from
16
Taiwanese rice. They reported 57.7 ºC and 65.1 ºC for To and Tp respectively and ΔHg of
11.5 J/g. Starches isolated from five indica rice cultivars with varying amylose content,
were studied by Sodhi and Singh in 2003 using DSC. Temperature ranges of 66.33-67.26
ºC, 69.74-71.94 ºC and 74.08-78.04 ºC representing To, Tp and Tc were reported while ΔHg
ranged from 8.16-11.88 J/g. A high degree of crystallinity, which provides structural
stability and makes the granule more resistant to gelatinization accounts for the high
transition temperatures (Barichello et al. 1990). They also suggested that the differences in
transition temperatures and ΔHg of the starches from the different rice cultivars may be
attributed to differences in amylose contents and their granular structures. Additionally,
they suggested that the starch samples with higher To, Tp, Tc values may be attributed to the
compact nature of their small starch granules and higher degree of molecular order of the
granules (Krueger et al. 1987). Further to this the presence of amylose lowers the melting
point of crystalline regions and the energy for starting gelatinization because amylopectin
plays a major role in starch granule crystallinity. More energy is needed to initiate melting
in the absence of amylose-rich amorphous regions (Kreuger et al. 1987).
Recrystallization of amylopectin branch chains has been reported to occur in a more
ordered manner in native starches than in stored starch gels and this may be the reason for
the occurrence of ΔH (retrogradation) at a lower temperature range than ΔH
(gelatinization) (Ward et al. 1994).
Varying ΔH (retrogradation) values of different
starches may be due to differences in amylose-amylopectin ratios, granular structures and
phosphate esters (Hizukuri et al. 1983; Kasemsuwan et al. 1995). Yamin et al. (1999) also
reported that amylopectin and intermediate materials play a significant role in starch
retrogradation during refrigerated storage and that the intermediate materials having longer
17
chains than amylopectin may also form longer double helices during re-association under
refrigerated storage conditions.
2.3.5
Pasting Properties
When starch is heated continuously in excess of water with stirring, the granules swell
irreversibly and burst as a result of breakage of the native starch structure. Amylose then
leaches out and the granules disintegrate to form a viscous material called paste (BeMiller
2007). Pasting properties of starch are key pointers to how a particular starch sample will
behave during processing and they can be used to assess the suitability of starch as a
functional ingredient in food and other products (Wani et al. 2012). The physical and
chemical characteristics of starch such as amylose:amylopectin ratio, granule size
distribution, average granule size and mineral content influence starch properties (Madsen
and Christensen 1996; Wani et al. 2010). Amylose and amylopectin play significant roles
in the gelatinization and pasting properties of rice starch with amylose being the major
factor influencing the physicochemical properties. While starch swelling is a property of
amylopectin, amylose tends to restrict it (Tester and Morrison 1990).
Starch pasting
characteristics are usually determined by employing a Brabender visco amylograph , rapid
visco analyzer (RVA), rotational rheometers or other viscometers capable of continuously
recording change in viscosity with temperature (Wickramasinghe and Noda 2008;
Tukomane and Varavinit 2008; Lin et al. 2009; Park et al. 2007; Li et al. 2008). The
important parameters commonly measured include pasting temperature, peak time, peak
viscosity (PV), trough viscosity, final viscosity (FV), setback viscosity (SB), and
breakdown viscosity (BD). Pasting temperature refers to the temperature at the onset of
18
increase in viscosity and it provides an indication of the minimum temperature required to
cook a given sample while peak viscosity reflects the water-binding capacity of starch and
often correlates with final product quality because the swollen and collapsed granules relate
to the texture of cooked starch (Wani et al. 2012). Breakdown viscosity is a measure of the
ease of disrupting swollen starch granules and gives an indication of the degree of stability
during cooking of starch (Adebowale and Lawal 2003).
Trough viscosity (hot paste
viscosity) is the viscosity recorded at the maximum temperature of the RVA test. Setback
viscosity, a measure of the tendency of starch paste to retrograde is an indicator of final
product texture and it is linked to syneresis (weeping) during freeze-thaw cycles (Wani et
al. 2012). Setback is also used to describe the increase in viscosity that occurs on cooling a
pasted starch (Fisher and Thompson, 1997). Final viscosity indicates the ability of the
material to form a viscous paste or gel after cooking and cooling. The rheology and
functional properties of starches isolated from five improved rice varieties from West
Africa was studied by Lawal et al. (2011). The peak viscosities of the starches were
reported to range between 147.48 and 209.17 RVU while hot paste (trough) viscosity was
from 123.83-157.08 RVU. The high values of setback 58.83-122.42 RVU and breakdown
viscosities 23.75-65.75 RVU observed in the study were attributed to the high amylose
content of NERICA starch samples. Similarly, upon studying the functional and physicochemical properties of flours and starches of African rice cultivars, Falade et al. (2014) also
reported high setback 150-174 RVU and breakdown viscosities 115.6-162.3 for two
African rice starches and attributed it to high amylose content of the starches. When the
breakdown in viscosity is high, the ability of the starch sample to withstand heating and
shear stress during cooking is low (Adebowale et al. 2005).
19
Peak viscosity, trough
viscosity, breakdown, final viscosity, setback and gelatinization temperature range of
values of 258.1-325.1 RVU, 160.6-178.2 RVU, 80.9-153.6 RVU, 307.7-322.3 RVU,
136.2-158.1 and 78.2-80.2 ºC respectively, were reported for a study on grain qualities and
their genetic derivation of seven New Rice for Africa (NERICA) varieties (Masahiro et al.
2008).
2.3.6
Swelling and Solubility Characteristics
The swelling capacity is a measure of the ability of starch to hydrate under specific
conditions such as temperature and water availability (Lawal et al. 2011). Starch swelling
occurs alongside loss of birefringence and precedes solubilization (Singh et al. 2004).
Amylose-lipid complexes inhibit swelling, and the swelling behavior of cereal starch is
primarily related to amylopectin structure Tester and Morrison (1990a).
It has been
reported that swelling power and solubility can be used to measure the extent of interaction
between starch chains and within the amorphous and crystalline domains of the starch
granule (Ratnayake et al. 2002). A greater swelling capacity is an indication of weaker
binding forces in the starch granules (Hoover et al. 1996). After studying five waxy and
ten normal rice starches for structure-function relationships, Vandeputte et al. (2003)
reported that the swelling power of one waxy and three normal rice starches was a function
of temperature. In the first swelling step, (55-85 ºC) amylose did not influence the swelling
of normal rice starches but granule swelling was increased by short amylopectin chains
(DP 6 to 9) between 55 ºC and 85 ºC.
The second swelling step (95-125 ºC) was
influenced by amylose leaching (decreased swelling power). Lawal et al. (2011) reported
of increase in swelling capacity and solubility with an increase in temperature for starches
20
isolated from five improved rice varieties from West Africa. When temperature was
increased in the presence of water, the mobility of starch molecules increased and as a
result the binding forces were weakened. This process enhanced water diffusion into the
starch granules and leaching of the soluble components of the starch granules thereby
leading to improved starch solubility. Amylose has been reported to constitute the larger
percentage of the amorphous component of the starch granules where water penetration
into the granule is more pronounced. Consequently, the highest swelling and solubility of
NERICA starch sample in this study was attributed to its high amylose component. In a
study on functional and physico-chemical properties of flours and starches of African rice
cultivars, Falade et al. (2014) reported swelling power values of 12.35 and 18.56 mL/g and
solubility values of 5.72 and 8.24% for two African rice cultivars. For sixty-three non
waxy starches studied, Tan and Corke (2002) reported a swelling power range of 11.4-23.5
mL/g.
2.4 In vitro Digestibility and Glycemic Attributes of Rice
The glycemic index (GI) and resistant starch (RS) content are two important indicators of
starch digestibility. Glycemic index is a numerical value, which indicates the postprandial
(after meal) blood glucose response of a test food rich in carbohydrate (i.e. a scale that
ranks carbohydrate-rich foods by their ability to raise blood glucose level) (Jenkins et al.
1987). A GI value greater than 70 is considered high while values between 56 and 69 are
regarded as medium. Values less than 56 are low. Glycemic load (GL) is a function of
carbohydrate intake and Glycemic Index. GL is the product of the grams of carbohydrates
in a serving of food and the GI of the food (Willett et al. 2002). For one serving of a food,
21
a GL greater than 20 is considered high while a GL of 11-19 is considered medium. A GL
of 10 or less is considered low. Using in vitro digestion as a basis of classification, Englyst
et al. (1992) categorized starch into three types, namely rapidly digestible (RDS), slowly
digestible (SDS) and resistant starch (RS) categories. Rapidly digestible starches refer to
starches that are digested after 20 minutes while slowly digestible starches are those
digested between 20 minutes and 120 minutes. Resistant starches are those that remain
undigested after 120 minutes. Enzymatic digestibility is affected by the morphological,
textural and rheological properties of food. Other factors that affect starch digestibility
include presence of proteins, lipids, anti-nutrients/inhibitors and non-starch polysaccharides
(Singh et al. 2010). Non-starchy substances such as proteins or lipids may limit the rate of
enzymatic hydrolysis by blocking the adsorption sites and therefore influence binding of
enzyme (Oates 1977). Hamaker and Bugusu (2003) suggested that protein fractions such
as albumins, globulins and glutenins, attach to the proteins bodies and form a matrix
around starch granules thereby acting as a barrier to amylases. Lehmann and Robin (2007)
reported that the crystalline structure of starches influences their hydrolysis. Amylose
forms complexes with lipids which alter the behavior of starches considerably. These
amylose-lipid complexes are known to be resistant to enzymatic attack (Holm et al. 1983).
Chung et al. (2011) studied the in vitro digestibility of rice starches differing in amylose
content and reported that rice starch from long grain having amylose content of 27% had
RDS content of 39% while glutinous rice starch having 4% amylose had 71% RDS. Miller
et al. (1992) reported the GI of white and brown types of three commercial rice varieties
(Doongara, Calrose and Pelde). Doongara with amylose content of 28% had GI of 64 for
the white rice and 66 for the brown rice. The other two varieties Calrose and Pelde had GI
22
values of 83 and 93 for white rice and 87 and 76 for brown rice, respectively. Sato et al.
(2010) reported a GI of 89 for Japanese Koshikari rice and Frei et al. (2003) reported
higher GIs (>90) for waxy rice cultivars compared to non-waxy cultivars. Ranawana et al.
(2009) reported that basmati rice cooked for longer times caused greater glycemic
response.
2.5
Conclusion
It is believed that O. glaberrima has not been fully exploited in terms of the many traits it
possesses for varietal improvement. It has been reported that there is a high level of
genetic diversity in O. glaberrima accessions of in both apparent amylose content and
pasting properties (Gayin et al. 2015). This diversity suggests that there may be a valuable
resource of potentially very important gene pool that may be of benefit in rice breeding
programs in this region. A detailed molecular analysis of promising accessions from such
as those with a high tendency to retrograde (high setback and potentially slower rate of
digestion due to retrogradation) or form viscous paste/gel after cooking and cooling (high
final viscosities) will ensure the use of the diversity in the accessions to assist in future
development of acceptable commercial cultivated rice varieties that have the desired
amylose content and pasting characteristics to meet consumer preferences and benefit
resource poor farmers and to ultimately reduce the over-dependency on rice imports by
African countries.
23
2.6
Significance of project
This research work will contribute to knowledge on rice by generating information on the
cooking quality and glycemic characteristics and physico-chemical properties and
molecular structure of starches from African rice accessions. The goal of this research is to
ultimately promote and add value to African rice and to identify African rice accessions
exhibiting promising traits for further improvement.
24
2.7
References (Chapters 1 and 2)
Adebowale, K. O., and Lawal, O. S. 2003. Microstructure, physicochemical properties and
retrogradation behavior of Mucuna bean (Mucuna pruriens) starch on heat moisture
treatments. Food Hydrocoll. 17: 265-272.
Atwell, W. A., Hood, L. F., Lineback, D. R., Varriano-Marston, E., and Zobel, H. F. 1988.
The terminology and methodology associated with basic starch phenomena. Cereal
Foods World 33: 306-311.
Bail, P. L., Bizot, H., Ollivon, M., Keller, G., Bourgaux, C., and Buléon, A., 1999.
Monitoring the crystallization of amylose-lipid complexes during maize starch
melting by synchrotron Xray diffraction. Biopolym. 50: 99-110.
Bao, J., Corke, H., and Sun, M. 2004. Genetic diversity in the physicochemical properties
of waxy rice (Oryza sativa L) starch. J.Sci.Food Agric. 84: 1299-1306.
Bao, J., Jin, L., Xiao, P., Shen, S., Sun, M., and Corke, H. 2008. Starch physicochemical
properties and their associations with microsatellite alleles of starch synthesizing
genes in a rice RIL population. J.Agric.Food Chem. 56: 1589-1594.
Bao, J. S. 2012. Toward understanding the genetic and molecular bases of the eating and
cooking qualities of rice. Cereal Foods World 57: 148-156.
Barichello, V., Yada, R. Y., Coffin, R. H., and Stanley, D. W. 1990. Low-temperature
sweetening in susceptible and resistant potatoes - Starch structure and composition.
J. Food Sci. 55: 1054-1059.
25
BeMiller, J. N. 2007. Carbohydrate chemistry for food scientists. 2nd ed. St. Paul, Minn.:
AACC Intl.: 389.
Bertoft, E. 2013. On the building block and backbone concepts of amylopectin structure.
Cereal Chem. 90: 294-311.
Bertoft, E., Koch, K., and Åman, P. 2012a. Structure of building blocks in amylopectins.
Carbohydr. Res. 361:105-113.
Bertoft, E., Koch, K. and Åman, P. 2012b. Building block organisation of clusters in
amylopectin from different structural types. Int.J.Biol.Macromol. 50: 1212-1223.
Bertoft, E., Piyachomkwan, K., Chatakanonda, P., and Sriroth, K. 2008. Internal
unit chain composition in amylopectins. Carbohydr. Polym. 74: 527-543.
Bertoft, E. 2007. Composition of clusters and their arrangement in potato amylopectin.
Carbohydr. Polym. 68: 433-446.
Bertoft, E., and Koch, K. 2000. Composition of chains in waxy-rice starch and its
structural units. Carbohydr. Polym. 41: 121-132.
Bhattacharya, P., Gosh,U., Chowdhuri, U. Roy, Chattopadhyay, P., and Gangopadhyay, H.
2004. Effects of different treatments on physico-chemical properties of rice starch.
J.Sci.Indus.Res. 63: 826-829.
26
Biliaderis, C. G., Page, C. M., Maurice, T. J., and Juliano, B. O. 1986. Thermal
characterization of rice starches: a polymeric approach to phase transitions of
granular starch. J. Agric. Food Chem. 34: 6-14.
Blanshard, J. M. V., Bates, D. R., Muhr, A. H., Worcester, D. L., and Higgins, J. S. 1984.
Small-angle neutron-scattering studies of starch granule structure. Carbohydr.
Polym. 4:427-442.
Bryant, R. J., Kadan R. S., Champagne, E. T., Vinyard B. T., and Boykin, D. 2001.
Functional and digestive characteristics of extruded rice flour. Cereal Chem. 78:
131-137.
Cai, Y., Liu, C., Wang, W., and Cai, K. 2011. Differences in physicochemical properties
of kernels of two rice cultivars during grain formation. J. Sci. Food Agric. 91:
1977-1983.
Carney, J. A. 2001. Black rice: the African origins of rice cultivation in the Americas.
Harvard University Press, Cambridge, Massachusetts.
Champagne, E.T. 2004. Rice: chemistry and technology, 3rd Edition. St Paul, MN, USA,
American Association of Cereal Chemists: 25-32.
Chmelik, J. 2001. Comparison of size characterization of barley starch granules
determined by electron and optical microscopy, low-angle laser light scattering and
gravitational field flow fractionation. J. Inst. Brewing 107:11-17.
27
Chung, H.J., Liu, Q., Lee, L., and Wei, D. 2011. Relationship between the structure,
physicochemical properties and in vitro digestibility of rice starches with different
amylose contents. Food Hydrocoll. 25: 968-975.
Cooke, D., and Gidley, M. J. 1992. Loss of crystalline and molecular order during starch
gelatinisation: Origin of the enthalpic transition. Carbohydr. Res. 227: 103-112.
Cuevas, R. P., and Fitzgerald, M. 2007. Linking starch structure to rice cooking quality.
IREC Farmers’ Newsletter. No. 177. Summer 2007-08: 16-17.
Dang, J. M. C., and Copeland, L. 2003. Imaging rice grains using atomic force
microscopy. J. Cereal Sci. 37: 165-70.
Diagana, B., Akindès, F., Savadogo, K., Reardon, T., and Staatz, J. 1999. Effects of the
CFA Franc devaluation on urban food consumption in West Africa: overview and
cross-country comparisons. Food Policy. 24: 465-478.
Englyst, H. N., Kingman, S., and Cummings, J. 1992. Classification and measurement of
nutritionally important starch fractions. Eur. J. Clin. Nutr. 46:33-50.
Falade, K. O., Semon, M., Fadairo, O.S., Oladunjoye, A. O., Orou, K. K. 2014. Functional
and physico-chemical properties of flours and starches of African rice cultivars.
Food Hydrocoll. 39: 41-50.
FAO. 1954. Rice and rice diets-a nutritional survey, revised edition. Rome, FAO: 78.
FAO. 2012. Rice Market Monitor - Volume XV-Issue No. 2 April 2012 Ed.
28
FAO. 2014. Rice Market Monitor - Volume XV1I-Issue No. 1 April 2014 Ed.
Falade, K. O., Semon, M, Fadairo, O.S., Oladunjoye, A, O., and Orou, K. K. 2014.
Functional and physico-chemical properties of flours and starches of African
rice cultivars. FoodHydrocoll. 39: 41-50.
Fisher, D. K., and Thompson, D. B. 1997. Retrogradation of maize starch after thermal
treatment within and above the gelatinization temperature range. Cereal Chem. 74:
344-351.
Frei, M., Siddhuraju, P., and Becker, K. 2003. Studies on the in vitro starch digestibility
and the glycemic index of six different indigenous rice cultivars from the
Philippines. Food Chem. 83: 395-402.
French D. 1984. Organization of starch granules. In: Whistler, R. L, Bemiller, J. N.,
Paschal, E. F. editors. Starch chemistry and technology. New York, N.Y.:
Academic Press. : 183-247.
Fuller, D., Sato, Y., Castillo, C., Qin, L., Weisskopf, A., Kingwell-Banham, E., Song, J.,
Ahn, S., and Etten, J. 2010. Consilience of genetics and archaeobotany in the
entangled history of rice. Archaeol. Anthropol. Sci. 2: 115-131.
Gallant, D. J., Bouchet, B., and Baldwin, P. M. 1997. Microscopy of starch: evidence of a
new level of granule organization. Carbohydr. Polym. 32: 177-191.
29
Gayin, J., Chandi, G. K., Manful, J., and Seetharaman, K. 2015. Classification of rice
based on statistical analysis of pasting properties and apparent amylose content:
the case of Oryza glaberrima accessions from Africa. Cereal Chem. 92: 22-28.
Ghesquière, A., Albar, L., Lorieux, M., Ahmadi, N., Fargette, D., Huang, N., McCouch, S.
R., and Notteghem, J. L. 1997. A Major Quantitative Trait Locus for Rice Yellow
Mottle Virus Resistance Maps to a Cluster of Blast Resistance Genes on
Chromosome12. Phytopathology 87: 1243-1249.
Hanashiro, I., Matsugasako, J., Egashira, T., and Takeda, Y. 2005. Structural
characterization of long unit-chains of amylopectin. J. Appl. Glycosci. 52: 233237.
Hirsch, R. 1998. La riziculture dans les pays de l’UEMOA: de la devaluation a la
liberalisation 1993-1997, AFD, Paris.
Hizukuri, S., Shirasaka, K., and Juliano, B. O. 1983. Phosphorus and amylose-branching
in rice starch. Starch/Stärke 35: 348-350.
Hizukuri, S., and Takagi, T. 1984. Estimation of the molecular weight for amylase by the
low angle laser-light-scattering technique combined with high-performance
chromatography. Carbohydr. Res. 134: 1-10.
Hizukuri, S. 1985. Relationship between the distribution of the chain length of
amylopectin and the crystalline structure of starch granules. Carbohydr. Res.
141: 295-306.
30
Holm, J., Bjorck, I., Ostrowska, S., Eliasson, A., Asp, N., and Larsson, K. 1983.
Digestibility of amylose-lipid complexes in-vitro and in- vivo. Starch/Stärke. 35:
294-297.
Hoover, R., and Manuel, H. 1995. A comparative study of the physicochemical properties
of starches from two lentil cultivars. Food Chem. 53: 275-284.
Hoover, R., Sailaja, Y., and Sosulski, F. W. 1996. Characterization of starches from wild
and long-grain brown rice. Food Res. Int. 29: 99-107.
Imberty, A., Buléon, A., Tran, V., and Pérez, S. 1991. Recent advances in knowledge of
starch structure. Starch/Stärke. 43: 375-384.
Jacobs, H., Eerlingen, R. C., Clauwaert, W., and Delcour, J. A. 1995. Influence of
annealing on the pasting properties of starches from varying botanical sources.
Cereal Chem. 72: 480-487.
Jenkins, P. J., and Donald, A. M. 1998. Gelatinization of starch: a combined
SAXS/WAXS/DSC and SANS study. Carbohydr. Res. 308: 133-147.
Jenkins, D., Thorne, M.J., Wolever, T., Jenkins, A.L., Rao, A.V., and Thompson, L.U.
1987. The effect of starch-protein interaction in wheat on the glycemic response
and rate of in vitro digestion. Am. J. Clin. Nutr. 45: 946-951.
Jones, M. P., Dingkuhn, M., Aluko, G. K., and Semon, M. 1997. Interspecific Oryza sativa
L. × O. glaberrima Steud. progenies in upland rice improvement. Euphytica 92:
237-246.
31
Juliano, B.O. 1998. Varietal impact on rice quality. Cereal Foods World 43: 207-222.
Juliano, B.O. 1992. Structure, chemistry and function of the rice grain and its fraction.
Cereal Foods World. 41: 827-832.
Juliano, B. O., and Goddard, M. S. 1986. Cause of varietal difference in insulin and
glucose responses to ingested rice. Plant Foods Hum. Nutr. 36: 35-41.
Juliano, B. O., and Bechtel, D. B. 1985. The rice grain and its gross composition. In Rice
chemistry and technology, St Paul, MN, USA, Am. Assoc. Cereal Chem.: 17-57.
Juliano, B. 0. 1972. The rice caryopsis and its composition. In D.F. Houston, ed. Rice
chemistry and technology, St Paul, MN, USA. Am. Assoc. Cereal Chem.: 1674.
Kalinga, D. N., Waduge, R., Liu, Q., Yada, R. Y., Bertoft, E., and Seetharaman, K. 2013.
On the differences in the granular architecture and starch structure between pericarp
and endosperm wheat starches. Starch/Stärke. 65: 791-800.
Kasemsuwan, T., Jane, J., Schnable, P., Stinard, P., and Robertson, D. 1995.
Characterization of the dominant mutant amylose-extender (Ae1–5180) maize.
Cereal Chem. 75: 457-464.
Khush, G.S. 1997. Origin, dispersal, cultivation and variation of rice. Plant Mol. Biol. 35:
25-34.
Klucinec, J.D., and Thompson D.B. 2002. Structure of amylopectins from ae-containing
maize starches. Cereal Chem. 79: 19-23.
32
Kong, X., Corke, H., and Bertoft, E. 2009. Fine structure characterization of amylopectins
from grain amaranth starch. Carbohydr. Res. 344: 1701-1708.
Krueger, B. R., Knutson, C. A., Inglett, G. E., and Walker, C. E. 1987. A differential
scanning calorimetry study on the effect of annealing on gelatinization behavior of
corn starch. J. Food Sci. 52: 715-718.
Laohaphatanaleart, K., Piyachomkwan, K., Sriroth, K., Santisopasri, V., and Bertoft, E.
2009. A study of the internal structure in cassava and rice amylopectin.
Starch/Stärke. 61: 557-569.
Lawal, O. S., Lapasin, R., Bellich, B., Olayiwola, T. O., Cesàro, A., Yoshimura, M. 2011.
Rheology and functional properties of starches isolated from five improved rice
varieties from West Africa. Food Hydrocoll. 25: 1785-1792.
Leelayuthsoontorn, P., and Thipayarat, A. 2006. Textural and morphological changes
of Jasmine rice under various elevated cooking conditions.
Food Chem. 96: 606- 613.
Liu, L., Lee, G., Jiang, L., and Zhang, J. 2007. Evidence for the early beginning
(c. 9000 cal. BP) of rice domestication in China: a response. The Holocene
17: 1059-1068.
Li, J., and Yeh, A. 2001. Relationships between thermal, rheological characteristics and
swelling power for various starches. J.Food Eng. 50: 141-148.
33
Lin, Q., Xiao, H., Zhao, J., Li, L., and Yu, F. 2009. Characterization of the pasting, flow
and rheological properties of native and phosphorylated rice starches. Starch/
Stärke 61: 709-715.
Lindeboom, N., Chang, P. R., and Tyler, R. T. 2004. Analytical, biochemical and
physico-chemical aspects of starch granule size, with emphasis on small granule
starches: a review. Starch/Stärke. 56: 89-99.
Luh, B.S., ed. 1991. Rice Utilization,Volume II. AVI, New York.
Ma, J., and Bennetzen, J.L. 2004. Rapid recent growth and divergence of rice nuclear
genomes. Proc. Natl. Acad. Sci. U.S.A. 101: 12404-12410.
Masahiro, K., Keitaro, S., Sumiko, N., and Ken’ichi, O. 2008. Grain qualities and their
genetic derivation of 7 New Rice for Africa (NERICA) varieties. J. Agric. Food
Chem. 56: 4605-4610.
Maji, A. T., Gana, A. S., and Ukwungwu, M. N. 2010. Responses of Oryza glaberrima
accessions to rice stresses and their morphological characteristics. African
J. General Agric. 6: 229-234.
Maji, T., Singh, B. N., and Akenova, M. E. 2001. Vegetative stage drought tolerance in O.
glaberrima Steud and O. sativa L. and relationship between drought parameters.
Oryza 38: 17-23.
Marshall, W. E., Normand, F. L., and Goynes, W.R. 1990. Effects of lipid-protein removal
on starch gelatinization in whole grain milled rice. Cereal Chem. 67: 458-463.
34
Madsen, M. H., and Christensen, D. 1996. Changes in viscosity properties of potato
starches during growth. Starch/Stärke 48: 245-249.
Masahiro, K., Keitaro, S., Sumiko, N., and Ken’ichi, O. 2008. Grain qualities and their
genetic derivation of 7 New Rice for Africa (NERICA) varieties. J. Agric.
Food Chem. 56: 4605-4610.
Miller, J. B., Pang, E., and Bramall, L. 1992. Rice: a high or low glycemic index food?
Am. J. Clin. Nutr. 56:1034-1036.
Mitchell, J. R. 2009. Rice starches: production and properties. In: BeMiller, J., and
Whistler, R. Editors. Starch chemistry and technology. New York: Academic
Press. : 569-579.
Mohapatra, D., and Bal, S. 2006. Cooking quality and instrumental textural attributes of
cooked rice for different milling fractions. J. Food Eng. 73: 253-259.
Nanri, A., Mizoue, T., Noda, M., Takahashi, Y., Kato, M. and Inoue, M. 2010. Rice intake
and type 2 diabetes in Japanese men and women: the Japan Public Health Centerbased Prospective Study. Am. J. Clin. Nutr. 92: 1468-1477.
Nayar, N. M. 2010. The history and genetic transformation of the African rice, Oryza
glaberrima Steud. (Gramineae). Curr. Sci. 99: 1681-1689.
Ndjiondjop, M. N., Albar, L., Fargette, D., Fauquet, C., and Ghesquiere, A.1999. The
genetic basis of high resistance to yellow mottle virus (RYMV) in cultivars of the
two cultivated rice species. Plant Dis. 83: 931-935.
35
Oko, A. O., and Ugwu, S. I. 2010. The proximate and mineral compositions of five major
rice varieties in Abakaliki, South-Eastern Nigeria. Int. J. Plant Physiol. Biochem. 3:
25-27.
Ong, M. H., and Blanshard, J. M. V. 1995. Texture determinants in cooked, parboiled rice.
I: Rice starch amylose and the fine structure of amylopectin. J. Cereal Sci. 21: 251260.
Park, I. M., Ibanez, A. M., Zhong, F., and Shoemaker, C. F. 2007. Gelatinization and
pasting properties of waxy and non-waxy rice starches. Starch/ Stärke 59: 388-396.
Peat, S.,Whelan,W. J.,and Thomas,G. J. 1952. Evidence of multiple branching in waxy
maize starch. J. Chem. Soc.: 4546-4548.
Pérez, S., and Bertoft, E. 2010. The molecular structures of starch components and their
contribution to the architecture of starch granules: A comprehensive review.
Starch/Stärke 62: 389-420.
PlowRight, R. A., Coyne, D. L., Nash, P., and Jones, M. P. 1999. Resistance to the rice
nematodes Heterodera sacchari, Meloidogyne graminicola and M. incognita in
Oryza glaberrima and O. glaberrima × O. sativa interspecific hybrids.
Nematology 1: 745-752.
Pomeranz, Y., and Webb, B.D. 1985. Rice hardness and functional properties. Cereal
Foods World 30: 784-788.
Poochinya, P., Onanong, N., Yupadee, P., and Patcharee, U. 2008. Morphological changes
of rice starch during grain development. J. Microscopy Soc. Thailand. 22: 76-79.
36
Prasert, W., and Suwannaporn, P. 2009. Optimization of instant jasmine rice process
and its physicochemical properties. J. Food Eng. 95: 54-61.
Preecha, P. 2008. Variation of the waxy microsatellite allele and its relation to amylose
content in wild rice (Oryza rufipogon Griff.). Asian J. Plant Sci. 7: 156-162.
National Nutrient Database for Standard Reference Release 27. 2014. Agricultural
Research Service, United States Department of Agriculture. Proc. Nat. Aca.
Sci.101: 12404-12410.
Ranawana, D. V., Henry, C. J. K., Lightowler, H. J., and Wang, D. 2009. Glycaemic index
of some commercially available rice and rice products in Great Britain. Int. J. Food
Sci. Nutr. 60:99-110.
Rashmi, S., and Urooj, A. 2003. Effect of processing on nutritionally important starch
fractions in rice varieties. Int. J. Food Sci. Nutr. 54: 27-36.
Ratnayake, W. S., Hoover, R., and Warkentin, T. 2002. Pea starch: composition, structure
and properties-a review. Starch/Stärke 54:217-234.
Sano, Y., Sano, R., and Morishima, H. 1984. Neighbour effects between two occurring
rice species, Oryza sativa and Oryza glaberrima. J. Appl. Ecol. 21: 245–254.
Sarla, N., and Swamy, B. P. M. 2005. Oryza glaberrima: A source for the improvement of
Oryza sativa. Curr. Sci. 89: 955-963.
37
Sato, S., Fukumura, K., Nishiyama, A., Inoue, Y., and Konishi, T. 2010. Glycemic index
and glucose utilization of rice vermicelli in healthy subjects. Biolog. Pharm. Bull.
33: 1385-1393.
Sauphanor, B. 1985. Some factors of upland rice tolerance to stem borers in West Africa.
Insect Sci. Appl. 6: 429-434.
Shobana, S., Malleshi, N. G.,Sudha, V., Spiegelman, D., Hong, B., Hu, F. B., Willett, W.
C., Krishnaswamy, K., and Mohan, V. 2011. Nutritional and sensory profile of two
Indian rice varieties with different degrees of polishing. Int. J. Food Sci. Nutr. 62:
800-810.
Singh N, Sandhu KS, Kaur M. 2004. Characterization of starches separated from Indian
chickpea (Cicer arietinum) cultivars. J. Food Eng. 63(4): 441-449.
Singh, N., Kaur, L., Sodhi, N. S., and Sekhon, K. S. 2005. Physicochemical, cooking and
textural properties of milled rice from different Indian rice cultivars. Food Chem.
89: 253-259.
Singh, N., Kaur, L., Sandhu, K. S., Kaur, J., and Nishinari, K. 2006. Relationships
between physicochemical, morphological, thermal, rheological properties of rice
starches. Food Hydrocoll. 20: 532-542.
Singh, J., Dartois, A., and Kaur, L. 2010. Starch digestibility in food matrix: a review.
Trends Food Sci.Technol. 21:168-180.
38
Sodhi, S., N., and Singh, N. 2003. Morphological, thermal and rheological properties of
starches separated from rice cultivars grown in India. Food Chem. 80:99-108.
Sweeney, M., and McCouch, S. 2007. The complex history of the domestication of rice.
Ann. Bot. 100: 951-957.
Takeda, Y., Hizukuri, S., and Juliano, B. O. 1987. Structures of rice amylopectins with
low and high affinities for iodine. Carbohydr. Res. 168: 79-88.
Tan, Y., and Corke, H. 2002: Factor analysis of physiochemical properties of 63 rice
varieties. J. Sci. Food Agric. 82: 745-752.
Tester, R. F., and Morrison, W. R. 1990. Swelling and gelatinization of cereal starches I.
Effects of amylopectin, amylose and lipids. Cereal Chem. 67: 551-557.
Tester, R. F., Karkalas, J., and Qi, X. 2004. Starch composition, fine structure and
architecture (review). J. Cereal Sci. 39: 151-165.
Tsai, M., Chin-Fung, L., and Cheng-Yi, L. 1997. Effects of granular structures on the
pasting behaviors of starches. Cereal Chem. 74: 750-757.
Tukomane, T., and Varavinit, S. 2008. Classification of rice starch amylose content from
rheological changes of starch paste after cold recrystallization. Starch/Stärke 60:
292-297.
Vamadevan, V., Bertoft, E., and Seetharaman, K. 2013. On the importance of organization
of glucan chains on thermal properties of starch. Carbohydr. Polym. 92: 16531659.
39
Vandeputte, G. E., Vermeylen, R., Geeroms, J., and Delcour, J. A. 2003. Rice starches.I.
structural aspects provide insight into crystallinity characteristics and gelatinization
behavior of granular starch. J. Cereal Sci. 38: 43-52.
Vandeputte, G. E., and Delcour, J. A. 2004. From sucrose to starch granule to physical
behaviour: a focus on rice starch. Carbohyd. Polym. 58: 245-266.
Vernet, R. 2002. Climate during the late holocene in the Sahara and the sahel: evolution
and consequences on human settlement. In droughts, food, and culture: ecological
change and food security in Africa’s later prehistory (ed. Hassan, F. A), Plenum
press, New York: 47-63.
Wang, L., Xie, B., Xiong, G., Du, X., Qiao, Y., and Liao, L. 2012. Study on the granular
characteristics of starches separated from Chinese rice cultivars. Carbohydr.
Polym. 87: 1038-1044.
Wani, A. A., Singh, P., Shah, M. A., Schweiggert-Weisz, U., Gul, K., and Wani, I. A.
2012. Rice starch Diversity: Effects on structural, morphological, thermal, and
physicochemical properties-A review. Comprehensive Reviews in Food Science
and Food Safety 11: 417-436.
Wani, I. A., Sogi, D. S., Wani, A. A., Gill, B. S., and Shivhare, U. S. 2010. Physicochemical properties of starches from Indian kidney bean (Phaseolus vulgaris)
cultivars. Int. J. Food Sci. Technol. 45: 2176-2185.
Ward, K. E. J., Hoseney, R. C., and Seib, P. A. 1994. Retrogradation of amylopectin
from maize and wheat starches. Cereal Chem. 71: 150-155.
40
Willett,W., Manson, J. and Simin, L. 2002. Glycemic index, glycemic load, and risk of
type 2 diabetes. Am. J. Clin. Nutr. 76: 274-280.
Yamin, F. F., Lee, M., Pollak, L. M., and White, P. J. 1999. Thermal properties of starch
in corn variants isolated after chemical mutagenesis of inbred lines B73. Cereal
Chem. 76: 175-181.
Yeh, A., and Yeh, L. 1993. Some characteristics of hydroxylpropilated and cross-linked
rice starch. Cereal Chem. 70: 273-276.
Zhang, Z., Zhao, S., and Xiong, S. 2010. Morphology and physicochemical properties of
mechanically activated rice starch. Carbohyd Polym 79: 341-348.
Zhu, F., Corke, H., and Bertoft, E. 2011. Amylopectin internal molecular structure in
relation to physical properties of sweetpotato starch. Carbohydr. Polym. 84: 907918.
41
CHAPTER 3: CLASSIFICATION OF RICE BASED ON STATISTICAL
ANALYSIS OF PASTING PROPERTIES AND APPARENT AMYLOSE
CONTENT: THE CASE OF Oryza glaberrima ACCESSIONS FROM AFRICA
42
Abstract
The diversity of 1,020 Oryza glaberrima rice accessions being kept at the Genetic
Resources Unit of the Africa Rice Center, with a varied range of apparent amylose content
and pasting properties, was explored using cluster analysis. Rice cultivars are usually
characterized according to grain dimensions, apparent amylose contents and gelatinization
temperatures, however, this work focused on grouping African rice accessions based on
their pasting properties and apparent amylose content. Using Ward method of hierarchical
cluster analysis, 1,020 rice accessions were initially distributed into five major clusters and
further into twenty-three sub clusters. The distribution pattern indicated that Clusters I, II,
III, IV and V formed 27.6%, 10.2%, 15.8%, 23.7% and 22.6% of the entire population,
respectively. Though some of the groups had similar apparent amylose content, their
pasting properties were very different, making it imperative for further investigations.
Peak viscosity highly correlated (p<0.01) with trough, breakdown and final viscosities in
all the five clusters while correlation between peak viscosity and apparent amylose content
was not significant within clusters II and IV. Additionally, this categorization serves as a
tool for exploring materials that can be employed in the development of rice cultivars for
specific end uses.
Key words: rice, oryza, glaberrima, sativa, apparent amylose, cluster, accessions, pasting
43
3.1 Introduction
Rice is one of the main cereal crops cultivated worldwide on more than 148 million
hectares in different agro-ecological areas and it serves as a primary source of food for
more than two-thirds of the world’s population (Singh et al. 2013; Ndjiondjop et al. 2010).
The genus Oryza comprises of twenty-two wild and two distinct types of domesticated rice
- Oryza sativa (Asian rice) and Oryza glaberrima (African rice). While O. sativa is
distributed globally but highly concentrated in Asia; O. glaberrima is mainly grown in the
West Africa sub region (Sweeney and McCouch 2007; Futakuchi et al. 2008).
The
advantages attributed to O. glaberrima over O. sativa include a preferred taste locally,
early maturity, better weed competitiveness, resistance to biotic stresses such as diseases,
insects and nematodes and abiotic stresses such as drought, flooding and salinity.
However, O. glaberrima is characterized by low yields as well as susceptibility to
shattering and poor resistance to lodging. Its red colored pericarp reduces the perception of
the overall quality when the grains are whitened after dehusking, resulting in a low value
on conventional markets. Additionally, it breaks easily during milling resulting in low
head rice yield of milled rice (Nayar 2010; Watanabe et al. 2002b; Sarla and Swamy 2005).
Although it is still popular in some areas because of its preference for the preparation of
some traditional foods as well as its tolerance to stresses in the field, many farmers in the
West African region are replacing O. glaberrima with O. sativa for want of higher yields
and better grain qualities (Linares 2002; Chakanda et al. 2013; Jones et al. 1997).
In the early 1990s, some O. glaberrima lines with desirable qualities namely, high tillering
ability, rapid seed growth and early maturity were selected from a screening exercise and
subjected to interspecific crosses with O. sativa (high yielding elite japonica upland lines)
44
leading to the development of NERICA (NEw RICe for Africa) lines for lowland and
upland ecologies at the Africa Rice Center. The concept underlying this breeding process
was to exploit the adaptability of O. glaberrima to local growing conditions and the high
yield of O. sativa. The NERICA varieties developed and disseminated have impacted
farmers’ livelihoods and contributed to poverty reduction in sub Saharan Africa. In spite of
this success, it is believed that O. glaberrima has not been fully exploited in terms of other
traits that were not focused on during the varietal improvement. These include amylose
content which is a major factor that influences texture and taste of cooked rice (Futakuchi
and Sié 2009; Zhang et al. 2010). To this end, a collection of Oryza glaberrima accessions
are being systematically characterized by Africa Rice Center (AfricaRice) for resistance
against major diseases and environmental stresses. To contribute to the broader objective
of this characterization process, this study seeks to classify 1,020 O. glaberrima accessions
from West Africa based on data of their pasting characteristics and apparent amylose
content using hierarchical cluster analysis technique. Accessions with potentially different
properties will be selected for further research on their functional, thermal and molecular
characteristics. This study may contribute to the selection of improved material for the
development of suitable O. glaberrima rice cultivars to meet diverse consumer preferences.
3.2 Materials and Methods
3.3 Materials
The O. glaberrima accessions used for the study were supplied by the Africa Rice Center
based in Cotonou, Benin. All the 1020 accessions were planted in July, 2010 and harvested
between October, 2010 and January, 2011.
45
The samples were dehusked in a Satake
dehusker (THU 35B, Satake Co. Ltd. Tokyo, Japan) and brown rice recovered polished in a
single pass friction rice pearler (BS08A Satake Co. Ltd. Tokyo, Japan). The samples were
milled into flour with a UDY cyclone mill (Fort Collins, Co., USA) fitted with a sieve of
0.5 mm mesh size.
3.3.1
Measurement of Pasting Properties
The pasting properties of rice flour samples were measured using Rapid Visco Analyzer
(RVA- SUPER 4, Newport Scientific Pty Ltd., Warriewood, Australia). About 3 g of flour
(12% moisture basis) from each rice sample was weighed directly into the aluminium
canister and mixed with 25 mL of distilled water resulting in a 9.5% suspension. The test
profile involved holding the sample at 50 °C for 1 min, heating to 95 °C at a rate of 12
°C/min, holding at 95 °C for 2.5 min, cooling to 50 °C at a rate of 12 °C/min and holding at
50 °C for 2.5 min. The rotating speed of the paddle was maintained at 160 rpm throughout
the run except for the first 10 s where paddle speed was held at 960 rpm. The pasting
parameters were determined from the RVA curve included peak viscosity (maximum
viscosity attained during heating), hot paste viscosity (trough) and final viscosity or cold
paste viscosity (viscosity at end of cooling period). Breakdown was calculated as the
difference between peak and trough viscosities while setback was calculated as the
difference between the cold paste or final viscosity and the peak viscosity. All analyses
were conducted in duplicate.
3.3.2
Measurement of Apparent Amylose Content
The apparent amylose content of the rice flour was determined in duplicate by using AACC
approved method 61-03 (AACC 2000).
46
3.4 Statistical Analysis
Hierarchical cluster analysis technique (Ward method) was applied to the data set for the
classification of the rice accessions.
The apparent amylose content and the pasting
parameters of the flour suspensions were used to classify 1,020 rice accessions into several
groups and sub groups. The values of the apparent amylose content and pasting parameters
were transformed into Z scores to facilitate their representation on one scale. Hierarchical
clustering method is based on the average linkage clustering which is a method of
calculating Euclidean distance between elements of each cluster. This statistical method
converts various characteristics of objects to quantitative measures (similarity distance) and
correspondingly clusters them at relatively close distances into a category (Lee et al. 2012).
The clustering was carried out using SPSS software IBM® SPSS® Statistics 21 (Windows)
for the descriptive analysis and Pearson correlations between pasting parameters and
apparent amylose content within the generated clusters were observed.
3.5 Results and Discussion
3.5.1
Pasting Properties of Rice Flour
The pasting properties of the glaberrima accessions are presented in Table 3.1. The
pasting temperature, representing the temperature of initial viscosity increase (beginning of
gelatinization) ranged from 86 to 96 °C with an average of 92 °C for the population. It was
observed that the pasting temperatures in this study were far higher than those reported by
Bao et al. 2006 which ranged from 67.5 to 81.5 °C with an average of 75 °C for some O.
sativa varieties. Peak viscosity ranged from 36.4 RVU (TOG 6481) to 230.7 RVU (TOG
16803) averaging 100.2 RVU for the entire population. The peak viscosity range was
47
narrower than earlier reports by Tan and Corke 2002 (92 to 319 RVU), Bao et al. 2006
(90.1 to 305.4 RVU) and Roferos et al. 2006 (105 to 308 RVU). Trough viscosity,
representing the lowest viscosity upon heating was highest at 183.5 RVU for (TOG 16803),
while the lowest value of 36.5 RVU was observed in accession IRGC 80809. The average
for the entire population was 208.5 RVU. The accession IRGC 80809 had the lowest final
viscosity (70.8 RVU), while TOG 16803 recorded the highest final viscosity (421.5 RVU).
The final viscosity range observed in this study is broader than earlier reports by Tan and
Corke 2002 (99 to 365 RVU) and Bao et al. (2006) (111.4 to 412.6 RVU). Breakdown
values for the accessions ranged from -0.5 RVU (IRGC 86808; TOG 12105) to 49.5 RVU
(TOG 10664). Accession TOG 5817 exhibited the lowest setback value (33 RVA) while
accession TOG 7507 had the highest setback value (224.8 RVU). The mean setback
viscosity of 108.4 RVU of the population was higher than the mean of 69.2 reported by
Bao et al. 2006. This difference suggests that O. glaberrima accessions have very different
pasting properties from O. sativa accessions making them an interesting subject for further
investigation.
3.5.2
Amylose Content of Rice Flours
The apparent amylose content (AAC) of the rice flours of the entire 1,020 population of the
Oryza glaberrima accessions ranged between 15.1 (IRGC 86741) and 29.6% (TOG 12440)
with a mean of 26.2% (Table 3.1). The results of this study showed a wider range of AAC
as compared to the findings of Watanabe et al. (2002a, b) where 157 O. glaberrima lines
screened for grain quality showed AAC ranging from 25 to 27%. However, in the analysis
of genotypic diversity in the starch physicochemical properties of 499 non-waxy Oryza
48
sativa lines, Bao et al. 2006 reported a much wider range of AAC (7.9-32.6%). Other
studies on Oryza sativa by Tan and Corke 2002 and Champagne et al. 1999 reported AAC
ranges of 6.3-28.2% and 10.3-24.9%, respectively. The sativas in comparison to the
glaberrimas seem to have a broader range of AAC which include waxy types of rice. No
waxy types were found in the glaberrimas analyzed in this study. Roferos et al. 2006 also
reported on the clustering of rice based on physical and physicochemical properties of
milled rice where AAC ranged from 12 to 33%. An earlier study applied linear correlation,
multiple regression, principal component and cluster analysis to show that sensory scores
of raw and cooked rice were determined by physical quality characteristics such as grain
translucency and chalkiness and pasting properties such as AAC and RVA pasting
viscosity respectively (Roferos and Julianao 1997).
Table 3.2 presents details of the
distribution as per the five clusters and 23 sub clusters. AAC within clusters ranged as
follows; cluster I = 24.7 to 29.4%, cluster II = 20.6 to 29.6%, cluster III = 21.7 to 26.8%,
cluster IV = 15.1 to 28.6%, cluster V = 21 to 28.8%. AAC is related to the cooking quality
of rice and other physico-chemical properties such as gel consistency, pasting viscosity and
gelatinization temperature. It is also considered to be one of the most important predictors
of the eating quality of cooked rice and various rice products (Juliano 1998; Bao 2012).
Milled rice cultivars may be generally classified on the basis of their AAC into waxy (12%), very low (2-12%), low (12-20%), intermediate (20-25%) or high (>25%) (Bao 2012).
The AAC of the entire population of O. glaberrima accessions analyzed suggests that they
fall into the categories of low, intermediate or high amylose types of rice. Cluster IV had
the widest range of AAC followed by cluster II, while cluster I had the narrowest range.
Clusters I, II, III and V had both intermediate and high amylose types of rice but cluster IV
49
included low, intermediate and high amylose types of rice.
The varied AAC of the
glaberrimas suggests that when cooked, they will possess diverse textural characteristics
since cooked rice kernels with high AAC are dry, fluffy, separate and hard whereas those
with low AAC become glossy, soft and sticky (Bao et al. 2006).
3.6 Classification of Rice Accessions
The pattern of distribution of the accessions into various clusters (C) and sub clusters (SC)
is shown in Figure 3.1. Using AAC and RVA pasting parameters of the rice flours as the
basis, the accessions were hierarchically clustered into five groups - C I, C II, C III, C IV
and C V. The distribution pattern indicates that the maximum number of accessions (282)
were included in C I, followed by C IV (242), C V (231), C III (161) and C II (104). All
five clusters had accessions with a large diversity of AAC and pasting properties. (Table
3.2; Figure 3.2)
3.7 Comparison of Clustering Pattern
3.7.1
Cluster I
The 282 accessions in cluster I formed 27.6% of the entire population. These accessions
were subsequently grouped into 5 sub clusters - SC IA, SC IB, SC IC, SC ID and SC IE with a population of 44 (15.6%), 84 (29.8%), 107 (37.9), 26 (9.2%) and 21 (7.4%),
respectively. The AAC of the accessions in all the sub clusters under cluster I put them
into a category of high amylose rice. All the sub clusters had very similar ranges of AAC.
However, their pasting characteristics differed greatly as shown in Figure 3.3 and Table
3.2. While peak and trough viscosities depicted strong diversity in all the 5 sub clusters,
SC IB, SC IC, SC ID and SC IE were close with respect to breakdown viscosity, but that of
50
SC IA was relatively very distinct. SC IA, SC IC, SC IE were likewise similar with respect
to final and setback viscosities, but SC IB and SC ID showed diversity.
3.7.2
Cluster II
Cluster II with 104 accessions comprised of the lowest proportion (10.2%) of the entire
population. Further grouping gave rise to three sub clusters SC IIA, SC IIB, and SC IIC
with populations of 26 (25%), 54 (52%) and 24 (23%) accessions, respectively. The range
of AAC for accessions in C II puts them into intermediate and high amylose types of rice.
C II had the broadest ranges of peak, trough, setback and final viscosities. However, it had
the second broadest AAC range and the narrowest pasting temperature range. SC IIA, and
SC IIB had similar ranges of AAC with SC IIC having the broadest range (Table. 3.2). SC
IIC also had the broadest range with respect to peak, final, setback and trough viscosities
while SC IIB had the narrowest range as regards peak, final and trough viscosities. Pasting
characteristics (Fig. 3.4) showed diversity in all the sub clusters in peak and final
viscosities, peak time and pasting temperatures.
3.7.3
Cluster III
Cluster III was made up of 161 accessions (Table 3.2) and formed 15.8% of the entire
population of the O. glaberrima analyzed and had five sub clusters namely SC IIIA, SC
IIIB, SC IIIC, SC IIID and SC IIIE making up 23%, 24.2%, 26.7%, 11.2% and 14.9% of
the accessions, respectively. Cluster III had the narrowest ranges of peak, trough, setback
and final viscosities. The AAC range of this cluster puts the accessions into a category of
intermediate to high amylose rice. While SC IIIA had the broadest apparent amylose
range, SC IIID had the narrowest range and that of SC IIIB and SC IIIE were similar.
51
Additionally SC IIIA had the broadest ranges with respect to peak, trough, breakdown and
final viscosities (Table 3.2). The sub cluster with the second broadest ranges of apparent
amylose, peak viscosity and breakdown viscosity was SC IIIC but it had the narrowest final
viscosity range. Figure 3.5 depicts the diversity within the five sub clusters as regards
pasting characteristics notably peak, trough, final and setback viscosities. This observation
is similar to the accessions in C I. Notwithstanding, a close similarity was observed in SC
IIIA, SC IIIB, SC IIIC and SC IIIE with reference to breakdown viscosity and peak time.
3.7.4
Cluster IV
23.7% of the entire population analyzed (242 accessions) were in cluster IV and the five
sub clusters SC IVA, SC IVB, SC IVC, SC IVD and SC IVE constituted 33.1%, 38.8%,
17.4%, 5.4% and 5.4% of the accessions, respectively (Table 3.2). Cluster IV had the
broadest range of apparent amylose, the second broadest peak viscosity range and the
fourth broadest ranges of both final and setback viscosities but the narrowest breakdown
viscosity range. The apparent amylose range of this cluster means that the accessions fall
under low, intermediate and high amylose rice. Peak, trough, breakdown, final and setback
viscosities depicted strong diversity in all the five sub clusters except in SC IVE (Figure
3.6), which though was distinct with respect to breakdown viscosity, showed very close
similarity to SC IVB with regards to peak, breakdown, final and setback viscosities. SC
IVA, SC IVB and SC IVC shared some similarities in AAC and peak time.
52
3.7.5
Cluster V
The 231 accessions in cluster V formed 22.6% of the total number of accessions analyzed.
As in clusters I, III, and IV there were five sub clusters in cluster V namely SC VA, SC
VB, SC VC, SC VD and SC VE which constituted 25.5%, 15.2%, 32.9%, 9.1% and 17.3%
of accessions, respectively (Table 3.2). The AAC of accessions in Cluster V ranged from
21 to 28.8% making them intermediate to high amylose rice types. All the sub clusters had
very similar AAC range except for SC VE. Cluster V was with the broadest range of final
viscosity and second broadest range of setback and breakdown viscosities and the third
broadest AAC range. The nature and extent of diversity in the pasting parameters among
the sub clusters in C V (Figure 3.7), generally varied from the other four clusters. Peak,
trough, breakdown, final and setback viscosities depicted diversity in all the five sub
clusters except in SC VE (Figure 3.6) which showed very close similarity to SC VA with
regards to peak and trough viscosities. Additionally, SC VA and SC VD shared some
similarities in AAC as well as SC VB and SC VC.
3.8 Correlations
Table 3.3 shows correlations between pasting parameters and AAC within the 5 clusters of
O. glaberrima rice accessions analyzed. Peak viscosity highly correlated with trough,
breakdown and final viscosities (p<0.01) in all the five clusters. Similarly, peak viscosity
correlated positively with setback viscosity (p<0.01), but correlated negatively with peak
time and pasting temperature (p<0.01) except CII. Within clusters II and IV correlation
between peak viscosity and AAC was not significant (p<0.05). Trough viscosity correlated
positively with peak and final viscosities but exhibited relatively lower correlations with
53
breakdown and setback viscosities (p<0.05) in the five clusters. However, trough viscosity
correlated negatively with peak time and pasting temperature except for clusters II and V
for peak time. As with peak viscosity, correlation with AAC at p<0.01 was not significant,
except for clusters I and III (p<0.05). Breakdown viscosity correlated negatively (p<0.01)
with both peak time and pasting temperature in all the five clusters but on the contrary,
correlated positively with peak and trough viscosities (p<0.01) except in clusters II and V
with trough viscosity (p<0.05). Correlations between breakdown viscosity and setback
viscosity were not significant in clusters I, III and V (p<0.05) and there was also no
significant correlation with final viscosity in clusters II and V.
Breakdown viscosity
correlated negatively with AAC in clusters IV and V but positively in clusters I and III,
however it was not significant for cluster II. Final viscosity correlated highly (p<0.01) with
setback viscosity in all the five clusters while correlation between setback viscosity and
apparent amylose was significant only in cluster I. Additionally, peak time also correlated
positively with pasting temperature in all five clusters.
3.9 Conclusion
This study has shown that there is a high level of genetic diversity in the accessions of O.
glaberrima from the West Africa sub region as evidenced by the variation in both AAC
and pasting properties in and within the clusters and sub clusters. This diversity suggests
that there may be a valuable resource of potentially very important gene pool that may be
of benefit in rice breeding programs in this region. Additionally, a detailed molecular
analysis of a selection of interesting accessions from the clusters such as those with a high
tendency to retrograde (high setback and potentially slower rate of digestion due to
54
retrogradation) or form viscous paste/gel after cooking and cooling (high final viscosities)
will ensure the use of the diversity in the accessions to assist in future development of
acceptable commercial cultivated rice varieties that have the desired amylose content and
pasting characteristics to meet consumer preferences and benefit resource poor farmers
who still depend on rice cultivation for their livelihood.
3.10
Acknowledgements
The Global Rice Science Partnership (GRiSP) and Africa Rice Center (AfricaRice) are
hereby acknowledged for providing funding for this study. Many thanks to the staff of the
Grain Quality Laboratory of the Africa Rice Center, Cotonou, Benin for assisting in
running experiments.
55
3.11
Tables and Figures
Table 3.1. Descriptive Statistics of Entire Population of 1,020 O. glaberrima Accessions
AAC and Pasting Parameters
Minimum value Maximum value
Mean
Apparent Amylose Content (%)
Std. Deviation
15.1
29.6
26.2
1.33
Peak Viscosity (RVU)
36.4
230.7
100.2
29.31
Trough Viscosity (RVU)
36.5
183.5
95.4
23.33
Breakdown Viscosity (RVU)
-0.5
49.5
4.8
7.45
Final Viscosity (RVU)
70.8
421.5
208.6
55.94
Setback Viscosity (RVU)
33.0
224.8
108.4
29.30
Peak Time (min)
5.7
7.0
6.4
0.48
Pasting Temperature (°C)
86.4
95.6
92.0
1.71
56
Table 3.2. Distribution of 1,020 O. glaberrima Rice Accessions Obtained Through Cluster Analysis Using SPSS#
Range
Cluster
I
Distribution (N)
AAC (%)
PV (RVU)
TV (RVU)
BDV (RVU)
FV (RVU)
SBV (RVU)
PT (min)
PTemp (°C)
C I (282)
24.7 - 29.4
67.3 - 123.8
66.7 – 124.2
-0.4 – 6.6
149.6 – 296.8
71.7 – 173.0
5.9 - 7.0
89.6 -94.5
SC IA (44)
25.3 - 29.4
76.8 - 115.6
73.3 – 113.4
1.0 – 6.6
160.1 – 225.4
71.7 – 124.5
5.9 - 6.5
90.7 - 93.6
SC IB (84)
24.8 - 27.9
67.3 - 87.4
66.7 – 87.3
-0.4 – 2.2
149.6 – 188.3
75.0 – 104.3
6.1 - 7.0
91.9 - 94.0
SC IC (107)
24.7 - 28.5
76.3 -101.6
76.6 – 101.8
-0.4 – 0.8
170.4 – 234.9
75.8 – 139.5
6.3 - 7.0
91.9 - 94.5
SC ID (26)
26.0 - 29.0
92.9 - 123.8
93.0 – 124.2
-0.3 – 2.2
217.7 – 296.8
118.0 – 173.0
6.3 - 7.0
91.5 - 93.6
SC IE (21)
25.4 - 27.7
89.4 – 107.9
89.7 – 105.8
-0.2 – 2.2
176.3 – 217.1
86.8 – 116.8
6.4 - 7.0
89.6 - 92.8
152.2 – 230.7
113.0 – 183.5
6.0 – 49.5
246.5 – 421.5
94.9 – 224.8
5.7 - 5.9
86.4 - 91.9
C II (104)
II
SC IIA (26)
23.0 - 29.2
137.8 – 182.7
125.3 – 160.9
6.0 – 32.4
308.1 – 394.8
54.8 – 224.8
5.7 - 6.2
88.3 - 91.9
SC IIB (54)
23.9 – 29.6
135.5 – 165.8
115.9 – 143.0
6.9 – 34.2
246.5 – 327.3
95.3 – 170.7
5.7 - 5.9
87.2 - 91.0
SC IIC (24)
20.6 - 28.5
152.2 – 230.7
113.0 – 183.5
21.9 – 49.5
247.0 – 421.5
94.9 – 190.8
5.7 - 5.9
86.4 - 89.2
80.9 – 125.5
80.4 – 122.2
-0.2 – 12.8
159.9 – 258.8
68.3 – 149.8
5.9 - 7.0
89.6 - 94.0
C III (161)
III
20.6 - 29.6
21.7 - 26.8
SC IIIA (37)
21.7 - 26.8
89.2 – 125.5
88.2 – 122.3
-0.2 – 12.8
228.3 – 258.8
120.8 – 149.8
5.9 - 7.0
90.7 - 93.6
SC IIIB (39)
23.2 - 26.6
89.8 – 110.8
89.8 – 103.1
-0.2 – 9.2
212.1 – 232.8
107.8 – 132.2
5.9 - 7.0
90.7 - 93.6
SC IIIC (43)
22.6 - 26.5
80.9 – 110.0
80.4 – 104.1
-0.2 – 11.8
194.3 – 211.5
92.1 – 119.1
5.9 - 7.0
89.9 - 94.0
SC IIID (18)
24.6 - 26.4
108.2 – 121.3
100.7 – 115.5
2.4 – 10.1
213.6 – 236.1
103.3 – 117.2
5.9 - 6.2
89.6 - 92.0
SC IIIE (24)
23.4 - 26.4
82.9 – 107.9
81.0 – 105.0
0.1 – 6.5
159.9 – 196.6
68.3 – 101.2
6.0 - 7.0
90.3 - 92.4
57
C IV (242)
IV
15.1 - 28.6
#
36.5 – 93.0
-0.5 – 1.2
70.8 – 205.8
33.0 – 122.0
6.3 - 7.0
91.5 - 95.6
SC IVA (80)
22.4 - 27.8
53.6 –73.4
54.0 – 73.8
-0.5 - 0.0
119.8 – 158.4
57.9 – 89.8
7.0 - 7.0
92.8- 95.2
SC IVB (94)
15.1 - 28.6
63.6 – 83.9
63.7 – 84.0
-0.4 – 0.2
135.8 – 180.4
62.3 – 105.0
6.8 - 7.0
91.5 - 94.7
SC IVC (42)
23.0 - 27.6
36.4 – 63.3
36.5 – 63.4
-0.4 - -0.1
70.8 – 118.7
33.0 – 67.8
7.0 - 7.0
92.4 - 95.6
SC IVD (13)
21.3 - 27.4
74.5- 93.1
74.3 – 93.0
-0.2 – 0.4
171.0 – 205.8
81.2 – 122.0
6.6 - 7.0
92.0 - 94.4
SC IVE (13)
21.7 - 25.6
50.8– 83.3
50.4 – 82.9
-0.4 – 1.2
123.4 – 176.0
60.4 – 98.0
6.3 - 6.7
91.9 - 94.0
96.3 – 152.3
87.1 – 141.4
0.8 – 22.2
194.2 – 343.5
84.8 – 197.7
5.7 - 6.4
87.1 - 92.4
C V (231)
V
36.4- 93.1
21.0 - 28.8
SC VA (59)
25.3 -28.8
117.3 – 152.3
104.9 – 139.3
6.3 – 22.2
212.4 – 275.3
84.8 – 137.7
5.7 - 6.0
87.1 - 91.1
SC VB (35)
25.7 - 28.8
114.3 -151.1
108.5 – 141.4
1.6 – 14.8
255.7 – 343.5
135.2 – 197.7
5.8 - 6.3
89.9 - 92.4
SC VC (76)
25.5 - 28.5
100.8 – 131.6
198.3 – 124.8
0.8 – 10.1
209.7 – 268.1
91.3 – 153.2
5.9 - 6.4
89.1 - 92.4
SC VD (21)
25.6 – 28.1
96.3 – 117.2
87.1 – 104.2
3.1 – 22.0
194.2 – 229.4
90.5 – 117.3
5.7 - 6.0
89.3 - 91.5
SC VE (40)
21.0 - 27.5
108.3 – 141.5
95.9 – 127.5
5.7 – 17.0
237.5 – 306.7
111.4 – 167.7
5.7 - 6.2
88.7 - 92.0
Ranges of values
C – Cluster
SC – Sub Cluster
FV – Final Viscosity
PV-Peak Viscosity
TV – Trough Viscosity
BDV – Breakdown Viscosity
SBV – Setback Viscosity
PT – Peak Time
PTemp – Pasting Temperature
AAC - Apparent Amylose Content
58
Table 3.3. Correlations between Pasting Parameters and Apparent Amylose Content within the 5 Clusters of 1,020 O. glaberrima
Accessions#
Breakdown
viscosity
Final
viscosity
Setback
viscosity
Peak
time
Pasting
Temp.
Apparent
Amylose
Content
1
1
1
1
1
0.992**
0.846**
0.955**
1.000**
0.924**
0.339**
0.688**
0.527**
0.400**
0.501**
0.765**
0.660**
0.646**
0.915**
0.723**
0.449**
0.217*
0.269**
0.749**
0.408**
-0.320**
NS
-0.382**
-0.258**
-0.275**
-0.526**
-0.500**
-0.500**
-0.701**
-0.384**
0.426**
NS
0.264**
NS
-0.175**
0.992**
0.846**
0.955**
1.000**
0.924**
1
1
1
1
1
0.218**
0.195*
0.252**
0.385**
0.132*
0.775**
0.780**
0.669**
0.915**
0.776**
0.467**
0.472**
0.323**
0.748**
0.519**
-0.220**
NS
-0.174*
-0.245**
NS
-0.484**
-0.247*
-0.400**
-0.697**
-0.227**
0.398**
NS
0.194*
NS
NS
0.339**
0.688**
0.527**
0.400**
0.501**
0.218**
0.195*
0.252**
0.385**
0.132*
1
1
1
1
1
0.145*
NS
0.190*
0.369**
NS
NS
-2.43*
NS
0.304**
NS
-0.838**
- 0.553**
-0.751**
-0.810**
-0.740**
-0.460**
-0.584**
-0.486**
-0.479**
-0.481**
0.330**
NS
0.308**
-0.244**
-0.319**
0.765**
0.660**
0.646**
0.915**
0.723**
0.775**
0.780**
0.669**
0.915**
0.776**
0.145*
NS
0.190*
0.369**
NS
1
1
1
1
1
0.919**
0.877**
0.909**
0.952**
0.926**
-0.130*
0.218*
NS
-0.249**
NS
-0.194**
NS
NS
-0.526**
0.131*
0.335**
NS
NS
NS
-1.46*
Peak Viscosity
Trough
viscosity
Trough viscosity
Cluster I (N=282)
Cluster II (N=104)
Cluster III (N=161)
Cluster IV (N=242)
Cluster V (N=231)
Peak
Viscosity
Breakdown
viscosity
Cluster,
Parameter
N=1,020
Cluster I (N=282)
Cluster II (N=104)
Cluster III (N=161)
Cluster IV (N=242)
Cluster V (N=231)
N=1,020
Final viscosity
Correlations
N=1,020
Cluster I (N=282)
Cluster II (N=104)
Cluster III (N=161)
Cluster IV (N=242)
Cluster V (N=231)
N=1,020
Cluster I (N=282)
Cluster II (N=104)
Cluster III (N=161)
Cluster IV (N=242)
Cluster V (N=231)
59
Cluster I (N=282)
Cluster II (N=104)
Cluster III (N=161)
Cluster IV (N=242)
Cluster V (N=231)
#
Setback Viscosity
NS
-0.243*
NS
0.304**
NS
0.919**
0.877**
0.909**
0.952**
0.926**
1
1
1
1
1
NS
0.397**
NS
-0.214**
0.215
NS
0.420**
0.394**
-0.335
0.384**
0.204**
NS
NS
NS
NS
Peak time
Cluster I (N=282)
Cluster II (N=104)
Cluster III (N=161)
Cluster IV (N=242)
Cluster V (N=231)
N=1,020
0.467**
0.472**
0.323**
0.748**
0.519**
-0.320**
NS
-0.382**
-0.258**
-0.275**
-0.220**
NS
-0.174*
-0.245**
NS
-0.838**
-0.533**
-0.751**
-0.810**
-0.740**
-0.130*
0.218*
NS
-0.249**
NS
NS
0.397**
NS
-0.214**
0.215**
1
1
1
1
1
0.538**
0.693**
0.509**
0.360**
0.628**
-0.340**
-0.289**
-0.427**
0.236**
0.164*
Pasting Temp.
Cluster I (N=282)
Cluster II (N=104)
Cluster III (N=161)
Cluster IV (N=242)
Cluster V (N=231)
N=1,020
0.449**
0.217*
0.269**
0.749**
0.408**
-0.526**
-0.500**
-0.500**
-0.701**
-0.384**
-0.484**
-0.247*
-0.400**
-0.697**
-0.227**
-0.460**
-0.584**
-0.486**
-0.479**
-0.481**
-0.194**
NS
NS
-0.526**
0.131*
NS
0.420**
0.394**
-0.335**
0.384**
0.538**
0.693**
0.509
0.360**
0.628**
1
1
1
1
1
-0.298**
NS
-0.247**
NS
NS
0.426**
NS
0.264**
NS
-0.175**
0.398**
NS
0.194*
NS
NS
0.330**
NS
0.308**
-0.244**
-0.319**
0.335**
NS
NS
NS
-0.146*
0.204**
NS
NS
NS
NS
-0.340**
-0.289**
-0.427**
0.236**
0.164*
-0.298**
NS
-0.247**
NS
NS
1
1
1
1
1
Apparent
Amylose
Content
N=1,020
Cluster I (N=282)
Cluster II (N=104)
Cluster III (N=161)
Cluster IV (N=242)
Cluster V (N=231)
N=1,020
** Correlation is significant at the 0.01 level (2-tailed)
* Correlation is significant at the 0.05 level (2-tailed)
60
O. glaberrima accessions
(N= 1,020)
SC IA
(N=44)
CI
C II
C III
C IV
CV
(N=282)
(N=104)
(N=161)
(N=242)
(N=231)
SC IB
(N=84)
SC ID
(N=26)
SC IC
(N=107)
SC IIIA
(N=37)
SC IE
(N=21)
SC IIIB
(N=39)
SC IIID
(N=18)
SC VA
(N=59)
SC IIIC
(N=43)
SC VD
(N=21)
SC IIIE
(N=24)
SC IVA
(N=80)
SC IIA
(N=26)
SC IIB
(N=54)
SC IIC
(N=24)
SC VB
(N=35)
SC IVB
(N=94)
SC IVD
(N=13)
SC VC
(N=76)
SC VE
(N=40)
SC VIC
(N=42)
SC IVE
(N=13)
Figure 3.1. Graphical Representation of Distribution of 1,020 African Rice Accessions into 5 Clusters (C) and 23 Sub Clusters (SC).
61
Figure 3.2. Distribution of Entire Population of 1,020 O. glaberrima Accessions
Based on Apparent Amylose Content and Pasting Properties.
62
C= cluster
.
Figure 3.3. Distribution of O. glaberrima Accessions in Cluster I Population Based
on Apparent Amylose Content and Pasting Properties.
63
SC=sub cluster
Figure 3.4. Distribution of O. glaberrima Accessions in Cluster II Population
Based on Apparent Amylose Content and Pasting Properties.
64
SC= sub cluster
Figure 3.5. Distribution of O. glaberrima Accessions in Cluster III Population Based
on Apparent Amylose Content and Pasting Properties.
65
SC= sub cluster
Figure 3.6. Distribution of O. glaberrima Accessions in Cluster IV Population Based on
Apparent Amylose Content and Pasting Properties.
66
SC= sub cluster
Figure 3.7. Distribution of O. glaberrima Accessions in Cluster V Population Based on
Apparent Amylose Content and Pasting Properties.
67
SC= sub cluster
3.12
References
AACC International. Approved Methods of Analysis, 11th Ed. Method 61-03.01.
Amylose content of milled rice. First approval October 15, 1997; Re-approval
November 3, 1999. AACC International: St. Paul, MN.
http://dx.doi.org/10.1094/AACCIntMethod-61-03.01.
Bao, J., Shena, S., Sun, M., and Corke, H. 2006. Analysis of genotypic diversity in the
starch physicochemical properties of non-waxy rice: Apparent amylose content,
pasting viscosity and gel texture. Starch/Stärke 58: 259-267.
Bao, J. S. 2012. Toward understanding the genetic and molecular bases of the eating and
cooking qualities of rice. Cereal Foods World 57: 148-156.
Chakanda, R., van Treuren, R., Visser, B., and van den Berg, R. 2013. Analysis of genetic
diversity in farmers’ rice varieties in Sierra Leone using morphological and AFLP
markers. Genet. Resour. Crop Evol. 60: 1237-1250.
Champagne, E. T., Bett, K. L., Vinyard, B. T., McClung, A. M., Barton, F. E., II,
Moldenhauer, K., Linscombe, S., and McKenzie, K. 1999. Correlation between
cooked rice texture and Rapid Visco Analyser measurements. Cereal Chem.
76:764-771.
Futakuchi, K., Fofana, M., and Sié, M. 2008a. Varietal differences in lodging resistance of
African rice (Oryza glaberrima Steud.). Asian J. Plant Sci. 7: 569-573.
68
Futakuchi, K. and Sié, M. 2009. Better exploitation of African Rice (Oryza glaberrima
Steud.) in varietal development for resource-poor farmers in West and Central
Africa. Agric. J. 4: 96-102.
Jones, M.P., Dingkuhn, M., Aluko, G. K. and Semon, M. 1997. Interspecific Oryza sativa
L. x Oryza. glaberrima Steud. progenies in upland rice improvement.
Euphytica 92: 237-246.
Juliano, B. O. 1998. Varietal impact on rice quality. Cereal Foods World 43: 207-222.
Lawal, O. S., Lapasin, R., Bellich, B., Olayiwola, T. O., Cesàro, A., and Yoshimura, M.
2011. Rheology and functional properties of starches isolated from five improved
rice varieties from West Africa. Food Hydrocoll. 25: 1785-1792.
Lee, I., Gyoung, We, G. J., Kim, D. E., Cho,Y.-S., Yoon, M-R., Shin, M., and Ko,
S. 2012. Classification of rice cultivars based on cluster analysis of hydration and
pasting properties of their starches. Food Sci. Technol. 48: 164-168.
Linares, O. F. 2002. African rice (Oryza glaberrima): History and future potential. Proc.
Natl. Acad. Sci. U.S.A. 99: 16360-16365.
Nayar, N.M. 2010. The history and genetic transformation of the African rice, Oryza
glaberrima Steud. (Gramineae). Curr. Sci. 99: 1681-1689.
Ndjiondjop, M. N., Manneh, B., Cissoko, M., Drame, N.K., Kakai, R. G., Bocco, R.,
Baimey, H., and Wopereis, M. 2010. Drought resistance in an interspecific
backcross population of rice (Oryza spp.) derived from the cross WAB56-104 (O.
sativa) × CG14 (O. glaberrima). Plant Sci. 179: 364-373.
69
Roferos, L. T., Felix, A. dR., and Juliano, B. O. 2006. The search for the grain quality of
raw and cooked IR 64 milled rice among Philippine Seed Board rice varieties.
Philipp. Agric. Sci.: 89:58-70.
Roferos, L. T., and Juliano, B. O. 1997. Chemometrics of grain quality of raw and cooked
Philippine milled rice. Philipp. Agric. Sci.: 80: 211-236.
Sarla, N., and Swamy, B. P. M. 2005. Oryza glaberrima: a source for the improvement of
Oryza sativa. Curr. Sci. 89: 955-963.
Singh, A., Singh, B., Panda, K., Rai, V. P., Singh, A. K., Singh, S. P., Chouhan, S. K., Rai,
V., Singh, P. K., and Singh, N. K. 2013. Wild rices of Eastern Indo-Gangetic plains
of India constitute two sub-populations harbouring rich genetic diversity. Plant
Omics J. 6: 121-127.
Sweeney, M., and McCouch, S. 2007. The complex history of the domestication of rice.
Ann. Bot. 100: 951-957.
Tan, Y., and Corke, H. 2002. Factor analysis of physicochemical properties of 63 rice
varieties. J. Sci. Food Agric. 82:745-752.
Watanabe, H., Futakuchi, K., Jones, M. P., Teslim, I., and Sobambo, B.A. 2002a.
Brabender viscogram characteristics of interspecific progenies of Oryza glaberrima
Steud. and O. sativa L. Nippon Shokuhin Kagaku Kogaku Kaishi. J. Jpn. Soc.
Food Sci. Technol. 49: 155-165.
70
Watanabe, H., Futakuchi, K., Teslim, I., and Sobambo, B. A. 2002b. Milling
characteristics and grain quality traits of interspecific progenies of Asian rice
(Oryza sativa) with African rice (Oryza glaberrima). Jpn. J. Tropical Agric. 46:
47-55.
Zhang, C.-H., Li, J.-Z., Zhu, Z., Zhang, Y.-D., Zhao, L., and Wang, C.-L. 2010. Cluster
analysis on Japonica rice (Oryza sativa L.) with good eating quality based on SSR
markers and phenotypic traits. Rice Sci.17: 111-121.
71
CHAPTER 4: PHYSICOCHEMICAL PROPERTIES, COOKING QUALITY AND
GLYCEMIC ATTRIBUTES OF AFRICAN RICE (Oryza glaberrima)
72
Abstract
Seven dehusked and polished African rice accessions (ARAs) were evaluated for their
grain quality attributes based on physicochemical, cooking quality, textural, pasting,
thermal and in vitro digestion properties. They were compared with two O. sativa (Asian
rice) varieties; Koshihikari (spp. japonica) and WITA 4 (spp.indica) and NERICA 4 (an
inter-specific cross between sativa x glaberrima). Based on amylose content the ARAs
examined were intermediate to high amylose rice type while NERICA 4 and the Asian rice
varieties were low amylose type. The thousand kernel weight and bulk density of the
ARAs were comparable to the Asian rice varieties which can be classified as mediumshaped kernels based on ISO classification system except for TOG 12440 which was bold.
Volume expansion ratio of the cooked kernels ranged between 2.5 and 3.5 for all samples
except for NERICA 4, which was significantly (p<0.05) higher (4.4). The ARAs had
significantly (p<0.05) lower cooking solids loss and harder texture compared to the Asian
rice varieties which were softer and more adhesive. The peak viscosities of the Asian rice
flours were significantly (p<0.05) higher but showed a higher breakdown tendency as
compared to the ARAs. Gelatinization transition temperatures (To, Tp, Tc) and enthalpy
were significantly (p<0.05) higher in ARA flours as compared with Asian rice. However,
melting transition temperatures of retrograded flour gels, were similar for all rice varieties,
while the melting enthalpy of the ARAs were higher than that of Asian rice. The expected
glycemic indices of both African and Asian rice were higher than 70. The study results
show differences in cooking characteristics between African and Asian rice varieties which
may be partially due to the high amylose content of the ARAs.
Keywords: amylose, texture, pasting, African rice, glycemic index
73
4.1 Introduction
One of the key cereal crops that is put under vast cultivation worldwide and serves as a
primary source of food for more than two-thirds of the world’s population is rice
(Ndjiondjop et al. 2010). Oryza sativa (Asian rice) and Oryza glaberrima (African rice)
are the two main types of domesticated rice that are cultivated by farmers globally (Linares
2002) but African rice which is indigenous to the West African sub region has seen less
development (Sarla and Swamy 2005; Sweeney and McCouch 2007). One of the main
advantages attributed to African rice in comparison to Asian rice is its resistance to a
number of biotic stresses such as diseases, insects and nematodes and abiotic stresses such
as drought, flooding and salinity.
Other advantages cited by producers include its
suitability for weaning foods, high protein content and good keeping quality of cooked
grains (Sarla and Swamy 2005). Nonetheless it is characterized by low yield and high
tendency to shatter, a situation that leads to high post-harvest losses. Conventional milling
also leads to high levels of broken rice because it breaks easily (Nayar2010; Watanabe et
al. 2002; Sarla and Swamy 2005). In spite of these challenges, some African farmers still
crop African rice because of its adaptability, ceremonial and cultural importance (Falade et
al. 2014). African rice can therefore be a valuable genetic source for breeding (Aluko et al
2004; Li et al. 2004; Watanabe et al. 2004) and a potential source of genes to enhance the
milling, cooking and eating quality of Asian rice (Sarla and Swamy 2005).
In the past, a rice variety (NERICA) was developed successfully by crossing high yielding
Asian rice lines with African rice lines that had some desirable characteristics.
The
adaptability of African rice to local climatic conditions and the high yield of Asian rice
74
were the underlying factors for the success of the new variety. Notwithstanding this
success, it is believed that African rice possesses other traits that can be tapped for future
grain improvement programmes. These include amylose content, which is a major factor
that influences texture and taste of cooked rice (Futakuchi and Sié 2009; Zhang et al.
2010). The Africa Rice Center (AfricaRice) has been systematically characterizing a
collection of African rice accessions that there may be a valuable resource of potentially
very important gene pool that may be of benefit in rice breeding programs. To support the
broader objective of this characterization process, this study seeks to evaluate the cooking
quality attributes of some African rice accessions that were selected from 1,020 rice
accessions based on their amylose and pasting data using Ward method of hierarchical
cluster analysis. Rice grain quality has been reported to be usually classified into four
components namely: milling efficiency, grain shape and appearance, cooking and edibility
characteristics and nutritional quality (Li et al. 2004). Grain quality characteristics vary
remarkably with regards to different subspecies and varieties of rice. This provides the
basis for categorising rice types that are preferred for ceremonial purposes or for use in
specific products such as flour and noodles (Li 2003). Additionally, consumers in different
areas of the world prefer different types of rice for general consumption purposes.
Characterizing quality attributes of different types of rice varieties is important for the
improvement in rice quality and rice-based products.
Two essential indicators that are commonly used to describe starch digestibility in relation
to health benefits are glycemic index (GI) and resistant starch (RS) content. Starch can be
categorized into three nutritional fractions, namely rapidly digestible (RDS), slowly
digestible (SDS) and resistant starch (RS). Rapidly digestible starches refer to starches that
75
are digested after 20 minutes while slowly digestible starches are those digested between
20 minutes and 120 minutes. Resistant starches are those that remain undigested after 120
minutes (Englyst et al. 1992). GI is a scale that is used to rank carbohydrate-rich foods by
their ability to raise blood glucose level when ingested compared to a reference food
(glucose) or white bread (Jenkins et al. 1987). The predominant type of rice consumed
worldwide is the polished or white rice (Nanri et al. 2010) and it is obtained by subjecting
paddy rice through a series of mechanized unit operations. The glycemic index value of a
specific white rice variety is influenced by the degree of processing, cooking time, and
amylose content (Foster-Powell et al. 2002). Different varieties of rice may therefore have
different GIs (Miller et al. 1992; Atkinson et al. 2008). GIs of cooked rice varieties
ranging from 47 (Ranawana et al. 2009) to 109 (Frei et al. 2003) have been previously
reported. In this study, the GIs of the samples are being investigated to ascertain farmers’
indirect claim of African rice having low GI.
In addition, the selected African rice
accessions were investigated in terms of physical and physicochemical properties of grains
and flours. Information generated from this study may contribute to the selection of
material for the development of suitable African rice cultivars to meet diverse consumer
preferences.
4.2 Materials and Methods
4.3 Materials
Ten dehusked and polished rice samples were provided by the Africa Rice Center based in
Cotonou, Benin for this study. Seven of the samples were African rice accessions (ARAs)
- IRGC 86741, IRGC 80787, IRGC 103759, RAM 123, TOG 7682, TOG 12440 and TOG
76
14365. The others are two Asian rice varieties – cv Koshihikari (spp. japonica) and
cv. WITA 4(spp. indica) and NERICA 4 (an interspecific cross between African rice x
Asian rice).
The seven ARAs were chosen from a population of 1,020 African rice
accessions from West Africa based on data of their pasting characteristics and apparent
amylose content using hierarchical cluster analysis technique. Portions of the samples
were milled into flour using UDY cyclone mill (Fort Collins, Co., USA) fitted with a sieve
of 0.5 mm mesh size.
4.4 Methods
Whole rice kernels were separated for evaluation of physicochemical, thermal, cooking and
in vitro digestibility properties.
4.4.1
Physicochemical Properties
4.4.2
1000-Kernel Weight
The method of Singh et al. (2005) was used for this determination.
4.4.3
Bulk Density
Whole rice kernels were transferred into a cup of known volume from a fixed height. A
straight-edge was thereafter used to level off the rice at the top of the cup. The bulk
density was calculated as the ratio of the weight of rice in the cup to the volume of the cup
and reported as g/mL.
77
4.4.4
Length-Width Ratio
Length-width ratio was measured using a digital caliper with an accuracy of 0.01mm. Fifty
head rice kernels of each sample were randomly selected and measurements of their
lengths and widths were manually taken.
4.4.5
Amylose Content
The apparent amylose content of the rice flour was determined in duplicate by using AACC
approved method 61-03 (AACC 2000).
4.4.6
Nitrogen
The Dumas combustion method was employed to measure the nitrogen content of rice flour
samples. Weighed rice flour samples (0.2 g) were combusted in a LECO Nitrogen/Protein
Analyzer (LECO, FP-528 Lakeview Avenue, St. Joseph, Michigan, United States) after
ethylenediaminetetraacetic acid (EDTA) standard was run to calibrate the equipment.
Nitrogen was converted into protein by multiplying the nitrogen determined by a factor of
5.7.
4.4.7
Total Starch
The total starch kit supplied by Megazyme (Wicklow, Ireland) was used for the
determination of total starch in the rice flour samples. Total starch was quantified in g/100
g of flour on dry weight basis.
78
4.4.8
Colour Measurement of Rice Kernels
Rice kernel colour was measured with a Minolta Chroma Meter (CM 3500-d, Konica
Minolta Sensing Inc., Mahwah, NJ, USA). The Minolta Chroma Meter, allocates each
sample a colour co-ordinate within the CIE (Commission Internationale l'Eclairage) L-a-b
three-dimensional (L*a*b*) colour space. CIE L* values represent brightness on a 0 (pure
black) -100 (pure white) unit scale, CIE a* values range from -60 (pure green) to +60 (pure
red), and CIE b* values range from -60 (pure blue) to +60 (pure yellow). Each CIE L*, a*,
b* value printed is averaged from three readings.
4.4.9
Cooking Characteristics
4.4.10 Optimum Cooking Time
The optimum cooking time (OCT) was determined using the method outlined by Juliano
and Bechel (1985) and modified by Mohapatra and Bal (2006). Approximately 100 mL of
distilled water was boiled in a beaker and 5 g of head rice samples was placed in it. Ten
minutes into boiling and every minute thereafter, 10 grains were removed and pressed
between two clean microscope glass slides. The appearance of a chalky/opaque core of the
grain was an indication of partially cooked grain. When at least 90% of the grains no
longer had uncooked cores, the sample was deemed to be optimally cooked. The sample
was then allowed to simmer for an additional 2 min to allow for the core of all grains to
gelatinize. OCT included the 2 min simmering time. Volume expansion, length expansion
and water uptake ratios (VER, LER, WUR respectively) were determined by placing 1 g of
head rice in 15 mL of boiling water till the OCT was reached. VER was reported as the
ratio of volume of cooked grains to the volume of uncooked grains while LER was the ratio
79
of the length of cooked grain to length of uncooked grain. WUR was the ratio of water
absorbed by the grains during cooking to the initial weight of uncooked rice. Solids loss
(SL) was determined by optimally cooking 2 g of head rice in 20 mL of distilled water and
transferring the gruel into a dry pre-weighed aluminium dish. The dish and gruel were
placed in an oven (100 °C) for 24 h to remove moisture. SL was calculated as the ratio of
increase in weight of dish to weight of rice sample. The average of three replications was
reported.
4.4.11 Textural Profile Analysis
Textural profile analysis (TPA) of the cooked rice was performed using a texture analyzer
(TA Exponent 32 Stable Micro Systems version 5,0,3,0, Texture Technologies Corp.,
Scarsdale, New York, USA) with a 5 kg load cell and a two-cycle compression according
to the procedure by Mohapatra and Bal (2006).
The rice samples were prepared by
optimally cooking 10 g in 200 mL of distilled water (till white core in grains disappeared).
At the end of the cooking, excess water was completely drained by employing a strainer
and the residual moisture on the grains blotted out. While the samples were still hot, grains
from each sample were selected onto the base of the equipment for testing and a two-cycle
compression force versus time program was used to compress samples. The probe was
programmed to compress the samples till 90% of the original cooked grain thickness,
return to its original position and then compress samples again (Juliano et al. 1984;
Meullenet et al. 1997). The probe was made to compress 5 grains arranged parallel to each
other with a pre-test and post-test speed of 1 mm/sec and a test speed of 0.5 mm /sec. The
parameters recorded from the test curves were hardness, adhesiveness and springiness.
80
Hardness is defined as the maximum peak force during the first compression cycle while
adhesiveness is the negative force area for the first bite and represents the work required to
overcome the attractive forces between the surface of a food and the surface of other
materials with which the food comes into contact. Springiness is defined as how well a
product physically springs back after it has been deformed during a first compression. The
spring back is measured at the down stroke of a second compression. The values reported
are means of ten replications.
4.4.12 Pasting Properties
The pasting properties of rice flour samples were measured using Rapid Visco Analyzer
RVA-4, (Newport Scientific Pty Ltd., Warriewood, NSW, Australia). 3 g of starch (12%
moisture basis) from each rice sample was weighed directly into the aluminium canister
and mixed with 25 mL of distilled water resulting in a 9.5% suspension. The test profile
involved equilibrating the sample at 50°C for 1 min, heating to 95°C at a rate of 12 °C/min,
holding at 95 °C for 2.5 min, cooling to 50 °C at a rate of 12 °C/min and holding at 50 °C
for 2.5 min. The rotating speed of the paddle was maintained at 160 rpm throughout the
run except for the first 10 s where paddle speed was held at 960 rpm. The under listed
pasting parameters were determined from the RVA pasting curve: peak viscosity
(maximum viscosity attained during heating), hot paste viscosity (trough) and final
viscosity or cold paste viscosity (viscosity at end of cooling period). Breakdown was
calculated as the difference between peak and trough viscosities while setback was
calculated as the difference between the final viscosity and the peak viscosity.
81
4.4.13 Differential Scanning Calorimetry
TA Instruments, Q2000 differential scanning calorimeter (DSC) equipped with a thermal
analysis data and recording software facility (TA Instruments, Universal Analysis 2000,
DE, USA) was used to measure the gelatinization properties of the rice flours. Flour
dispersion in water (1:5) was allowed to equilibrate for at least 3 h at room temperature
before sample was loaded. In all measurements, the thermogram was recorded with an
empty aluminum pan as a reference and the scanning temperature range was programmed
from 20-120 °C at a heating rate of 10 °C/minute. Scanned samples were stored at 4 °C for
14 days and then re-scanned using the same temperature profile. The onset (To), peak (Tp)
and conclusion (Tc) transition temperatures were reported while the enthalpy of
gelatinization (ΔH) was estimated by integrating the area between the thermogram and a
base line under the peak and was expressed as J/g of dry flour. Determinations were done
in duplicates.
4.4.14 In vitro Digestibility and Expected Glycemic Index of Optimally Cooked Rice
In vitro digestibility was conducted on optimally cooked rice kernels using the method
outlined by (Englyst et al. 1992). The ten rice samples were optimally cooked as per their
OCTs, freeze-dried and milled using UDY mill (Fort Collins, CO, USA) fitted with a 0.5
mm mesh. A fresh blend of three enzymes namely pancreatin from porcine pancreas
(Sigma-Aldrich P-1625, with activity 3x USP/g), invertase from baker’s yeast Saccharomyces sereviciae (1450, Sigma-Aldrich) and amyloglucosidase (Megazyme
International, Ireland with activity 200 U/mL p-nitrophenyl β-maltoside) was prepared
according to Englyst et al. (1992) protocol for the hydrolysis of cooked rice flour. Based
82
on their dry weight the flour samples weighed for the hydrolysis process contained 0.7 g of
starch. 10 mL of 0.1 M sodium acetate buffer (pH-5.2) was initially added to the weighed
samples and incubated in a water bath (37 °C) for 5 min after which 5 mL of the enzyme
mix was added. The sample-enzyme mixture was then incubated for 2 h and from start, 0.1
mL aliquots were pipetted from the reaction mass every 20 min into Eppendorf tubes
containing 0.9 mL of 80% ethanol to halt the hydrolysis reaction. The amount of glucose
in these aliquots was determined by glucose oxidase/peroxidase assay (GOPODMegazyme) and converted to starch by multiplying by a factor of 0.9. The hydrolyzed
starch (expressed on dry weight basis) was categorized into rapidly digestible starch
(RDS), slowly digestible starch (SDS) and resistant/residual starch (RS) according to the
description of (Englyst et al. 1992). RDS represents glucose released at 20min x 0.9; SDS
as (glucose released at 120min - glucose released at 20min) x 0.9 and RS as total starch (RDS+SDS). The kinetics of hydrolysis of the flour samples was described by the nonlinear first-order equation C= C∞ (1- ℮-kt) developed by Goni et al. (1997). C is the starch
concentration hydrolyzed at a time t; C∞ the equilibrium concentration at the final time (120
min); k is the kinetic constant and t is the chosen time. The hydrolysis index (HI) was
obtained by dividing the area under the hydrolysis curve (AUC) of the samples by that for
the white bread as a reference sample as reported by Goni et al. (1997). The AUC was
governed by the equation:
AUC= C∞ (tf-to) - (C∞/k) [1-e-k (tf-to)]. Where tf is the final time and to is the initial time. The
eGI was calculated from the equation: eGI= 8.198+0.862 (HI) as described by Granfeldt et
al. (1992).
83
4.5 Statistical Analysis
The statistical programme Statgraphics Centurion XV, Version 15.1.02 (StatPoint,
Warrenton, VA, USA) was employed to conduct statistical analysis. One-way analysis of
variance was performed to determine significant differences between any pair of means at
the 95 % confidence level and multiple range tests were done to determine which means
were different.
4.6 Results and Discussion
4.7 Physicochemical Properties
The physicochemical properties of the rice samples studied are presented in Table 4.1. The
thousand-kernel weight which is a measure of seed/grain size ranged from 15.9 to 20.0 g.
NERICA 4 had the biggest seed size (20.0g) while TOG 7682 had the smallest seed size
(15.9g). The bulk densities of the samples were quite similar but that of Koshihikari
(0.93 g/mL) was the highest. The length-width (L:W) ratio which signifies grain size and
shape ranged from 1.71 to 3.56 and corresponded to the two Asian rice varieties
(Koshihikari and WITA 4 respectively). The apparent amylose content of the samples
showed significant variations (p<0.05) among the samples with the ARAs and NERICA 4
having significantly higher proportions (p<0.05) than the Asian rice varieties (Table 4.1).
Milled rice cultivars may be generally classified on the basis of their amylose content into
waxy (1-2%), very low (2-12%), low (12-20%), intermediate (20-25%) or high (>25%)
(Bao 2012). Thus, by this classification the ARAs belong to intermediate to high amylose
rice types while NERICA 4 and the Asian rice varieties are low amylose types. It was
similarly observed that the crude protein content which ranged from 5.1 to 11.3% were
84
higher in all the ARAs and NERICA 4 than the Asian rice varieties. Total starch content of
the ARAs and the Asian rice varieties were similar and ranged from 82.5 to 87.3 %. The L
index of the color measured signifies lightness (whiteness) and values indicate that the
appearances of the ARAs were comparable to the Asian rice varieties.
4.8 Cooking and Textural Properties
The cooking and textural characteristics of the ten rice samples are presented in Table 4.2.
The optimum cooking time (OCT) of the samples varied from 16 to 23 min. TOG 7682
which recorded the least bulk density (0.86 g/mL) and least 1000-kernel weight (15.9g)
also had the least cooking time of 16 min. Similarly, Koshihikari which had the highest
bulk density (0.93 g/mL) recorded the longest cooking time of 23 min.
Significant
differences were found among the samples in the loss of solids during cooking but it was
generally observed that the Asian rice varieties had higher solids loss.
Significant
variations were observed among the rice samples with respect to water uptake ratio,
volume expansion ratio (VER) and length expansion ratio (LER). Hardness in texture test
was observed to be significantly higher (p<0.05) in the ARAs which interestingly had
higher amylose concentration. In contrast to hardness, the adhesiveness values were found
to be higher in the Asian rice samples. Springiness values were marginally higher in most
of the ARAs and NERICA than in Asian rice varieties.
4.9 Pasting and Thermal Characteristics
Figure 4.1 and Table 4.3 depict the RVA profiles and pasting characteristics of the rice
flour samples. The samples had a peak time ranging from 5.5 min to 7.0 min. The peak
85
viscosity of the flour samples ranged from 48.5 to 271.6 RVU and were significantly
(p<0.05) lower in the ARAs. The trough viscosity, which is a measure of the ability of the
paste to withstand breakdown during the heating phase, varied widely among the samples
while final viscosity was significantly higher in the Asian rice samples than most of the
ARAs.
Peak viscosity indicates the water-holding capacity of a sample, while final
viscosity, the most commonly used parameter to define a particular sample’s quality,
indicates the ability of the material to form a viscous paste or gel after cooking and cooling.
While the ARAs showed stability, breakdown viscosity was higher in the Asian rice
varieties. The setback viscosity, which is a measure of the tendency of starch paste to
retrograde, varied widely among the samples.
DSC thermograms of flour from the rice samples are shown in Figure 4.2 and melting
parameters of the same are presented in Tables 4.4 and 4.5. The onset of gelatinization
temperatures of the ARAs and NERICA 4 were significantly higher than Asian rice
varieties. A similar trend of the ARAs and NERICA 4 having significantly (p<0.05) higher
melting temperatures than the Asian rice varieties was observed for both Tp and Tc. The
gelatinization temperature range (Tc-To) of the samples was narrow and the gelatinization
enthalpies (ΔH) defined as the energy required to unwind the double helices in amylopectin
range was also narrow (6.1-10.1) J/g. The ΔH values of the ARAs were significantly
higher than NERICA 4 and the two Asian rice varieties suggesting that the amylopectin
crystals of the ARAs are more stable in comparison to the Asian rice (Vamadevan et al.
2013). Table 4.5 shows the melting temperatures of the retrograded flour gels. To, Tp, Tc
and ΔH values were considerably lower than their corresponding flours. In addition, To, Tp
and Tc showed very narrow ranges but with regards to the retrograded flour gels ΔH values
86
of the ARAs and NERICA 4 were significantly higher than the two Asian rice samples
suggesting more stability even in the retrograded state.
4.10
In vitro Digestibility of Cooked Rice Samples
Figure 4.3 shows the in vitro starch hydrolysis profiles of the optimally cooked and freezedried rice samples. The hydrolysis trend was similar for all the samples and TOG 14365
had the highest rate of hydrolysis compared with the other samples until about 100 min
while TOG 7682 had the lowest rate. Approximately 76-86 % of the starch in the samples
had been hydrolyzed by 120 min of hydrolysis. This is comparable to 70 % that was
reported for rice for the same time of hydrolysis (Zhang et al. 2006). The quantum of RDS
in the cooked rice samples during hydrolysis ranged from 41.3 % to 51.4 % and
corresponded to IRGC 80787 and RAM 123 (Table 4.6). The levels of SDS in the samples
were generally lower than the levels of RDS and similarly RS levels were lower than the
RDS. Table 4.7 shows the kinetic constant, hydrolysis index and eGI of the cooked and
freeze-dried rice. The hydrolysis constants and the eGI of the ARAs were comparable to
the Asian rice varieties despite the variations observed.
4.11
Discussion
Grain size is an important quality trait in rice and also a major determinant of grain weight
(Fan et al. 2006). Grain weight is important to farmers because it is one of the most stable
components of yield (Li et al. 2004). Grain weight, which is commonly denoted by
1,000- kernel weight in breeding applications, is determined by the length, width and
thickness of the grain (Evans 1972). Preference for rice grain characteristics vary with
consumer groups and rice grain shape has been reported to play a major role in consumer
87
preferences in different countries and populations (Kesavan et al. 2013). The range of
1000- kernel weights observed in this study was comparable to what was reported for a
range of popular rice varieties in Malaysia (Thomas et al. 2103) and some Indian rice
cultivars (Singh et al. 2005). The bulk densities observed for the samples in this study
were also similar to the range reported by Singh et al. (2005) but the length:width ratios for
the ARAs were lower. The grain length and shape of rice are important characteristics to
consumers because they determine the physical appearance and affect the cooking quality
of the grain. Based on the International Standards Organization (ISO), the rice samples in
this study were classified as slender, medium or bold (short). While NERICA 4 and all the
ARAs except TOG 12440 were medium, Koshihikari was bold/short and WITA 4 was
slender.
The ARAs were significantly higher in amylose content in comparison to NERICA 4 and
the Asian rice varieties. Differences in amylose content of different rice cultivars may be
attributed to environmental factors, genotype and cultural practices as well as climatic and
soil conditions (Kim and Wiesenborn 1995; Morrison and Azudin 1987). Amylose is also
considered to be one of the most important predictors of the eating quality of cooked rice
and various rice products (Juliano 1998; Bao 2012). Additionally, the content of amylose
in rice is a major factor that affects texture and taste of cooked rice (Zhang et al. 2010;
Futakuchi and Sié, 2009).
It has been reported that rice that has high amylose
concentration upon cooking exhibits high volume expansion, becomes less tender and
harder when it cools while low amylose rice cooks moist and sticky (Juliano 1985). The
optimum cooking time showed a negative correlation with amylose content (r=-0.6399,
p<0.05) and a weak positive correlation with bulk density of milled rice samples (r=0.5192,
88
p<0.05) (Table 4.8). The relationship between bulk density and cooking time suggested
that the rice samples with higher bulk density exhibited slower rate of water uptake which
eventually caused a longer cooking time.
The samples with higher amylose content
required relatively shorter cooking time. Amylose content and bulk density thus influenced
the optimum cooking time of the rice samples. This was consistent with what was reported
by Singh et al. (2005). Amylose correlated positively with hardness and springiness but
negatively with adhesiveness. The positive correlation between amylose and hardness of
cooked rice is consistent with what was reported by Singh et al. (2005) and Yu et al.
(2009). Cooked rice that has high amylose concentration easily hardens whereas cooked
rice with lower amylose content hardens at a much slower rate (Yu et al. 2009). Amylose
has been reported to leach out during cooking to form a coated film on rice grains and the
rapid retrogradation of amylose results in the hardness of cooked rice grains
(Leelayuthsoontorn and Thipayarat, 2006). High amylose rice types have been reported to
have a dry, firm, and fluffy texture when cooked (Noda et al. 2003; Mohapatra and Bal
2006).
Peak viscosity, the maximum viscosity attained by gelatinized starch during heating in
excess water, is determined by granular swelling, amylose leaching and friction between
granules and it also signifies water binding capacity of starch granules (Shimelis et al.
2006). The granular swelling behavior of cereal starches is primarily a property of their
amylopectin content (Tester and Morrison 1990), and since the Asian rice varieties had
higher amylopectin content than the ARAs it was expected that the former will have higher
peak viscosity. The final viscosity which indicates the ability of the material to form a
viscous paste or gel after cooking and cooling was similar in NERICA 4 and the Asian rice
89
varieties but relatively lower in the ARAs except IRGC 80787 and TOG 12440.
Breakdown viscosity measures the susceptibility of the cooked starch to fragmentation and
that the higher breakdown viscosity, the lower the ability of the sample in question to
withstand heating and shear stress during cooking (Adebowale et al. 2005). The low or
(negative) breakdown of the ARAs seems plausible due to the high enthalpy of
gelatinization (ΔH) indicating a stable structure (Table 4.4). This suggests that the high
amylose content of the ARAs may have restricted the extent of granular swelling and that
granules were not fully disintegrated, hence their lower peak viscosities. Setback viscosity
is commonly used to describe the increase in viscosity that occurs on cooling a starch paste
(Fisher and Thompson 1997) and it determines the retrogradation tendency of the paste
(Zaidul et al. 2007). Total setback viscosity did not show any particular trend among the
ARAs and Asian rice samples. The differences in To, Tp, Tc and ΔH in starches from
different rice cultivars may be attributed to differences in granular architecture and glucan
composition (Vamadevan and Bertoft, 2014).
Based on studies that used rice starches with a wide range of amylose concentration, it has
been established that the extent of digestibility of rice starch is related to amylose content
(Chung et al. 2010; Frei et al. 2003; Zhu et al. 2011). Additionally, factors such as extent
of granule swelling after cooking, inhibition of granule swelling by amylose-lipid complex
and the speed and extent of retrogradation of amylose molecules are important. However,
it has been also reported that rice varieties with similar amylose concentration can possess
different physicochemical (gelatinization) properties which can affect starch digestion rate
and blood glucose. The ARAs due to their higher amylose content were expected to show
a much lower eGI than the Asian rice varieties but this was not observed. Consequently
90
using amylose content alone to predict starch digestibility and glycemic response can be
inaccurate (Panlasigui et al. 1991).
Table 4.9 shows the correlations between amylose and OCT and thermal and pasting
properties of the flour samples. Correlation between OCT and the thermal parameters as
well as the pasting parameters measured were not significant (p<0.05) except for peak and
trough viscosities. Amylose correlated positively with the enthanlpy of gelatinization and
all the transition temperatures. Correlation was not significant (p<0.05) between amylose
and final and total setback viscosities. However, amylose correlated negatively with peak,
trough and breakdown viscosities but positively with setback viscosity. Multiple linear
regression models to describe the relationship between amylose, OCT and cooking and
textural parameters are presented in figure 4.4.
4.12
Conclusion
This study established the African rice accessions as medium shaped high amylose rice
type with high protein content. Their relatively higher amylose concentration influenced
some of the cooking quality and textural parameters measured. This reflected in the ARAs
having relatively harder texture of cooked grains. The high enthalpy of gelatinization (ΔH)
of the ARAs indicates a more stable structure that was relatively more resistant to granule
swelling to preserve the granule integrity and exhibit shear stability. This resulted in
negligible breakdown. Additionally, the observed differences between the ARAs and the
Asian rice varieties in the properties are likely due the differences in the amylose
concentration. The high amylose did not however influence the estimated GI of ARAs.
91
4.13
Acknowledgements:
A special appreciation is hereby extended to the Global Rice Science Partnership (GRiSP)
and Africa Rice Center (AfricaRice) for the provision of funding and rice samples for this
work.
92
4.1 Tables and Figures
Table 4.1. Physicochemical Characteristics of African and Asian Rice Kernels and Flours#
Accession
BD
(g/mL)
MC (%)
TK W(g)
IRGC 80787
5.07bcde
17.63e
0.87bc
IRGC 86741
4.09ab
16.73c
IRGC 103759
4.55bc
RAM 123
L:W
Colour
a*
AC (%)
P (%)
TS (%)
2.15c
28.73d
8.18g
85.20c
65.3
2.2
13.8
0.86ab
2.35d
25.47c
11.29j
84.34b
59.4
4.3
16.5
16.82c
0.88cd
2.14c
28.23d
7.79f
84.83c
65.8
3.8
17.0
5.36de
19.09g
0.90e
2.14c
29.72de
7.55e
85.06c
69.7
1.3
13.7
TOG 7682
5.18bcde
15.94a
0.86a
2.22c
28.51d
10.35i
82.49a
64.3
1.0
14.7
TOG 12440
5.59e
16.26b
0.86ab
1.91b
28.25d
7.22d
84.32b
61.9
3.7
14.7
TOG 14365
5.25cde
17.36d
0.86ab
2.16c
31.07e
9.42h
84.10b
59.9
4.7
17.0
L
*
b*
African rice
O. sativa x O. glaberrima cross
4.58abc
20.02h
0.90de
2.75e
23.76b
6.96c
86.43e
66.4
0.9
16.3
Koshihikari
4.69abcd
18.11f
0.93f
1.71a
17.00a
5.06a
85.92d
66.0
-1.3
12.7
WITA 4
3.59a
17.35d
0.87abc
3.56f
22.88b
6.16b
87.31f
69.5
-1.1
11.8
NERICA 4
Asian rice
MC - Moisture Content
TKW – Thousand Kernel Weight
BD - Bulk Density
L/W - Length/Width Ratio AC - Amylose Content
P- Protein
#
Values with the same letters are not significantly different within columns
93
TS – Total Starch
Table 4.2. Cooking and Textural Properties of African and Asian Rice Kernels#
OCT
SL
Hardness Adhesiveness Springiness
WUR
VER
LER
(min.)
(%)
(g)
(g.sec)
(mm)
African rice
IRGC 80787
18.0
3.25abc
3.00b
3.00bc
1.43abc
6039ef
39.4b
0.74d
IRGC 86741
19.0
2.96a
2.61a
3.25cde
1.42ab
5937def
0.8a
0.50ab
IRGC 103759
19.0
3.78cd
3.19bc
3.50de
1.44abc
5229c
14.3a
0.72d
RAM 123
17.0
4.48ef
3.38cd
2.56ab
1.62e
5428cde
12.2a
0.79d
TOG 7682
16.0
3.01ab
2.60a
2.88bc
1.42ab
6122f
0.8a
0.55bc
TOG 12440
21.0
5.00f
3.43cd
2.38a
1.51bcd
4142b
12.3a
0.72d
TOG 14365
17.0
3.56bc
3.06b
3.25cde
1.55de
5303cd
0.7a
0.50ab
19.0
4.33de
3.50d
4.36f
1.52cde
3825b
22.8ab
0.60c
Koshihikari
23.0
6.04g
3.40cd
3.67e
1.60de
2729a
149.1c
0.43a
WITA 4
17.0
5.76g
3.39cd
3.04bcd
1.41a
3761b
40.3b
0.56bc
O. sativa x O. glaberrima cross
NERICA 4
Asian rice
OCT – Optimum Cooking Time
SL – Solids Loss
WUR – Water Uptake Ratio
VER – Volume Expansion Ratio
LER – Length Expansion Ratio
#
Values with the same letters are not significantly different within columns
94
Table 4.3. Pasting Properties of Flours from African and Asian Rice Kernels#
PKT
PV
TV
FV
BDV
SB
Sample
(min) (RVU) (RVU) (RVU) (RVU) (RVU)
African rice
IRGC 80787
5.9
171.6h
156.5f
335.1g
15.1c
163.5i
IRGC 86741
7.0
60.9b
65.2b
136.2b
0*a
75.4c
IRGC 103759
6.8
99.6d
101.1d
206.2d
0*a
106.7g
RAM 123
6.0
104.9e
100.3d
203.5d
4.6b
98.6f
TOG 7682
7.0
48.5a
51.1a
94.5a
0*a
46.0b
TOG 12440
5.8
144.4f
129.6e
276.4f
14.8c
132.0h
TOG 14365
7.0
78.5c
80.4c
165.3c
0*a
86.8d
O. sativa x O. glaberrima cross
5.5
166.2g
126.5e
260.5e
39.7d
94.4ef
Koshihikari
6.0
271.6i
153.2f
260.6e
118.5f
-11.0a
WITA 4
5.8
173.0h
128.2e
262.7e
44.8e
89.7de
NERICA 4
Asian rice
PKT= Peak Time; PT= Pasting Temperature; PV= Peak Viscosity; TV= Trough
Viscosity;
FV= Final Viscosity; BDV= Breakdown Viscosity; SB= Setback Viscosity
#
Values with the same letters are not significantly different within columns
*
0 indicates the stability of the paste, i.e. there was no breakdown
95
Table 4.4. DSC Gelatinization Temperatures and Enthalpy of Flours from African
and Asian Rice Kernels#
Accession
African rice
To (°C)
Tp (°C)
Tc (°C)
ΔH (J/g)
MR (°C)
IRGC 80787
71.6e
75.5de
83.2f
9.1d
11.6
IRGC 86741
71.1d
75.1cd
82.5d
7.4b
11.4
IRGC 103759
71.6e
76.5f
83.8h
8.3c
12.2
RAM 123
70.7c
75.0c
82.9e
9.6ef
12.2
TOG 7682
70.8c
75.7e
84.8j
9.8fg
14.0
TOG 12440
71.1d
75.5de
84.1i
10.1g
13.0
TOG 14365
71.4e
75.6de
83.4g
9.2de
12.0
72.0f
76.3f
82.0c
7.5b
10.0
Koshihikari
62.9b
69.1b
76.1b
7.7b
13.2
WITA 4
62.2a
68.0b
75.6a
6.1a
13.4
O. sativa x O. glaberrima cross
NERICA 4
Asian rice
To - Onset of Gelatinisation Temperature
Tp - Peak of Gelatinisation Temperature
Tc - Conclusion of Gelatinisation Temperature
ΔH - Enthalpy of Gelatinization
MR= Melting Range (Tc-To)
#
Values with the same letters are not significantly different within columns
96
Table 4.5. Transition Parameters of Stored African and Asian Rice Flour Gels#
Tp (°C)
Tc (°C)
ΔH (J/g)
MR (°C)
% Retrgd1
43.5abc
58.4c
73.0e
4.8f
29.5
52.7
IRGC 86741
42.6a
58.3c
71.7bc
3.8cd
29.1
51.4
IRGC 103759
43.3abc
58.1c
72.0cd
4.2de
28.7
50.6
RAM 123
44.4bc
58.0bc
71.7bc
4.1d
27.3
42.7
TOG 7682
43.8abc
58.1c
72.7de
4.8f
28.9
49.0
TOG 12440
43.0ab
58.1c
72.1cd
4.7ef
29.1
46.5
TOG 14365
44.6c
57.4bc
68.4a
3.3c
23.8
35.9
Sample
African rice
To (°C)
IRGC 80787
O. sativa x O. glaberrima cross
43.7abc
58.0bc
72.2cde
3.9d
28.5
52.0
Koshihikari
43.4abc
55.7a
71.8bc
0.3a
28.4
3.9
WITA 4
44.1abc
57.0b
71.0b
0.90b
26.9
14.8
NERICA 4
Asian rice
To = Onset Temperature; Tp = Peak Temperature; Tc = Conclusion Temperature;
ΔH = Enthalpy of Gelatinization. MR= Melting Range (Tc-To);
1
Percentage Retrogradation (calculated as ΔH of retrograded gel/ΔH of starch granules * 100)
#
Values with the same letters are not significantly different within columns
97
Table 4.6. Nutritional Starch Fractions during Hydrolysis of Cooked African and Asian
Rice#
Sample
African rice
RDS (%)
SDS (%)
RS (%)
IRGC 80787
41.3a
36.8e
21.9e
IRGC 86741
44.5b
31.8d
23.8g
IRGC 103759
49.9cd
30.2c
20.0c
RAM 123
51.4e
27.7a
21.0d
TOG 7682
42.5a
39.7f
17.9b
TOG 12440
49.6c
30.0bc
20.6cd
TOG 14365
50.9de
28.5ab
20.6cd
49.3c
29.0abc
21.8e
Koshihikari
45.1b
41.3g
13.8a
WITA 4
49.5c
27.6a
23.0f
O. sativa x O. glaberrima cross
NERICA 4
Asian rice
#
Values with the same letters are not significantly different within columns
RDS =Rapidly Digestible Starch; SDS = Slowly Digestible Starch;
RS = Resistant/Residual Starch
98
Table 4.7. Kinetic Constant, Hydrolysis Index and Expected Glycemic Index
of cooked African and Asian rice#
Kinetic
Hydrolysis
Sample
eGI
Constant (k)
Index (HI)
African rice
IRGC 80787
0.0376a
73.2a
71.3a
IRGC 86741
0.0438b
74.1ab
72.4b
IRGC 103759
0.0488c
79.7c
76.9ef
RAM 123
0.0525e
79.8c
77.1ef
TOG 7682
0.0364a
76.3abc
74.0c
TOG 12440
0.0490c
79.1c
76.4de
TOG 14365
0.0513de
79.8c
77.0ef
0.050cd
78.1c
75.8d
Koshihikari
0.0369a
83.7d
77.5f
WITA 4
0.0514de
77.5bc
76.5de
O. sativa x O. glaberrima cross
NERICA 4
Asian rice
#
Values with the same letters are not significantly different within
99
Table 4.8. Correlations between Amylose; Protein and Cooking and Textural Properties#
Amylose
Protein Bulk
OCT
Density
Solids
Loss
WUR
VER LER Hardness
Adhesiveness
Amylose
Protein
0.5872
Bulk Density -0.5914
-0.6814
OCT
-0.6399
-0.5206 0.5192
Solids Loss
-0.6661
-0.8886 0.5407
0.5212
WUR
NS
-0.8638 0.5865
NS
0.7750
VER
-0.4826
NS
NS
NS
NS
NS
LER
NS
NS
0.5905
NS
NS
0.5323
Hardness
0.7577
0.8387
-0.5653 -0.6555 -0.8844 -0.7546 NS
Adhesiveness -0.8161
Springiness
#
0.5255
NS
NS
-0.7020 0.7201
0.6841
0.6439
NS
NS
NS
-0.6907
NS
NS
NS
NS
NS
NS
NS
NS
Correlation is significant at the 0.05 level
100
NS
Springiness
Table 4.9. Correlations between Amylose and Thermal and Pasting Properties#
To
Tp
Tc
ΔH
Peak
Trough
Final Breakdown Setback Total
viscosity viscosity viscosity viscosity viscosity setback
Amylose 0.7572 0.7370 0.8277 0.6777
OCT
#
NS
NS
NS
NS
-0.7568
-0.4660
NS
-0.9016
0.6418
NS
0.6803
0.5575
NS
NS
NS
NS
Correlation is significant at the 0.05 level
101
350
300
IRGC 80787
IRGC 86741
250
Viscosity (RVU)
IRGC 103759
RAM 123
200
TOG 7682
150
TOG 12240
TOG 14365
100
NERICA 4
50
KOSHIHIKARI
WITA 4
0
0
5
10
Time (min)
Figure 4.1. RVA Pasting Profiles of African and Asian Rice Flour samples.
102
15
Heat Flow (W/g)
-3
–––––––
–––––––
–––––––
–––––––
–––––––
–––––––
–––––––
–––––––
–––––––
––––
IRGC 80787.001
IRGC 86741.001
IRGC 103759.001
RAM 123.001
TOG 7682.001
TOG 12440.001
TOG 14365.001
NERICA 4.001
KOSHIHIKARI .001
WITA 4.001
-4
-5
Exo Up
20
40
60
80
Temperature (°C)
Figure 4.2. DSC Thermograms of African and Asian Rice Flour.
103
100
120
Universal V4.7A TA Instruments
80
Starch hydrolyzed (%)
IRGC 80787
60
IRGC 86741
IRGC 103759
RAM 123
40
TOG 7682
TOG 12440
TOG 14365
20
NERICA 4
KOSH'RI
WITA 4
0
0
40
80
120
Time (min)
Figure 4.3. Hydrolysis Kinetics of Cooked and Freeze-Dried African and Asian Rice Kernels.
104
Plot of Fitted Model
24
6.9
22
5.9
Solids Loss
OCT
Plot of Fitted Model
20
4.9
3.9
18
2.9
16
16
20
24
Amylose
28
16
32
OCT = 27.0311 - 0.319824*Amylose
r=-0.6399
20
28
32
Solids Loss = 8.87102 - 0.176512*Amylose
r=-0.6661
Plot of Fitted Model
Plot of Fitted Model
6600
6600
5600
5600
Hardness
Hardness
24
Amylose
4600
4600
3600
3600
2600
2600
16
18
20
OCT
22
16
24
Hardness = 11657.0 - 365.893*OCT
r=-0.6555
20
24
Amylose
28
32
Hardness = -721.243 + 211.394*Amylose
r= 0.7577
Plot of Fitted Model
Plot of Fitted Model
180
6.9
150
Adhesiveness
Solids Loss
5.9
4.9
3.9
120
90
60
30
0
2.9
16
18
20
OCT
22
16
24
Solids Loss = -0.922698 + 0.276376*OCT
r=0.5212
20
24
Amylose
28
32
Adhesiveness = 258.6 - 8.70015*Amylose
r=-0.8161
Figure 4.4. Multiple Linear Regression Models to Describe the Relationship between Amylose,
OCT and Cooking and Textural Parameters.
105
4.15
References
AACC International. Approved Methods of Analysis, 11th Ed. Method 61-03.01.
Amylose content of milled rice. First approval October 15, 1997; Re-approval
November 3, 1999. AACC International: St. Paul, MN.
http://dx.doi.org/10.1094/AACCIntMethod-61-03.01.
Adebowale, K. O., Olu-Owolabi, I. O., Olawunmi, E. K., and Lawal, O. S. 2005.
Functional properties of native, physically and chemically modified breadfruit
(Artocorpus altilis) starch. Industrial Crops and Products 21: 343-351.
Aluko, G., Martinez, C., Tohme, J., Castano, C., Bergman, C., and Oard, J.H. 2004. QTL
mapping of grain quality traits from the interspecific cross Oryza sativa×O.
glaberrima. Theor. Appl. Genet.109: 630-639.
Atkinson, F. S., Brand-Miller, J. C., and Foster-Powell, K. 2008. International tables of
glycemic index and glycemic load values. Diabetes Care 31: 2281-2283.
Bao, J. S. 2012. Toward understanding the genetic and molecular bases of the eating and
cooking qualities of rice. Cereal Foods World 57: 148-156.
Chung, H. J., Liu, Q. A., Wang, R. L., Yin, Y. L and Li, A. K. 2010. Physicochemical
properties and in vitro starch digestibility of cooked rice from commercially
available cultivars in Canada. Cereal Chem. 87: 297-304
Englyst, H. N., Kingman, S., and Cummings, J. 1992. Classification and measurement of
nutritionally important starch fractions. Eur. J. Clin. Nutr. 46: 33-50.
106
Evans, L. T. 1972. Storage capacity as a limitation on grain yield. In: Banos Los (Ed) Rice
breeding. International Rice Research Institute, Manila: 499-511.
Falade, K. O., Semon, M., Fadairo, O.S., Oladunjoye, A. O., Orou, K. K. 2014. Functional
and physico-chemical properties of flours and starches of African rice cultivars.
Food Hydrocoll. 39: 41-50.
Fan, C., Xing, Y., Mao, H., Lu, T., Han, B., Xu, C., Li, X., and Zhang, Q. 2006. GS3, a
major QTL for grain length and weight and minor QTL for grain width and
thickness in rice, encodes a putative transmembrane protein. Theor.Appl.Genet.
112: 1164-1171.
Fisher, D. K., and Thompson, D. B. 1997. Retrogradation of maize starch after thermal
treatment within and above the gelatinization temperature range. Cereal Chem. 74:
344-351.
Foster-Powell, K., Holt, S. H. A., and Brand-Miller, J.C. 2002. International table of
glycemic index and glycemic load values. Am. J. Clin. Nutr.76:5-56.
Frei, M., Siddhuraju, P., and Becker, K. 2003. Studies on the in vitro starch digestibility
and the glycemic index of six different indigenous rice cultivars from the
Philippines. Food Chem. 83: 395-402.
Futakuchi, K., and Sié, M. 2009. Better exploitation of African Rice (Oryza glaberrima
Steud.) in varietal development for resource-poor farmers in West and Central
Africa. Agric. J. 4: 96-102.
107
Juliano, B. O., Perez, C. M., Alyo-Shin, E. P., Romanov, V. B., Blakeney, A. B., Welsh, L.
A., Choudhury, N. H., Lilia Delgado, L., Iwasaki, T., Shibuya, N., Mossman, A. P.,
Siwi, B., Damardjati, D. S., Suzuki, H., and Kimura, H. 1984. International
cooperative test on texture of cooked rice. J. Texture Stud. 15: 357-376.
Juliano, B. O. 1985. Criteria and tests for rice grain qualities in: Rice: Chemistry and
Technology, 3rd Ed. St Paul, MN, USA. Am. Assoc. Cereal Chem.: 443-524.
Juliano, B.O. 1998. Varietal impact on rice quality. Cereal Foods World 43: 207-222.
Kesavan, M., Song, J. T., and Seo, H. S. 2013. Seed size: a priority trait in cereal crops.
Physiol.Plantarum. 147: 113-120.
Kim, Y. S., and Wiesenborn, D. P. 1995. Starch noodle quality as related to genotypes.
J. Food Sci. 61: 248-252.
Lawal, O. S., Lapasin, R., Bellich, B., Olayiwola, T. O., Cesàro, A., Yoshimura, M. 2011.
Rheology and functional properties of starches isolated from five improved rice
varieties from West Africa. Food Hydrocoll. 25: 1785-1792.
Leelayuthsoontorn, P., and Thipayarat, A. 2006. Textural and morphological changes of
Jasmine rice under various elevated cooking conditions. Food Chem. 96: 606-613.
Li, J., Thomson, M., and McCouch, S. R. 2004. Fine Mapping of a grain-weight
quantitative trait locus in the peri-centromeric region of rice chromosome 3.
Genetics 168: 2187-2195
108
Li, J., Xiao, J., Grandillo, S., Jiang, L., Wan, Y., Deng, Q., Yuan, L., and McCouch, S.R.
2004. QTL detection for rice grain quality traits using an interspecific backcross
population derived from cultivated Asian (O. sativa L.) and African
(O. glaberrima S.) rice. Genome 47: 697-704.
Meullenet, J-F. C., Carpenter, J. A., Lyon, B. G., and Lyon, C. E. 1997. Bi-cyclical
instrument for assessing texture profile parameters and its relationship to sensory
evaluation of texture. J. Text. Stud. 28: 101-118.
Miller, J.B., Pang, E., and Bramall, L. 1992. Rice: a high or low glycemic index food?
Am. J. Clin. Nutr. 56: 1034-1036.
Mohapatra, D., and Bal, S. 2006. Cooking quality and instrumental textural attributes of
cooked rice for different milling fractions. J. Food Eng. 73: 253-259.
Morrison, W. R., and Azudin, M. N. 1987. Variation in the amylose and lipid contents and
some physical properties of rice starches. J. Cereal Sci. 5: 35-37.
Nanri, A., Mizoue, T., Noda, M., Takahashi, Y., Kato, M., and Inoue, M. 2010. Rice
intake and type 2 diabetes in Japanese men and women: the Japan Public Health
Center-based prospective study. Am. J. Clin. Nutr. 92: 1468-1477.
Ndjiondjop, M. N., Manneh, B., Cissoko, M., Drame, N. K., Kakai, R. G., Bocco, R.,
Baimey, H., and Wopereis, M. 2010.
Drought resistance in an interspecific
backcross population of rice (Oryza spp.) derived from the cross WAB56-104 (O.
sativa) × CG14 (O.
glaberrima). Plant Sci. 179: 364-373.
109
Noda, T., Nishiba, Y., Sato, T., and Suda, I. 2003. Properties of starches from several lowamylose rice cultivars. Cereal Chem. 80: 193-197.
Ranawana, D. V., Henry, C. J. K., Lightowler, H. J., and Wang, D. 2009. Glycaemic index
of some commercially available rice and rice products in Great Britain. Int. J. Food
Sci. Nutr. 60:99-110.
Sarla, N., and Swamy, B.P.M. 2005. Oryza glaberrima: a source for the improvement of
Oryza sativa. Curr. Sci. 89: 955-963.
Shimelis, E. A., Meaza, M., and Rakshit, S. 2006. Physico-chemical properties, pasting
behavior and functional characteristics of flours and starches from improved bean
(Phaseolus vulgaris L.) varieties grown in East Africa. Agric Eng Int:
CIGR Journal. 8: 1-18.
Singh, N., Kaur, L., Sodhi, N. S., and Sekhon, K. S. 2005. Physicochemical, cooking and
textural properties of milled rice from different Indian rice cultivars. Food Chem.
89: 253-259.
Sweeney, M., and McCouch, S. 2007. The complex history of the domestication of rice.
Ann. Bot. 100: 951-957.
Thomas, R., Wan-Nadiah, W. A., and Bhat, R. 2013. Physiochemical properties,
proximate composition, and cooking qualities of locally grown and imported rice
varieties marketed in Penang, Malaysia. Int. Food Res. J. 20: 1345-1351.
Tester, R. F., and Morrison, W. R. 1990. Swelling and gelatinization of cereal starches I.
Effects of amylopectin, amylose and lipids. Cereal Chem. 67: 551-557.
110
Vamadevan, V., Bertoft, E., Seetharaman, K. 2013. On the importance of organization of
glucan chains on thermal properties of starch. Carbohydr. Polym. 92: 1653-1659.
Vamadevan, V., and Bertoft, E. 2014. Structure-function relationships of starch
components. Starch/Stärke. 66: 1-14
Watanabe, H., Futakuchi, K., Teslim, I., and Sobambo, B. A. 2002. Milling characteristics
and grain quality traits of interspecific progenies of Asian rice (Oryza sativa) with
African rice (Oryza glaberrima). Jpn. J.Tropical Agric., 46:47-55.
Watanabe, H., Futakuchi, K., Jones, M.P., and Sobambo, B.A. 2004. Grain protein content
of African rice (Oryza glaberrima Steud) lines and Asian rice (O. sativa L.)
varieties in West Africa. Oryza 41: 35-38.
Yu, S., Ma, Y., and Sun, D. 2009. Impact of amylose content on starch retrogradation and
texture of cooked milled rice during storage. J. Cereal Sci. 50: 139-144.
Zaidul, I. S. M., Nik Norulaini, N. A., Mohd Omar, A. K., Yamauchi, H., and Noda, T.
2007. RVA analysis of mixtures of wheat flour and potato, sweet potato, yam, and
cassava starches. Carbohyd. Polym. 69: 784-791
Zhang, G., Ao, Z., and Hamaker, B. R. 2006. Slow digestion property of native cereal
starches. Biomacromolecules. 7: 3252-3258.
Zhang, C.-H., Li, J.-Z., Zhu, Z., Zhang, Y.-D., Zhao, L., and Wang, C.-L. 2010. Cluster
analysis on Japonica rice (Oryza sativa L.) with good eating quality based on SSR
markers and phenotypic traits. Rice Sci. 17: 111-121.
111
Zhu, L.-J., Liu, Q.-Q., Wilson, J. D., Gu, M.-H and Shi, Y.-C. 2011. Digestibility and
physicochemical properties of rice (Oryza sativa L.) flours and starches differing in
amylose content. Carbohydr. Polym. 86: 1751-1759.
112
CHAPTER 5: THERMAL AND MOLECULAR CHARACTERIZATION OF
STARCHES ISOLATED FROM AFRICAN RICE (Oryza glaberrima)
113
Abstract
African rice (Oryza glaberrima) and Asian rice (Oryza sativa) are two distinct types of
domesticated rice with the former having been less extensively studied.
With the
expansion of rice food products in Sub-Saharan Africa, characterization of starches from
African rice is needed for product development. This study was designed to investigate the
physical and molecular properties of starches isolated from seven African rice accessions
(ARAs) in comparison with two O. sativa samples; cv Koshihikari (spp. japonica) and cv
WITA 4 (spp. indica) and an inter specific cross between sativa x glaberrima cv NERICA
4. The granule size distribution of the starches examined ranged from 2-22 μm and showed
tightly packed polyhedral shaped granules. A wide range of amylose content (17-30%)
was found among rice starches with the ARAs having relatively higher proportions. The
energy required to gelatinize starches or melt retrograded starch gels varied among starches
and was observed to be lower in NERICA 4 and the Asian rice varieties. Significant
(p < 0.05) differences in the peak, trough, final, breakdown and total setback viscosities
were observed among the starches.
The starches also showed significant (p < 0.05)
differences in the ratio of the absorbance to scattering coefficient (K/S) values when
exposed to iodine vapor, suggesting differences in the rigidity of the glucan chains in the
granules. Starches from all the ARAs had higher K/S values than the other samples.
Additionally, all the starches displayed the typical 'A' type x-ray diffraction pattern. The
significant differences in physico-chemical properties observed between the ARAs and
Asian rice varieties are likely due to differences in amylose content which may influence
the cooking and eating quality of the ARAs. The results also suggest that the structure and
granular architecture of the ARA starches may differ from the Asian rice starches.
Keywords: Rice starch, iodine, molecular characteristics, physical characteristics, x-ray,
amylose
114
5.1 Introduction
Rice is one of the main cereal crops cultivated worldwide and millions depend on it as food
staple. Oryza sativa (Asian rice) and Oryza glaberrima (African rice) are the two distinct
types of domesticated rice with O. sativa having a global distribution particularly in Asia.
However, African rice, which has the ability to withstand a wide range of environmental
stress conditions, is cultivated mainly in the West Africa sub region (Sweeney and
McCouch 2007). Factors such as consumer preferences, genetic character and variation in
climatic conditions have led to the development of many rice cultivars and the diversity in
these rice cultivars is associated with different cooking qualities and physical properties
(Wani et al. 2012). Of major interest is starch, the predominant component of rice (90%
dry basis) and its physicochemical properties, including apparent amylose content,
amylopectin structure, gelatinization temperatures and pasting viscosities that determine
the eating, cooking and milling qualities of rice. Several benefits of rice starch over other
starch sources (e.g. hypo-allergenicity, creaminess and spreadability, smoothness, bland
flavor, small granules, acid resistance, and relatively good freeze-thaw stability of its gels)
make it a vital ingredient in the formulation of a variety of food products and other
applications (Lawal et al. 2011; Wani et al. 2012). These exceptional attributes make rice
starch one of the most suitable for industrial purposes (Vandeputte and Delcour 2004).
Since the development of new rice cultivars affects physico-chemical properties of starch
as reported by several researchers (Ahmed et al. 2008; Lawal et al. 2011; Wang et al.
2012), it is expedient that the characteristics and functionalities of starches from new
cultivars are studied so that those with desirable functional and physico-chemical
properties are selected for specific end uses.
115
The NERICA (NEw RICe for Africa) rice varieties (an interspecific cross between
O. sativa x O. glaberrima) developed in the 1990s and disseminated to farmers has
contributed to poverty reduction in sub Saharan Africa. The development of this variety
harnessed desirable qualities such as high tillering ability, rapid seed growth and early
maturity. Despite the success of this project, it is believed that African rice has not been
fully exploited in terms of other traits that were not focused on during the varietal
improvement. These include amylose content (and for that matter starch), which is a major
factor that influences texture and taste of cooked rice (Futakuchi and Sié 2009). To this
end, a collection of African rice accessions is being systematically characterized by Africa
Rice Center (AfricaRice) for resistance against major diseases and environmental stresses.
Objective: Since the basic structure of African rice starches have not been characterized
yet, this study focuses on investigating the basic physical, thermal and molecular
characteristics of seven African rice accessions that were selected from a population of
1,020 through cluster analysis. The cluster analysis of the 1,020 accessions was based on
apparent amylose and pasting data of their flour.
116
5.2 Materials and Methods
5.2.1
Samples
Rice samples were supplied by the Africa Rice Center based in Cotonou, Benin. They
consisted of seven African rice accessions (ARAs) - IRGC 86741, IRGC 80787, IRGC
103759, RAM 123, TOG 7682, TOG 12440 and TOG 14365. In addition were NERICA 4
(an interspecific cross between O. sativa x O. glaberrima) and two O. sativa varieties –
cv. Koshihikari (spp. japonica) and cv.WITA 4 (spp. indica) for comparison purposes
5.2.2
Methods
5.2.3
Starch Isolation
Starch was isolated from the rice samples using the method of Waduge et al. (2010) with
some modifications. The rice kernels were frozen in liquid nitrogen for 2 min and milled
for 1 min using a Coffee and Spice Grinder with stainless steel grinding blades,
(SmartGrindTM model CBG100SC - Black and Decker Corporation, Towson, Maryland,
USA) into flour. Fifty grams of rice flour was then transferred into 1 L sodium borate
buffer (12.5 mM, pH 10 and containing 0.5% sodium dodecyl sulfate (SDS) {w/v} and
0.5% Na2S2O5 {w/v}). The flour-buffer suspension (1:20 {w/v}) was stirred at a moderate
speed for 5 min to extract proteins. The suspension was subjected to centrifugation at
900 × g for 5 min and the residue recovered. The protein extraction step was repeated, and
the residue was washed three times with double distilled water and recovered by
centrifugation (900 × g for 5 min) each time. Thereafter, the residue was suspended in
double distilled water and the starch slurry was first passed through four layers of
117
cheesecloth to trap coarse particles and then through nylon material with a mesh size of 70
μm. The coarse mass in the cheesecloth after the filtration was further ground into finer
particles to recover more starch. The starch-water suspension was centrifuged at 1600 × g
for 10 min, after which the brown protein layer that settled on top of the starch layer was
carefully removed with a spatula. The starch was then suspended in double distilled water
again and subjected to centrifugation. This step in the isolation process was repeated till all
the particles that formed the brown layer were removed from the surface of the starch
residue. The starch fraction was finally washed with ethanol (98%) followed by acetone
and allowed to air dry in a fume hood. Starch yield was expressed as the dry weight per
weight of rice flour.
5.2.4
Physical Characteristics:
5.2.5
Starch Damage
AACC Method 76-31.01 was employed in measuring level of starch damage in the rice
starch samples using the Megazyme Starch Damage Test Kit (Megazyme International Ltd,
Bray, Ireland).
5.2.6
Nitrogen
The Dumas combustion method for the quantitative determination of total nitrogen was
employed to measure the nitrogen content of rice starch samples. Weighed starch samples
(0.2 g) were combusted in a LECO Nitrogen/Protein Analyzer (LECO, FP-528 Lakeview
Avenue, St. Joseph, Michigan, United States) after ethylenediaminetetraacetic acid
118
(EDTA) standard was run to calibrate the equipment. Nitrogen was converted into protein
by multiplying the nitrogen determined by a factor of 5.7.
5.2.7
Particle Size Determination
Particle size distribution of starch samples was determined with Mastersizer 2000 (Malvern
Instruments Ltd, Worcestershire, UK). Before analysis was carried out, the motor speed
was set to 2000 rpm and the dispersion unit was flushed with distilled water. The tank was
completely flushed three times with tap water and once with distilled water. The tank was
refilled with distilled water and it was ensured that no bubbles were left in the tubes to and
from the sample chamber. Starch samples were added with care to avoid lump formation
and in enough quantities to produce laser obscuration range between 12% and 14%. A
refractive index of 1.52 was used for the starches in the dispersed phase, 1.33 for the
aqueous continuous phase and sample absorption of 0.001.
A 300 Reverse Fourier
(convergent beam) lens was employed for the analyses. The Mastersizer 2000 software
controls the system during the measurement process and analyzes the scattering data to
calculate a particle size distribution.
5.2.8
Morphology
5.2.9
Scanning Electron Microscopy
The rice starch granule characteristics was studied with Hitachi S-570 Scanning Electron
Microscope (Hitachi Scientific Instruments, Rexdale, Ontario,Canada). The starch samples
were initially thinly spread on double-sided adhesive tapes attached to aluminum specimen
stubs. The samples were then coated with Gold/Palladium dust to a thickness of 15 nm in
119
approximately 2 min. using Emitech K550X Sputter Coater (Quorum Technologies Ltd,
Ashford, Kent, United Kingdom). The working distance used on the scanning electron
microscope was 15 mm with an accelerating voltage of 10 kV. Samples were placed onto
the stage mount using stub tweezers and tightened with Allen key. Images were brought
into focused by increasing the magnification slowly and adjusting the scan speed until the
images were sharp and stretches removed.
Digital images of the samples were then
acquired.
5.2.10 Wide Angle X-ray Diffraction Scattering
Sample preparation was done by hydrating 0.1 g quantities of the isolated rice starches over
a saturated solution of K2SO4 with aw 0.97 in an air-tight desiccator at room temperature
for 30 days. The hydrated starches were then tightly packed by firmly pressing samples
into the groove of a quartz plate sample holder. X-ray diffraction measurements were
obtained with a Rigaku X-ray diffractometer (Rigaku-Denki Co., Tokyo, Japan). The
operating conditions included a target voltage of 40 kV, a current of 44 mA, step time of
0.95 s and scanning range was from 3-35° at a speed of 1.00°/min. The divergence slit
width, sampling width and scatter slit width were 0.5°, 0.5° and 0.03°, respectively. Jade
software (Material Data Inc. CA, USA) was used to estimate the relative crystallinity by
engaging a Gaussian function for curve fitting. The ratio of the area under the peaks to the
total area under the curve (peaks + amorphous background) was reported as percent
crystallinity of the starches.
120
5.2.11 Reflectance Spectra after Exposure of Starch to Iodine Vapor
The reflectance spectra and color {(L* (white/black), a* (green/red), and b* (blue/yellow)}
of the samples was determined using the Kubetta and Munk equation:
K
S
=
(1−R)2
2𝑅𝑅
Equation 1
where K is the absorption coefficient, S is the scattering coefficient and R is the reflectance
expressed as a fraction between 0 and 1 (Billmeyer and Saltzman 1981). The K/S values of
the samples (representing the absorption corrected to the light scattering) were measured
over a wavelength range of 400 to 700 nm using a CM 3500-d spectrophotometer (Konica
Minolta Sensing Inc., Mahwah, NJ, USA) equipped with SpectraMagic NX CM-S 100
software at 10 nm intervals. Exactly 0.2 g of sample equilibrated to aw 0.97 and exposed to
iodine for 24 h was spread over the sample cell as a thick layer and the measurements were
taken. Sample equilibrated to the same aw but not exposed to iodine vapor was used as a
reference.
5.2.12 Thermal Properties
5.2.13 Rapid Visco Analyzer
The pasting properties of rice starch samples were measured using Rapid Visco Analyzer
RVA-4, (Newport Scientific Pty Ltd., Warriewood, NSW, Australia). 3 g of starch (12%
moisture basis) from each rice sample was weighed directly into the aluminium canister
and mixed with 25 mL of distilled water resulting in a 9.5% suspension. The test profile
involved equilibrating the sample at 50°C for 1 min, heating to 95°C at a rate of 12 °C/min,
holding at 95 °C for 2.5 min, cooling to 50 °C at a rate of 12 °C/min and holding at 50 °C
121
for 2.5 min. The rotating speed of the paddle was maintained at 160 rpm throughout the
run except for the first 10 s where paddle speed was held at 960 rpm. The under listed
pasting parameters were determined from the RVA pasting curve; peak viscosity
(maximum viscosity attained during heating), hot paste viscosity (trough) and final
viscosity or cold paste viscosity (viscosity at end of cooling period). Breakdown was
calculated as the difference between peak and trough viscosities while setback was
calculated as the difference between the final viscosity and the peak viscosity.
5.2.14 Differential Scanning Calorimetry
TA Instruments, Q1000 differential scanning calorimeter (DSC) equipped with a thermal
analysis data and recording software facility (TA Instruments, Universal Analysis 2000,
DE, USA) was used to measure the gelatinization properties of the rice flours. Starch
dispersion in water (1:3) was allowed to equilibrate for at least 3 h at room temperature
before sample was loaded. In all measurements, the thermogram was recorded with an
empty aluminum pan as a reference and the scanning temperature range was programmed
from 20-120 °C at a heating rate of 10 °C/minute. Scanned samples were stored at 4 °C for
14 days and then re-scanned using the same temperature profile. The onset (To), peak (Tp)
and conclusion (Tc) transition temperatures were reported while the enthalpy of
gelatinization (ΔH) was estimated by integrating the area between the thermogram and a
base line under the peak and was expressed as J/g of dry starch. Determinations were done
in duplicate.
122
5.2.15 Molecular Composition of Rice Starches
5.2.16 Chain Length and β-Limit Dextrins
Size distribution of starch components and their β-limit dextrins were investigated using
Gel Permeation Chromatography (GPC). Rice starch samples (8 mg) were dissolved in
90% DMSO (200 μL), heated in warm water (80 °C) for 5 min and then left overnight to
stir at room temperature (25 °C). Warm water (800 μL) was used to dilute the samples.
The size-distribution of the components of the starch samples was determined by
fractionating them (400 μL) on a Sepharose CL-2B column (1.6 × 32 cm) (GE-Healthcare,
Uppsala, Sweden). The eluent (0.01 M NaOH) was pumped through the column at a rate
of 0.5 mL/min. The total carbohydrate content of the fractions collected (1 mL) was
determined by using the phenol-sulfuric acid reagent (Dubois et al. 1956). The wavelength
maxima (λmax) of the glucan-iodine complex was determined with a WPA Spectrawave
S800 Diode Array Spectrophotometer (Biochrom Ltd., Cambridge, United Kingdom) by
initially neutralizing the collected fractions with 1 mL of 0.01 M HCl and thereafter adding
100 μL of 0.01 M I2-0.1 M KI solution. β-Limit dextrins (β-LD) of the rice starches were
prepared according to the method outlined by Bertoft (2004). Starch (8 mg) was dissolved
in 90% DMSO (200 μL) and left overnight to stir. β-amylase (4 U/mg, 2 μL) in 100 μL
(0.01 M NaOAc buffer; pH 6) was thereafter added and the resulting mixture was again
stirred overnight at room temperature (25 °C). Water was added to dilute the mixture to 1
mL and then 400 μL was injected onto the Sepharose CL-2B column.
123
5.2.17 Chain Length of Debranched Starches
GPC as described by Sargeant (1982) was used to analyze debranched rice starches.
Samples (4 mg) were dissolved in 90% DMSO (100 μL) by heating gently in a water bath
(80 °C) for 5 min. The dissolved starch was allowed to gently stir overnight at room
temperature (25°C). The solution was diluted with water (80 °C) (750 μL) and thereafter
sodium acetate buffer (0.01 M, pH 5.5, 100 μL) was added.
Exactly 1 μL each of
isoamylase (465 U/mL) and pullulanase M1 (925 U/mL) (Megazyme International,
Wicklow, Ireland) was added and the mixture incubated overnight with gentle stirring at
room temperature (25°C). The reaction was halted by placing samples in a boiling water
bath for 5 min. Fifty μL of 5 M NaOH was added after which the sample was diluted to
1.5 mL with deionized water and filtered. The filtered sample (1 mL) was applied on a
Sepharose CL-6B column (1 × 90 cm; GE Healthcare, Uppsala, Sweden) with NaOH
(0.5 M) as eluent pumped at a rate of 1 mL/min. The total carbohydrates content of the
fractions collected (1 mL) was analyzed by using the phenol-sulfuric acid reagent (Dubois
et al. 1956). The amylose and amylopectin portions of the samples were determined by
dividing the chromatogram at the lowest point between the two peaks as reported by
(Sargeant 1982). The amylose fraction was further divided into short and long chains.
5.2.18 Amylose
The amylose concentration of the samples was determined using GPC as described under
section 5.2.17 above.
124
5.2.19 Statistical Analysis
The statistical programme Statgraphics Centurion XV, Version 15.1.02 (StatPoint,
Warrenton, VA, USA) was employed to conduct statistical analysis. One-way analysis of
variance was performed to determine significant differences between any pair of means at
the 95.0% confidence level and multiple range tests were done to determine which means
were different.
5.3 Results
5.3.1
Starch Yield and Composition
The recovery of starch based on the whole weight of the dehulled rice samples is presented
in Table 5.1. The recoveries were in the range of 65.7-72.9%. Protein and fat contents of
all the starch samples were below 1% attesting to their purity. The starch damage level in
all the samples was 1.2% or less (Table 5.1) indicating that the milling step in the starch
extraction process did not adversely affect the structural integrity of the starch granules.
5.3.2
Morphology and Structure of Rice Starch Granules
Scanning electron micrographs of all the samples revealed that the morphologies of their
granules were similar.
The granules were found to be of similar sizes, angular and
polygonal/polyhedral in shape and possessed the same tight mode of packing (Figure 5.1).
Grooves that seem to be points of attachment between the granules were observed on the
surfaces of the granules. The granule size distribution curves of the samples displayed
unimodal shapes with peak size of about 5-6 μm (Figure 5.2). The granule sizes of the
starch samples ranged from 2-22 μm with Koshihikari, WITA 4 and RAM 123 having
125
similar size distributions with peak size of about 5 μm. The other samples possessed a
peak size of approximately 6 μm, with the exception of TOG 14365, which had a peak size
of 6.6 μm.
The X-ray diffraction patterns of the native and iodinated rice starches are illustrated in
Figures 5.3 and 5.4 respectively. The diffractograms showed regions of noticeable peaks
centered on 2θ 15°, a doublet on 17° and 18° and another single peak at 23°. This
diffraction pattern is consistent with cereal starches generally classified as type “A”,
(Imberty and Perez 1988). All the starch samples in this investigation displayed the typical
‘A’ type x-ray diffraction pattern. Exposure of the starch samples to iodine vapor resulted
in changes to some characteristics notably the relative crystallinity. There was a significant
increase in the relative crystallinity of all the iodinated starches by 2-5%, except for TOG
7682 (Figure 5.4). The diffraction pattern was not affected by the interaction between the
iodine and the starch samples, however, shifts in some of the peaks were observed. The
reduction in the intensities of peaks at 15°, 17° and 18° 2θ° were more obvious in NERICA
4 and all the ARAs than in WITA 4 however, no such effect was seen with Koshihikari. At
20° 2θ there was an increase in the peak intensity for all the starches due to the formation
of a single helical complex with iodine but the 23° 2θ peak remained intact for all the
starch samples.
The reflectance spectra of all the iodinated starches showed λmax at 550 nm (Figure 5.5).
Significantly higher K/S intensity values (p < 0.05) were observed for all the ARAs in
comparison to the other samples over the entire wavelength range that measurement was
taken (400-700 nm). The differences in the K/S values signify variances in the population
of glucan polymers available in the granules of the starches to form inclusion complexes
126
with iodine (Saibene et al. 2008). At λmax 550, the K/S intensity values were highest for
TOG 14365 (35.2) and lowest for Koshihikari (13.6) and, interestingly, all the ARAs were
higher in K/S value than NERICA4, WITA 4 and Koshihikari.
5.3.3
Thermal Properties of Starch Granules
The RVA profiles and pasting characteristics of the starch samples are presented in Figure
5.6 and Table 5.2 respectively. The samples had a peak time ranging from 5.0 min (TOG
14365) - 5.7 min (IRGC 86741). Pasting occurs when the starch granules absorb sufficient
water and swell after gelatinization. The pasting temperatures differed significantly (p <
0.05) and ranged from 71.1 °C (Koshihikari) to 78.6 °C (IRGC 80787). The peak viscosity
of the starches ranged from 231.5 to 308.9 RVU corresponding to RAM 123 and
Koshihikari, respectively. The trough viscosity, which is a measure of the ability of the
paste to withstand breakdown during the cooling phase, was from 134.8 RVU (RAM 123)
to 187.6 RVU (IRGC 80787) while final viscosity, which is the change in the viscosity
after holding cooked starch at 50°C, ranged from 266.3 to 332.3 RVU corresponding to
Koshihikari and IRGC 80787, respectively. Peak viscosity indicates the water-holding
capacity of the starch, while final viscosity, the most commonly used parameter to define a
particular sample’s quality, indicates the ability of the material to form a viscous paste or
gel after cooking and cooling. Breakdown viscosity was lowest for IGRC 86741 (68.7
RVU) and highest for Koshihikari (124.8 RVU). The total setback viscosity, which is a
measure of the tendency of starch paste to retrograde, was from 82.2 to 166 RVU
corresponding to Koshihikari and IRGC 103759, respectively.
127
Table 5.3 and the bar graph in Figure 5.7 show the transition temperatures (To, Tp, Tc) and
gelatinization enthalpies (ΔH) of the native, granular starches. The onset temperatures of
the ARAs and NERICA 4 were significantly higher than WITA 4 and Koshihikari. While
WITA 4 recorded the lowest To of 59.8 °C, TOG 14365 had the highest To of 69.1 °C. A
similar trend was observed for both Tp and Tc. Gelatinization enthalpy (ΔH) of starch
granules ranged from 13.3-15.8 J/g (Table 5.3) with the two Asian rice starches having the
lowest ΔH values. ΔH and the narrow melting range (Tc-To) of the ARAs suggests that
their amylopectin crystals are more homogeneous and uniformly stable in comparison to
the Asian rice (Ratnayake et al. 2001).
Figure 5.8 and Table 5.4 show the properties of the retrograded starch gels. To, Tp, Tc and
ΔH values were considerably lower than their respective native starches. To and Tp ranged
from 42.9 to 44.1 °C and 52.5 to 54.6 °C, respectively. While ΔH for the retrograded
starches ranged from 6.4-10.8 J/g, Tc was from 65.5 to 72.3 °C. While To, Tp, and ΔH of
the gelatinization of the native granules were considerably higher for the ARAs than the
Asian rice starches (Figure 5.7), differences for the retrograded starch samples occurred
mainly in Tc and ΔH (Figure. 5.8).
5.3.4
Molecular Structure of Rice Starches
The amylose concentration of the starch samples ranged from 17% for Koshihikari to
31.1% for TOG 14365 (Table 5.1). These values were obtained by analysing their chain
profiles with GPC on Sepharose CL-6B (Figure 5.9). It was observed that all the ARAs
and NERICA4 had significantly higher proportions (p<0.05) of amylose than the Asian rice
varieties. The gel permeation chromatograms also showed the relative proportions of long
128
(LCAM) and short chains (SCAM) of amylose (Figure 5.9). The relative amounts and ratios
of LCAM:SCAM of the starches are presented in Table 5.1. The ARAs generally had
significantly higher amounts of both long and short chains of amylose (p< 0.05) than
Koshihikari and WITA 4. All starches, except Koshihikari, recorded LCAM:SCAM ratios of
0.6 to 0.74 due to higher amounts of short amylose chains than long amylose chains.
Koshihikari possessed a high LCAM:SCAM ratio of 0.87, because it had nearly equal
amounts of LCAM and SCAM.
Gel permeation chromatograms of whole rice starches and their corresponding β-limit
dextrins (β-LDs) fractionated on Sepharose CL-2B along with their carbohydrate contents
and λmax values are presented in Figure 5.10. The elution profiles of the chromatograms
showed a bimodal distribution with the first/ narrow peak, which corresponds to the large
amylopectin molecules, eluted at the void volume of the column. The second/ broad peak
representing amylose (smaller glucan molecules) eluted after the void volume of the
column. The λmax values of the samples in the amylopectin portion ranged from about 560
to 585 nm. In this region the ARAs and NERICA 4 recorded higher λmax values than
Koshihikari and WITA 4. This reflects the presence of long external chains which are
available to complex with iodine in the ARAs and NERICA 4 (Table 6.1). A similar
observation was made in the amylose region and this suggests that the ARAs and NERICA
4 have longer amylose chains available for iodine complexing than Koshihikari and WITA
4. The λmax values of the β-LDs from all the samples were observed to be lower than the
corresponding whole starches in the amylose region. However, in the amylopectin region
higher λmax values were observed for β-LDs from Koshihikari. It is important to note that
iodine can form complexes with external and internal chains of amylopectin (Chauhan and
129
Seetharaman 2013; Vamadevan et al. 2015) but after β-amylolysis, the relative abundance
of long internal chains of amylopectin could shift the spectrum to higher λmax values.
Additionally, the availability of internal glucan chains for iodine binding could be
influenced by the nature of branching pattern. The maltose peak was the result of the
removal of external chains of the branched glucans and linear polymers of the starch by βamylase.
5.4 Discussion
The alkaline extraction procedure used in this study resulted in recoveries between 65.7
and 72.9%. A wide range of variation in rice starch recovery from different rice cultivars
(59% to 71.6%) have been reported (Mohan et al. 2005; Wang and Wang 2004). Yields of
70.0% to 73.77% were reported for five newly released improved rice varieties from West
Africa (Lawal et al. 2011) and the yields in this investigation compares well. Lawal et al.
(2011) reported that starch recovery is a function of the genetic composition of the rice
itself, the environment under which it was cultivated and the starch extraction method
employed. The method of starch isolation can significantly affect the pasting, rheological
and textural properties of starch paste (Zhong et al. 2009). Isolation of starch from rice is
quite different from other starches due to its unique protein composition (Wani et al. 2012).
The separation of starch from protein, fiber and lipid are the key elements in the isolation
process and a number of important steps have to be considered. These include avoidance
of amylolytic or mechanical damage to the starch granules, effective deproteinization of the
starch, minimizing loss of small granules and avoidance of starch gelatinization (Schulman
and Kammiovirta 1991).
130
Starch damage commonly occurs during milling of cereal grains and examples of the
damage are broken starch granules with exposed interior and granule fragments (Dhital et
al. 2010; Hasjim et al. 2009). Reports indicate that different milling or grinding processes
result in different degrees of damage to starch granules in flour depending on the
mechanical forces and temperature during the grinding process (Nishita and Bean 1982;
Nowakowski et al. 1986; Hatcher et al. 2002). Starch swelling can be improved through
water absorption when there is a low degree of damage to starch granules but excessive
damaged starch granules have increased solubility in cold water than the intact starch
granule and undergo less swelling during cooking. This is due to the degradation of starch
molecules (mainly amylopectin) and starch crystalline structure that occurs during
grinding, leading to the production of highly soluble, low molecular-weight molecules,
(Tester 1997; Morrison et al. 1994; Becker et al. 2001). Since milling is necessary to
disrupt protein and cell wall structures to facilitate the release of starch granules from the
grain matrices, it is essential that care is taken not to cause damage to starch molecules for
better starch characterization (Gidley et al. 2010). Damaged starch is known to affect
thermal characteristics of starch. High levels cause a decrease in gelatinization temperature
and enthalpy (Morrison et al. 1994). Starch damage levels of less than 1.2% were obtained
for all samples in this study.
The morphological characteristics of the starch granules analyzed in this study (Figure 5.1)
were largely consistent with what has earlier been reported in literature (Li and Yeh 2001;
Singh et al. 2003; Raina et al. 2007; Lawal et al. 2011; Wang et al. 2012) except for the
wider granule size distribution of 2-22 μm. Using a laser particle size analyzer, Wang et al.
2012 found no significant distinctions between various rice starches in terms of their size,
131
shape and distribution. The range of median size for indica and japonica rice starch
granules, which were angular and polygonal in shape, were reported to be around 6.2-7.7
μm and 6.7-7.8 μm, respectively.
Using the same procedure for determining size
distribution, Li and Yeh (2001) reported a size range of 2-10 μm with an average granule
size of 6.4 μm for Taiwan rice starch granules, which had polyhedral shape. Polyhedral
shaped granules with polygonal edges having size ranging between 3.1 and 7.8 μm was
reported by Raina et al. (2007) for Oryza sativa rice varieties using scanning electron
microscopy (SEM). Similarly, Lawal et al. (2011) used SEM to study starch isolated from
five improved rice varieties from West Africa and reported granules with irregular
polyhedral shapes that had sizes ranging from 1.5 to 6.1 μm. The granular structure of rice
starches vary in shape and size among different cultivars and rice types; non-waxy, waxy,
and long-grain (Wani et al. 2012).
All the starch samples investigated in this study had similar diffraction patterns, A-type
crystal. Lawal et al. (2011) reported X-ray diffraction patterns for NERICA that showed
no doublet but a single peak at 2θ 17.1° and a less intense and broader peak at 2θ 23.1°.
This was inconsistent with the findings on the NERICA 4 sample in this study. The
relative crystallinity (Figure 5.4), which was calculated as the ratio of the area under
crystalline peaks to the total area under the diffractogram (representing both crystalline and
amorphous structures), did not differ significantly among the samples (p<0.05).
The
relative crystallinities of the samples in this study were very similar to that of improved
rice varieties from West Africa as reported by Lawal et al. (2011), but marginally lower
than the crystallinity of indica rice starches with various amylose contents (He et al. 2006).
The observed decrease in the peak intensities of the X-ray diffractograms of the starch
132
samples at 2θ 15°, 17° and 18° after exposure to iodine vapor (Figure 5.4) have been
reported in other cereal starches. Saibene et al. (2008) reported a decrease in peak intensity
at 2θ 15°, 17°, 18° and 23° of corn starch and a loss of resolution in peaks around 2θ 2224° in potato starch. Similarly, Annor et al. (2014) and Cheetham and Tao (1998) found a
decrease in peak intensities in millet starches and normal maize starches, respectively, with
the disappearance of some peaks in high amylose maize starch. While the iodine did not
seem to have any observed effect on 2θ 23° in all the starch samples, the decrease in
intensities of the characteristic peaks 2θ 15°, 17°, 18° was more pronounced in NERICA 4
and the ARAs (Figure 5.4) than the two Asian rice starches (Koshihikari and Wita 4). The
decrease in peak intensities is thought to be due to the absorption of X-rays by high
molecular weight iodine molecules (Saibene et al. 2008).
The presence of glucan polymers in starch granules and their ability to form single helices
in the presence of iodine determine the extent of interaction with iodine (Waduge et al.
2010). The capacity of iodine to form inclusion complexes depends on the flexibility and
availability of the glucan polymers within the granule, while penetration of iodine into the
starch granule most likely depends on surface features of the starch granules such as pores
or channels (Kim and Huber 2008). The K/S intensity value (defined as the ratio of light
absorbance coefficient to light scattering coefficient) is indicative of differences in the
flexibilities of the glucan chains within the starch granules to form inclusion complexes
with iodine (Saibene et al. 2008). The higher K/S intensities of the starches from the ARAs
suggest that the population of flexible glucan polymers available to form inclusion
complexes with iodine was substantially greater than in the Asian rice and NERICA 4
starches (Figure 5.5). The formation of iodine-glucan complex when the granular starches
133
were exposed to iodine vapor affected their crystallinity to different extents. K/S at λmax
550 nm correlated highly (0.9389) with amylose content at p < 0.05 (Table 5.5) and the
equation of the fitted model is
K/S at λmax550 = -12.6625 + 1.53405*amylose.
Pasting properties of starch are important indicators of how the starch will behave in the
course of processing and they are used in determining the suitability of starch in different
foods and other applications (Wani et al. 2012). When starch is continuously heated in
excess water with stirring, granules swell and burst as a result of the breakdown of the
native starch structure. Amylose then leaches out of the disintegrated granules leading to
the formation of a homogeneous mass called paste (BeMiller 2007). All the starch samples
in this study exhibited similar pattern and trend but significant differences were found
among the parameters that were measured (Figure 5.6). The temperature at the onset of
viscosity increase (pasting temperature) was significantly lower in the two Asian rice
starches than in the ARAs and NERICA 4 (Table 5.2). Adebowale et al. (2005) reported
that breakdown viscosity measures the susceptibility of the cooked starch to fragmentation
and that the higher the breakdown in viscosity, the lower the ability of the starch sample to
withstand heating and shear stress during cooking.
The high breakdown viscosity
exhibited by Koshihikari suggests that it will have a low capacity to withstand heating and
shear stress compared to other starches. Setback viscosity is commonly used to describe
the increase in viscosity that occurs on cooling a starch paste (Fisher and Thompson 1997).
It is defined as the difference between the breakdown viscosity and the viscosity at 50 °C
and it determines the retrogradation tendency of the paste (Zaidul et al. 2007). The high
amylose content of the ARAs and NERICA 4 starches explains their high setback and
134
breakdown values. In this study amylose is correlated (p<0.05) with the pasting parameters
measured. The correlations between amylose and peak, final, setback and total setback
viscosities at (p<0.05) were -0.65, 0.51, 0.87 and 0.90 respectively (Figure 5.11, Table 5.5)
and the equation of the multiple linear regression model to describe the relationships are
Peak viscosity = 358.423 - 3.70361*amylose, Final viscosity = 226.873 +
2.74866*amylose and Total setback = 3.98736 + 5.06745*amylose respectively.
The transition temperatures (To, Tp and Tc) and enthalpies of gelatinization (ΔH) of the
starches in this study are in line with other published data. To values ranging from 53.3°C
to 75.9°C have been reported for non-waxy rice starch (Wang et al. 2010; Wickramasinghe
and Noda 2008), likewise Tp ranging from 61.8°C to 80°C (Wang et al. 2010;
Wickramasinghe and Noda 2008). Tc from 70.9°C to 85.4°C (Wang et al. 2010) and ΔH
from 3.7 J/g to 17.7 J/g (Singh et al. 2007; Vandeputtee et al. 2003) have been reported.
Yamin et al. (1999) reported that variations observed in To, Tp, Tc and ΔH in starches from
different rice cultivars might be attributed to differences in amounts of longer chains in
amylopectins. These longer chains require a higher temperature to completely dissociate
than that required for shorter double helices. Additionally, Vamadevan et al. (2013) also
reported that the internal structure of amylopectin correlates with To of starches from
different botanical sources. The higher transition temperatures and ΔH observed in the
ARA starches compared to the Asian rice starches may therefore be due to differences in
their amylopectin internal structures. Chung et al. (2011) reported that the differences in
To, Tp, Tc and ΔH in starches from different rice cultivars may be attributed to differences
in amylose content, granular structure, molecular weight, and amylose/amylopectin ratio.
135
A linear relationship was observed between the amylose content of the starches and their
gelatinization temperatures.
Correlation between amylose and To, Tp, Tc and ΔH at
(p<0.05) was 0.74, 0.73, 0.7 and 0.64 respectively (Table 5.5). Figure 5.12 shows the
multiple linear regression models to describe the relationship between To, Tp, Tc, ΔH and
amylose and the equation of the fitted models. When the gelatinized starches were stored
for 14 days at 4°C to monitor retrogradation, To, Tp, Tc and ΔH for the retrograded gels
were found to have reduced drastically when compared to the native granules.
The
percentage retrogradation (calculated as ΔH of retrograded gel/ΔH of starch granules x
100) was between 44.8 and 71.5% (Table 5.4) and the least retrogradation was observed in
Koshihikari starch possibly because it had the lowest amylose concentration. The amylose
content of the starches reported in this study (Table 5.1) was consistent with the range
reported by Lawal et al. (2011) for improved rice varieties in West Africa and within the
range of values reported for rice starches from other sources (Peisong et al. 2004;
Wickramasinghe and Noda 2008; Wang et al. 2010; Yu et al. 2012). Generally, the ratio of
amylose to amylopectin is crucial in affecting many physico-chemical parameters of
starches, since the amorphous component of starches is made up mainly of amylose,
(Lawal et al. 2011).
5.5 Conclusion
This study sheds light on the structural and thermal characteristics of starches isolated from
seven African rice accessions that were selected from a population being screened for
further development at the Africa Rice Center. These characteristics provide important
information for useful applications of the starches and may also influence the advanced
136
selection of accessions of interest for the development of suitable varieties for cooking
quality and specific end use. The Asian rice starches, apart from having similar physical
and morphological characteristics as ARA starches, differed in their molecular composition
(amylose content) and structure, which might have influenced their functional properties.
But, NERICA 4 in many ways had properties similar to the starches from the ARAs. The
differences in physicochemical properties that were observed are likely due to differences
in amylose content. The results also suggest differences between the structure and granular
architecture of starches from the ARAs and Asian rice varieties.
5.6 Acknowledgements:
A special appreciation is hereby extended to the Global Rice Science Partnership (GRiSP)
and Africa Rice Center (AfricaRice) for the provision of funding and rice samples for this
work.
137
5.7 Tables and Figures
Table 5.1. Characteristics and Composition of Starches Isolated From African Rice
Accession and Asian Rice Varieties#
Starch Starch
AMP
AM
LCAM:
LCAM SCAM
Sample
yield damage
(%)
(%)
SCAM
(%)
(%)
African rice
IRGC 80787
70.9
1.02a
71.3
28.7d
11.2cd
16.9c
0.67bcd
IRGC 86741
66.5
1.16d
74.5
25.5c
10.8c
14.7b
0.74de
IRGC 103759
68.2
1.18def
71.8
28.2d
10.7c
17.5cd
0.62ab
RAM 123
72.3
1.21f
70.3
29.7de
10.8c
19.0d
0.57a
TOG 7682
72.9
1.06b
71.5
28.5d
11.0c
17.5cd
0.63abc
TOG 12440
65.7
1.02ab
71.8
28.2d
10.8c
18.0cd
0.60ab
TOG 14365
67.2
1.11c
68.9
31.1e
12.2d
17.5c
0.70cd
O. sativa x O. glaberrima cross
69.7
1.17de
76.2
23.8bc
10.5bc
13.2b
0.8ef
Koshihikari
71.7
1.20ef
83.0
17.0a
7.9a
9.1a
0.87f
WITA 4
69.2
1.15cd
77.1
22.9b
9.5b
13.4b
0.71d
AM – Amylose
AMP - Amylopectin LCAM – Long Chain Amylose
NERICA 4
Asian rice
SCAM – Short Chain Amylose
#
Values with the same letters are not significantly different within columns
138
Table 5.2. Pasting Properties of Native Starches Isolated From African Rice Accessions
and Asian Rice Varieties#
PKT
(min)
PT
(°C)
PV
(RVU)
TV
(RVU)
FV
(RVU)
BDV
(RVU)
SB
(RVU)
TSB
(RVU)
IRGC 80787
5.2b
78.6e
277.5e
187.6e
332.3d
89.9b
54.8de
144.7cde
IRGC 86741
5.7d
77.8cd
251.4cd
182.7e
315.0c
68.7a
63.7e
132.4bc
IRGC 103759
5.2b
78.1d
253.6cd
145.4ab
311.4c
108.2de
57.8de
166.0g
RAM 123
5.0a
77.8cd
231.5a
134.8a
274.0ab
96.8c
42.5c
139.3cd
TOG 7682
5.2b
77.8cd
263.3d
165.5cd
318.2c
97.8c
54.9de
152.7ef
TOG 12440
5.2b
78.9e
235.6ab
139.3ab
285.0b
96.3c
49.4cd
145.7de
TOG 14365
5.0a
77.9cd
253.5cd
149.8b
308.4c
103.7d
54.9de
158.6fg
Sample
African rice
O. sativa x O. glaberrima cross
5.3b
77.7c
284.5e
175.1de
307.7c
109.3e
23.2b
132.5bc
Koshihikari
5.2b
71.1a
308.9f
184.1e
266.3a
124.8f
-42.5a
82.2a
WITA 4
5.4c
73.1b
248.3bc
153.3bc
275.0ab
94.9bc
26.8b
121.7b
NERICA 4
Asian rice
PKT= Peak Time; PT= Pasting Temperature; PV= Peak Viscosity; TV= Trough Viscosity;
FV= Final Viscosity; BDV= Breakdown Viscosity; SB= Setback Viscosity;
TSB= Total Setback Viscosity
#
Values with the same letters are not significantly different within columns
139
Table 5.3. Gelatinization Parameters of Native Starches Isolated From African Rice
Accessions and Asian Rice Varieties#
Sample
African rice
To (°C)
Tp (°C)
Tc (°C)
ΔH (J/g) MR (Tc-To)
IRGC 80787
69.0ef
73.6d
83.0g
15.1bcd
14.0
IRGC 86741
68.6cd
72.7c
81.5cd
15.1bcd
12.9
IRGC 103759
68.6c
73.6d
82.2f
15.8d
13.6
RAM 123
68.5c
72.9c
81.8e
15.2bcd
13.3
TOG 7682
68.8cde
73.3d
81.7de
15.3bcd
12.9
TOG 12440
68.9def
73.3d
82.2f
15.6d
13.3
TOG 14365
69.1f
73.4d
81.4c
15.4cd
12.3
O. sativa x O. glaberrima cross
70.3g
74.4e
82.3f
14.5bc
12.0
Koshihikari
60.5b
66.9b
77.0b
14.3ab
16.5
WITA 4
59.8a
65.8a
75.0a
13.3a
15.2
NERICA 4
Asian rice
To = Onset Temperature; Tp = Peak Temperature; Tc = Conclusion Temperature;
ΔH = Enthalpy of Gelatinization. MR= Melting Range,
#
Values with the same letters are not significantly different within columns
140
Table 5.4. Melting Parameters of Retrograded Starch Gels of African Rice Accessions
and Asian Rice Varieties#
MR
RT
Sample
To (°C)
Tp (°C)
Tc (°C) ΔH (J/g)
(°C)
(%)
African rice
IRGC 80787
44.1c
54.6e
71.3g
9.8bc
27.2
64.9
IRGC 86741
43.1ab
53.5bcd
71.1f
10.8c
28.0
71.5
IRGC 103759
44.0c
53.8cde
72.3h
10.8c
28.3
68.4
RAM 123
43.7bc
53.8cde
70.0d
9.2b
26.3
60.5
TOG 7682
43.8bc
53.2abc
70.1d
10.0bc
26.3
65.4
TOG 12440
42.8a
52.5a
70.0d
10.1bc
27.2
64.7
TOG 14365
43.1ab
52.9abc
70.4e
10.6c
27.3
68.8
43.1ab
54.2de
69.2c
9.2b
26.1
63.4
Koshihikari
42.9a
52.8ab
65.5a
6.4a
22.6
44.8
WITA 4
42.9a
52.4ab
67.5b
7.3a
24.6
54.9
O. sativa x O. glaberrima cross
NERICA 4
Asian rice
To = Onset Temperature; Tp = Peak Temperature; Tc = Conclusion Temperature; ΔH =
Enthalpy of Gelatinization MR= melting range, RT%=Percentage retrogradation after 14
days of storage at 4°C.
#
Values with the same letters are not significantly different within columns
141
Table 5.5. Correlations between Amylose Concentration and Some Measured Parameters#
Amylose
Amylose
TO
TP
TC
ΔH
Peak
viscosity
Final
viscosity
Breakdown
viscosity
Setback
Total
setback
K/S@
λmax550
CNS
CIDS
#
0.7406
0.7278
0.6960
0.6429
-0.6547
TO
TP
TC
ΔH
Peak
viscosity
Final
Breakdown Setback Total
K/S@
CNS
viscosity
viscosity
setback λmax550
0.9953
0.9721 0.9824
0.7063 0.7232 0.7651
NS
NS
NS
NS
0.5130
0.6709 0.6812 0.6616
NS
0.8651
NS
NS
NS
NS
0.7040 0.6791 0.6360 0.5668
0.5343
-0.6982
NS
0.6547
-0.6951
0.8998
0.7371 0.7385 0.6807 0.6800
-0.5962
0.6191
NS
0.9389
NS
0.7963 0.7765 0.7590 0.6891
NS
NS
NS
0.4726
-0.6239
NS
0.5997
-0.5711
NS
NS
NS
NS
NS
NS
0.4553 0.5069
NS
NS
NS
NS
0.8970
0.9041 0.8463
NS
NS
NS
NS
NS
NS
NS
NS
Correlations significant at p<0.05
CNS = Crystallinity of Starch; CIDS= Crystallinity of Iodinated Starch; To = Onset Temperature; Tc = Conclusion Temperature;
Tp = Peak Temperature; ΔH = Enthalpy of Gelatinization
142
CIDS
Figure 5.1. SEM Micrographs of Isolated Starch from African Rice Accessions and
Asian Rice Varieties.
143
18
16
14
IRGC 80787
IRGC 86741
12
Volume (%)
IRGC 103759
RAM 123
10
TOG 7682
8
TOG 12440
TOG 14365
6
NERICA 4
Koshihikari
4
WITA 4
2
0
0
5
10
15
20
25
30
35
Granule size (um)
Figure 5.2. Particle Size Distribution of Starches Isolated from African Rice Accessions
and Asian Rice Varieties.
144
0.006
0.005
IRGC 80787
IRGC 86741
Intensity (cps)
0.004
IRGC 103759
RAM 123
0.003
TOG 7682
TOG 12440
0.002
TOG 14365
NERICA 4
0.001
KOSHIHIKARI
WITA 4
0
5
10
15
20
25
30
35
2θ (degree)
Figure 5.3. Wide Angle X-Ray Diffractograms of Starches Isolated from African Rice
Accessions and Asian Rice Varieties.
145
Intensity (cps)
IRGC 80787
IRGC 86741
CNS=31.3
CIS=35.4
TOG 12440
IRGC 103759
CNS=32.1
CIS=36.7
TOG 14365
CNS=30.7
CIS=35.6
RAM 123
CNS=31.2
CIS=32.9
CNS=30.5
CIS=36.1
Koshihikari
NERICA 4
CNS=29.5
CIS=31.2
TOG 7682
CNS=31.6
CIS=34.5
WITA 4
CNS=31.0
CIS=35.1
Diffraction angle 2θ
Figure 5.4. Wide Angle X-Ray Diffractograms of Starch and Iodinated Starches from African and Asian Rice.
(Intensity (cps) vrs. Diffraction Angle (2θ degree)).
Black and blue lines represent starch and iodinated starches respectively; CNS = % crystallinity of starch;
CIS = % crystallinity of iodinated starch
Arrows show the shift in peaks at 15°, 17° and 18° 2θ° observed in the iodinated ARAs.
146
CNS=32.5
CIS=32.2
CNS=29.5
CIS=31.2
40
35
IRGC 80787
30
IRGC 86741
IRGC 103759
K/S value
25
RAM 123
20
TOG 7682
TOG 12440
15
TOG 14365
10
NERICA 4
KOSHIHIKARI
5
WITA 4
0
400
450
500
550
600
650
700
Wavelength (nm)
Figure 5.5. K/S Spectra of Starches Isolated from African Rice Accessions and Asian Rice. Varieties
147
350
100
IRGC 80787
90
IRGC 86741
IGRC 103759
250
80
200
70
150
RAM 123
Temp. (°C)
Viscosity (RVU)
300
TOG 7682
TOG 12440
TOG 14365
60
100
NERICA 4
50
50
KOSHIHIKARI
WITA 4
40
0
0
5
10
15
Time (min)
Figure 5.6. RVA Pasting Profiles of Starches Isolated from African Rice Accessions
and Asian Rice Varieties.
148
Temp.
To
Tp
Tc
ΔH= 13.3 J/ga
WITA 4
ΔH= 14.3 J/gab
Koshihikari
ΔH= 14.5 J/gbc
NERICA 4
TOG 14365
ΔH= 15.4 J/gcd
TOG 12440
ΔH=15.6 J/gd
TOG 7682
ΔH= 15.3 J/gbcd
RAM 123
ΔH= 15.2 J/gbcd
IRGC 103759
ΔH=15.8 J/gd
IRGC 86741
ΔH= 15.1 J/gbcd
ΔH= 15.1 J/gbcd
IRGC 80787
55
60
65
70
75
80
85
Figure 5.7. Thermal Properties of Starches Isolated from African Rice Accessions and Asian
Rice Varieties.
149
Tp
To
Tc
ΔH= 7.3 J/g a
WITA 4
ΔH= 6.4 J/g a
Koshihikari
ΔH= 9.2 J/g b
NERICA 4
TOG 14365
ΔH= 10.6 J/g c
TOG 12440
ΔH=10.1 J/g bc
TOG 7682
ΔH= 10.0 J/g bc
ΔH= 9.2 J/g b
RAM 123
ΔH=10.8 J/g c
IRGC 103759
IRGC 86741
ΔH= 10.8 J/g c
IRGC 80787
ΔH= 9.8 J/g bc
40
45
50
55
60
Temperature (°C)
65
70
Figure 5.8. Thermal Properties of Retrograded Starch Gels from
African Rice Accessions and Asian Rice Varieties.
150
75
8
Carbohydrates (wt. %)
7
6
AMP
SCAM
LCAM
5
4
3
2
1
0
40
60
80
100
120
140
160
180
Fraction number
KOSHIHIKARI
NERICA 4
WITA 4
IRGC 86741
TOG 7682
IRGC 103759
IRGC 80787
TOG 14365
TOG 12440
RAM 123
Figure 5.9. Gel-Permeation Chromatograms of Debranched Rice Starches
Fractionated on Sepharose CL-6B.
(LCAM = Long Chain Amylose
SCAM = Short Chain Amylose
AMP = Amylopectin Chains)
151
35
700
IRGC 80787
35
M
30
650
IRGC 86741
700
M
30
650
35
700
IRGC 103759
650
25
15
10
450
5
5
0
400
15
25
35
45
55
65
0
75
35
15
25
35
45
55
65
450
5
400
0
35
700
TOG 14365
M
650
25
600
20
15
0
400
65
75
400
15
25
35
45
55
65
75
Figure 5.10. Gel-Permeation Chromatograms of Whole Rice Starches
and Their Corresponding β-LDs on Sepharose CL-2B.
Graph: Carbohydrate content (Weight %) on primary axis; λmax (nm)
on secondary axis vrs. Fraction number.
65
450
5
35
700
650
600
20
45
55
65
75
55
65
35
WITA 4
700
M
30
650
25
600
550
450
5
400
0
500
10
450
400
15
25
35
45
55
65
75
450
5
0
400
15
25
35
45
55
65
M=Maltose; Line graphs (
------- ) represent carbohydrate
content (weight %)
Symbols (
● represent λmax values (nm))
152
75
500
10
35
45
15
500
25
35
20
15
15
25
550
550
0
400
15
M
25
5
450
75
600
10
0
55
30
500
450
45
650
15
5
35
Koshihikari
20
450
10
400
25
NERICA 4
M
500
500
0
0
15
700
35
550
15
5
75
550
10
5
65
600
500
10
55
55
25
25
15
45
45
650
20
35
35
30
30
550
25
25
M
30
15
400
15
75
700
TOG 12440
20
10
450
600
550
500
10
650
25
15
500
10
700
M
600
550
15
15
500
TOG7682
30
20
550
35
650
600
20
20
550
M
30
25
25
600
20
700
RAM 123
M
30
25
600
35
75
Plot of Fitted Model
Plot of Fitted Model
340
320
300
Final viscosity
Peak viscosity
320
280
300
260
280
240
260
220
16
20
24
amylose
28
16
32
Peak viscosity = 358.423 - 3.703*amylose
r=-0.6547
24
amylose
28
32
Final vis. = 226.873 + 2.748*amylose
r= 0.5130
Plot of Fitted Model
Plot of Fitted Model
180
50
160
Total setback
70
30
setback
20
140
10
120
-10
100
-30
-50
80
16
20
24
amylose
28
setback = -131.55 + 6.452*amylose
r=0.8651
32
16
20
24
amylose
28
32
Total setback = 3.98 + 5.067*amylose
r=0.8998
Figure 5.11. Correlation between Amylose Concentration and Pasting Parameters.
153
Plot of Fitted Model
Plot of Fitted Model
71
75
69
73
peak temp
onset temp
67
71
65
69
63
67
61
65
59
16
20
24
amylose
28
16
32
Onset temp = 49.8519 + 0.657902*amylose
20
24
amylose
28
32
Peak temp = 58.3673 + 0.517402*amylose
Plot of Fitted Model
Plot of Fitted Model
84
16
82
enthalpy
concl temp
15
80
14
78
13
76
74
12
16
20
24
amylose
28
32
16
Concl. temp = 69.4606 + 0.43016*amylose
20
24
amylose
28
32
Enthalpy = 11.6829 + 0.123362*amylose
Figure 5.12. Multiple Linear Regression Models to Describe the Relationship between Amylose
and Thermal Parameters of Isolated Starches.
154
5.8 References
Adebowale, K. O., Olu-Owolabi, I. O., Olawunmi, E. K., and Lawal, O. S. 2005.
Functional properties of native, physically and chemically modified breadfruit
(Artocorpus altilis) starch. Industrial Crops and Products 21: 343-351.
Annor, G. A., Marcone, M., Bertoft, E., and Seetharaman, K. 2014.
Physical and
molecular characterization of millet starches. Cereal Chem. 91; 286-292.
Ahmed, J., Ramaswamy, H. S., Ayad, A., and Ali, I. 2008. Thermal and dynamic
rheology of insoluble starch from basmati rice. Food Hydrocoll. 22: 278-287.
Becker, A., Hill, S. E., and Mitchell, J. R. 2001. Milling-A further parameter affecting the
rapid visco analyser (RVA) profile. Cereal Chem. 78: 166-172.
Bertoft, E. 2004. On the nature of categories of chains in amylopectin and their connection
to the super helix model. Carbohydr. Polym. 57: 211-224.
Billmeyer, F.W., and Saltzman, M. 1981. The coloring of materials in industry. Principles
of color Technology, (Eds.), Wiley, New York: 133-172.
Chauhan, F., and Seetharaman, K. 2013. On the organization of chains in amylopectin.
Starch/Stärke. 65: 191-199.
Cheetham, N. W. H., and Tao, L. 1998. Variation in crystalline type with amylose content
in maize starch granules: An X-ray powder diffraction study. Carbohydr. Polym.
36: 277-284.
155
Chung, H.J., Liu, Q., Lee, L., and Wei, D. 2011. Relationship between the structure,
physicochemical properties and in vitro digestibility of rice starches with different
amylose contents. Food Hydrocoll. 25: 968-975.
Dhital, S.; Shrestha, A. K., and Gidley, M. J. 2010. Effect of cryo-milling on starches:
functionality and digestibility. Food Hydrocoll. 24: 152-163.
Fisher, D. K., and Thompson, D. B. 1997. Retrogradation of maize starch after thermal
treatment within and above gelatinization temperature range. Cereal Chem. 74:
344-351.
Futakuchi, K., and Sié, M. 2009. Better exploitation of African Rice (Oryza glaberrima
Steud.) in varietal development for resource-poor farmers in West and Central
Africa. Agric. J. 4: 96-102.
Gidley, M. J., Hanashiro, I., Hani, N. M., Hill, S. E., Huber, A., Jane, J.-L., Liu, Q., Morris,
G., Rolland-Sabaté, A., Striegel, A. M., and Gilbert, R. G. 2010. Reliable
measurements of the size distributions of starch molecules in solution: Current
dilemmas and recommendations. Carbohydr. Polym. 79: 255-261.
Hasjim, J., Srichuwong, S., Scott, M. P., Jane, J.-L. 2009. Kernel composition, starch
structure, and enzyme digestibility of opaque-2 maize and quality protein maize. J.
Agric. Food. Chem. 57: 2049-2055.
Hatcher, D. W., Anderson, M. J., Desjardins, R. G., Edwards, N. M., and Dexter, J. E.
2002. Effects of flour particle size and starch damage on processing and quality of
white salted noodles. Cereal Chem.79: 64-71.
156
He, G. Q., Song, X. Y., Ruan, H., and Chen, F. 2006. Octenyl succinic anhydride modified
early indica rice starches differing in amylose content. J. Agric. Food Chem. 54:
2775-2779.
Imberty, A., and Perez, S. 1988. A revisit to the three-dimensional structure of B-type
starch. Biopolym. 27: 1205-1221.
Kim, H.-S., and Huber, K. C. 2008. Channels within soft wheat starch A- and B-type
granules. J. Cereal Sci. 48: 159-172.
Lawal, O. S., Lapasin, R., Bellich, B., Olayiwola, T. O., Cesàro, A., and Yoshimura, M.
011. Rheology and functional properties of starches isolated from five improved
rice varieties from West Africa. Food Hydrocoll. 25: 1785-1792.
Li, J. Y., and Yeh, A. I. 2001. Relationships between thermal, rheological characteristics
and swelling power for various starches. J. Food Eng. 50: 141-148.
Mohan, B.H., Gopal, A., Malleshi, N.G., and Tharanathan, R.N. 2005. Characteristics of
native and enzymatically hydrolyzed ragi (Eleusine coracana) and rice (Oryza
sativa) starches. Carbohyd. Polym. 59: 43-50.
Morrison, W. R., Tester, R. F., and Gidley, M. J. 1994. Properties of damaged starch
granules II. Crystallinity, molecular order and gelatinisation of ball-milled starches.
J. Cereal Sci. 19: 209-217.
Nishita, K. D., and Bean, M. M. 1982. Grinding methods: Their impact on rice flour
properties. Cereal Chem. 59: 46-49.
157
Nowakowski, D., Sosulski, F. W., and Hoover, R. 1986. The effect of pin and attrition
milling on starch damage in hard wheat flours. Starch/Stärke 38: 253-258.
Peisong, H., Haijun, Z., Zhiying, D., Zhang, L., and Dianxing, W. 2004. Starch
digestibility and the estimated glycemic score of different types of rice differing in
amylose contents. J. Cereal Sci. 40: 231-237.
Raina, C. S., Singh, S., Bawa, A. S., and Saxena, D. C. 2007. A comparative study of
Indian rice starches using different modification model solutions. LWT-Food
Science and Technology. 40: 885-892.
Ratnayake, W.S., Hoover, R., Shahidi, F., Perera, C., and Jane, J. 2001. Composition,
molecular structure, and physicochemical properties of starches from four field pea
(Pisum sativum L.) cultivars. Food Chem. 74: 189-202.
Saibene, D., Zobel, H. F., Thompson, D. B., and Seetharaman, K. 2008. Iodine‐binding in
granular starch: Different effects of moisture content for corn and potato starch.
Starch/Stärke. 60: 165-173.
Sargeant, J. 1982. Determination of amylose: Amylopectin ratios of starches.
Starch/Stärke. 34: 89-92.
Schulman, A.H., and Kammiovirta, K. 1991. Purification of barley starch by protein
extraction. Starch/Stärke 43: 387-389.
Singh, N., Singh, J., Kaur, L., Sodhi, N. S., and Gill, B. S. 2003. Morphological, thermal
and rheological properties of starches from different botanical sources. Food
Chem. 81: 219-231.
158
Singh, N., Nakaura, Y., Inouchi, N., and Nishinari, K. 2007. Fine structure, thermal and
viscoelastic properties of starches separated from Indica rice cultivars.
Starch/Stärke 59: 10-20.
Sweeney, M., and McCouch, S. 2007. The complex history of the domestication of rice.
Ann. Bot. 100: 951-957.
Tester, R. F. 1997. Properties of damaged starch granules: Composition and swelling
properties of maize, rice, pea and potato starch fractions in water at various
temperatures. Food Hydrocoll. 11: 293-301.
Vamadevan, V., Bertoft, E., and Seetharaman, K. 2013. On the importance of organization
of glucan chains on thermal properties of starch. Carbohydr. Polym. 92: 16531659.
Vamadevan, V., and Bertoft, E. 2015. Structure-function relationships of starch
components. Starch/Stärke. 67: 55-68.
Vandeputte, G. E.,Vermeylen R., Geeroms, J., and Delcour, J. A. 2003. Rice starches I.
Structural aspects provide insight into crystallinity characteristics and gelatinization
behavior of granular starch. J. Cereal Sci. 38: 43-52.
Vandeputte, G. E., and Delcour, J. A. 2004. From sucrose to starch granule to physical
behaviour: a focus on rice starch. Carbohydr. Polym. 58: 245-266.
Waduge, R., Xu, S., and Seetharaman, K. 2010. Iodine absorption properties and its effect
on the crystallinity of developing wheat starch granules. Carbohydr. Polym. 82:
786-794.
159
Wang, L. F., and Wang, Y. J. 2004. Rice starch isolation by neutral protease and high
intensity ultrasound. J. Cereal Sci. 39: 291-296.
Wang, L., Xie, B., Shi, J., Xue, S., Deng, Q., Wei, Y., and Tian, B. 2010. Physicochemical
properties and structure of starches from Chinese rice cultivars. Food Hydrocoll.
24: 208-216.
Wang, L., Xie, B., Xiong, G., Du, X., Qiao, Y., and Liao, L. 2012. Study on the granular
characteristics of starches separated from Chinese rice cultivars. Carbohydr. Polym.
87: 1038-1044.
Wani, A. A., Singh, P., Shah, M. A., Schweiggert-Weisz, U., Gul, K., and Wani, I. A.
2012. Rice starch Diversity: Effects on structural, morphological, thermal, and
physicochemical properties-A review. Comprehensive Reviews in Food Science
and Food Safety. 11: 417-436.
Wickramasinghe, H. A. M, and Noda, T. 2008. Physicochemical properties of starches
from Sri Lankan rice varieties. Food Sci Technol Res 14: 49-54.
Yamin, F.F., Lee, M., Pollak, L.M., and White, P.J. 1999. Thermal properties of starch in
corn variants isolated after chemical mutagenesis of inbred lines B73. Cereal
Chem. 76:175-181.
Yu, S., Ma, Y., Menager, L., and Sun, D. 2012. Physicochemical properties of starch and
flour from different rice cultivars. Food Bioprocess Technol. 5: 626-637.
160
Zaidul, I. S. M., Norulaini, N. A. N., Omar, A. K. M., Yamauchi, H., and Noda, T. 2007.
RVA analysis of mixtures of wheat flour and potato, sweet potato, yam, and
cassava starches. Carbohydr. Polym. 69:784-791.
Zhong, F., Li, Y., Ibáñez, A. M., Oh, M. H., McKenzie, K. S., and Shoemaker, C. 2009.
The effect of rice variety and starch isolation method on the pasting and rheological
properties of rice starch pastes. Food Hydrocoll. 23: 406-414.
161
CHAPTER 6: UNIT AND INTERNAL CHAIN PROFILES OF AFRICAN RICE
(Oryza glaberrima) AMYLOPECTIN
162
Abstract
African rice and Asian rice are two distinct types of domesticated rice. Little is known
about structure of starch in African rice and its impact on rice quality.
This work
investigated the fine structure of the starches isolated from African rice in comparison to
Asian rice (globally consumed rice). The unit chain profiles of amylopectins and their
ϕ, β-limit dextrins from two African rice accessions (ARAs), namely TOG 12440 and
IRGC 103759, were compared with two Asian rice samples, namely cv Koshihikari
(spp. japonica) and cv WITA 4 (spp. indica) and NERICA 4 (an inter specific cross
between sativa x glaberrima). The debranched starch samples were studied by highperformance anion-exchange chromatography. The average chain lengths (CL), internal
chain length (ICL), external chain length (ECL), and total internal chain length (TICL) of
the amylopectins showed narrow ranges among all the rice investigated. The short:long
(S:L) chain ratio ranged between 12.1-13.8, and the relative molar concentration of Bchains was approximately 50% in all samples, and thus the ratio of A:B-chains was 1.0.
Significant differences were, however, observed in the proportions of very short
"fingerprint" B-chains (Bfp, DP 3-7). The ARAs and NERICA 4 had higher levels of
Bfp-chains than the Asian rice but the major group of short B-chains (DP 8-25) was similar
in all samples.
Positive correlations (p<0.05) were observed between ECL and
gelatinization transition temperatures and enthalpy. The study results strongly suggest that
the structure of amylopectin of African rice is different from that of Asian rice and this may
account for the higher gelatinization enthalpies observed in the ARAs. The differences
observed also suggest possible dissimilarities in the fine structure (clusters and building
blocks) of their amylopectins.
Key words: African rice, amylopectin, φ,β-limit dextrins, unit chain, internal chain
163
Introduction
Oryza sativa (Asian rice) and Oryza glaberrima (African rice) are the two distinct types of
domesticated rice with Asian rice having a global distribution particularly in Asia. African
rice is cultivated mainly in the West Africa sub region. The predominant component of
rice is starch (90% dry basis) and its composition (including apparent amylose content and
amylopectin structure) and physicochemical properties (i.e. gelatinization temperature and
pasting viscosity) determine the eating, cooking and milling qualities of rice (Juliano 1998;
Bao 2012).
Starch contains three types of bio-macromolecules: amylopectin (highly branched
molecule), amylose (almost linear but may have few branches molecule), and possibly an
intermediate molecule, which is found in some kinds of starch as in oat and some mutant
maize genotypes exhibiting a structure intermediate to those of amylose and amylopectin
(Wang and White 1994; Wang et al. 1993). Variations exist in the contents of these three
components in starches from different sources, but amylopectin is commonly considered as
the major component in storage starch and accounts for up to 85% by weight (Nakamura
2002).
Amylose is a linear polymer that is made up of almost entirely α-1,4-linked
D-glucopyranosyl units, however, several amylose molecules have between 0.3% and 0.5%
α-1,6-linked D-glucopyranose branches (Whistler and BeMiller 1997). Amyloses from
rice starch have been reported to have average degree of polymerization (DP) = 920 to
1110 and they are slightly branched with 2 to 5 chains on the average (Takeda et al. 1986).
Amylopectin unit chains are broadly divided into A-, B- and C-chains. A-chains which do
not carry other chains are attached through their reducing end to B-chains, while B-chains
164
are linked in the same way and carry one or more A-chains. Additionally, each molecule
contains a single C-chain that contains the only reducing group of the amylopectin
molecule and carries other chains (Peat et al. 1952).
The building block backbone model (Bertoft 2013) and the cluster model proposed by
(French 1972; Nikuni 1978) and modified by Hizukuri (1986) are the two key models of
amylopectin structure that are discussed in recent times. In the two-directional backbone
model, the short chains are anchored perpendicular to the long chains which are mainly
present in the amorphous region and do not participate in the formation of the crystalline
lamellae whereas in the cluster model, the clusters are interconnected by long chains which
participate in the formation of the crystalline and amorphous lamellae. The amylopectin
molecule consists of three types of unit chains, denoted as A-, B- and C-chains. Hizukuri
(1986) suggested that the A-chains are the shortest chains, B1-chains with a DP of 20-24
are short while B2-chains (DP 42-48) and B3-chains (DP69-75) are much longer. Exoacting enzymes such as phosphorylase and β-amylase have been used to isolate the internal
part of amylopectin through the removal of external chains. β-amylase produces maltose
by acting on glucan chains from their non-reducing ends. Due to the inability of β-amylase
to by-pass a branch point in the amylopectin molecule, the resulting β-limit dextrin with the
branches denotes the internal part of the amylopectin molecule. Phosphorylase a from
rabbit muscle produces φ-limit dextrins (Manners 1989).
Practically, isolation of the
internal part of amylopectin is done by treating amylopectin successively with both
enzymes to generate φ,β-limit dextrins (φ,β-LD), in which the A-chains are reduced into
maltosyl stubs, and the external segments of the B-chains into a single residue.
Debranching of the φ,β-LD reveals the internal chains, which can be analysed by
165
techniques such as anion-exchange chromatography (Bertoft 2004).
Zhu and Bertoft
(1996) reported the use of limit dextrins in studying the internal structure and ratio of A:B
chains in amylopectin molecules. After analyzing the internal unit chain distribution in
amylopectins from different botanical sources, Bertoft et al. (2008) categorized the internal
chains (which all are B-chains), into short (BS) and long chains (BL). The BS-chains were
further divided into very short “fingerprint” Bfp-chains with a degree of polymerisation
(DP) 3-7 and BSmajor, with DP ∼8-25. On the other hand,
BL-chains were divided into
B2-chains (DP ∼26–50) and B3-chains (DP > 50).
Bertoft et al. (2008) and Manners (1989) classified amylopectin into external and internal
chains where the external chains are found in the crystalline part of the amylopectin
molecule and the internal chains are mainly present in the amorphous lamellae. The entire
A-chains are external since they do not carry other chains while the B-chains are involved
in both external and internal divisions. From the internal chain profile the total internal
chain length (TICL), which is defined as the average length of the B-chains without their
external chains, can be estimated (Bertoft 1991). The length of the segments of the chains
that extend from the outermost branch to the non-reducing end of the chains (ECL) and the
average internal chain length (ICL), defined as the length of the chain segment between
two branches in the amylopectin molecule, can be estimated from a comparison of the
average chain length (CL) of the amylopectin and CL of the limit dextrin (Manners 1989).
The internal structure of amylopectins from diverse botanical sources have been reported
over the past years (Bertoft 2013; Zhu et al. 2011; Kong et al. 2009; Laohaphatanaleart et
al. 2009; Bertoft et al. 2008; Klucinec and Thompson 2002 and Bertoft and Koch 2000).
166
Bertoft et al. (2008) on the basis of internal chain profile categorized amylopectins into
four structural types, of which medium amylose-containing Asian rice was found to be type
2. Since the basic structure of African rice starches have not been characterized yet, this
study focuses on investigating the internal structure of two accessions of African rice
amylopectins in the form of φ,β-limit dextrins. They were compared with two Asian rice
varieties and an interspecific cross between African rice and Asian rice.
6.2
Materials and Methods
6.2.1
Materials
Five dehusked and polished rice samples were provided by the Africa Rice Center based in
Cotonou, Benin for this study. Two of the samples were African rice accessions (ARAs) IRGC 103759 and TOG 12440. The others are two O. sativa varieties – cv Koshihikari and
cv. WITA 4 and NERICA 4 (an interspecific cross between African rice x Asian rice). All
chemicals and solvents used were of ACS certified grade.
6.2.2
Methods
6.2.3
Starch Isolation
Starch was isolated from the rice samples using the method of Waduge et al. (2010) with
some modifications as previously described in chapter 4.
6.2.4
Isolation of Amylopectin from Rice Starch
The method described by Klucinec and Thompson (1998) was employed with some
modifications to isolate amylopectin from the rice starches. Forty mL of 90% dimethyl
167
sulfoxide (DMSO) was used to disperse 2 g of granular rice starch in a water bath (90 °C)
for 3 h under nitrogen atmosphere. During this time the mixture was constantly stirred
with a magnetic stirrer at low speed. The dissolved starch was gradually cooled to 80 °C
after which 4 portions of ethanol (98%) (160 mL) were added drop-wise with the aid of a
peristaltic pump (Easy-Load®Masterflex® Model 7518-12) to precipitate the starch out of
solution. The starch-ethanol mixture was slowly cooled down to room temperature (25 °C)
and then centrifuged at 8000 x g for 15 min at 4 °C. The supernatant was discarded and the
precipitate (non-granular starch) was washed with ethanol. The non-granular starch was
again dissolved in DMSO (90%) (56 mL) in a water bath at 90 °C while was constantly
stirred for 1h under nitrogen atmosphere. The temperature was cooled to 80 °C and a
vigorously stirred mixture of 1-butanol (23.5 mL) and isoamyl alcohol (23.5 mL) in 324
mL of water was added drop-wise using a peristaltic pump for a period of 1-1.5 h under
nitrogen. Thereafter the mixture was continuously stirred for an additional 30 min under
nitrogen.
The mixture was allowed to cool gradually overnight to room temperature
(25 °C) in an insulated foam box. The mixture was gently agitated and centrifuged at
10,000 x g for 30 min at 4 °C after which the supernatant (which is the amylopectin
portion) was cautiously decanted and concentrated to 50 mL in a rotary evaporator at
60 °C. A mixture of 1-butanol-isoamyl alcohol-water (1:14) was added to the concentrated
supernatant to precipitate any residual amylose after which the centrifugation step was
repeated. The supernatant (amylopectin) was decanted and concentrated to 50 mL in a
rotary evaporator at 60 °C. Amylopectin was precipitated out of solution with 3 portions of
methanol (150 mL) overnight at room temperature after which centrifugation at 22, 360 x g
for 20 min at 4 °C was followed. The precipitate was collected and dissolved in warm
168
water (20 mL) and 3 portions of ethanol (98%) (60 mL) was used to re-precipitate
amylopectin for 1 h at room temperature. The precipitated sample (amylopectin) was
finally centrifuged at 22, 360 x g for 20 min at 4 °C. The precipitate was collected,
dissolved in warm water (20 mL) and freeze-dried.
6.2.5
Determination of Purity of Amylopectin by Gel Permeation Chromatography
The purity of the isolated amylopectin fractions isolated from rice starch was determined
by analyzing debranched samples by gel-permeation chromatography according to the
procedure reported by (Laohaphatanaleart et al. 2009) on a column (1.6 cm x 90 cm) of
Sepharose CL-6B (Pharmacia, Uppsala, Sweden) eluted with 0.5 M aqueous NaOH at
0.5 mL/min. Purified amylopectin (2.0 mg) was dissolved in 90% DMSO (50 μL) by
employing heat for about 15 min in boiling water bath with constant magnetic stirring. The
amylopectin solution was then diluted with 400 μL warm water followed by the addition of
50 μL sodium acetate buffer (pH 5.5, 0.01 M). The solution was allowed to cool to room
temperature after which 1 μL each of isoamylase and pullulanase were added to initiate
debranching of the amylopectin. The debranching reaction lasted for 12 h with constant
stirring at room temperature. Thereafter, 50 μL NaOH (5 M) was added to halt the reaction
by inactivating the debranching enzymes. Aliquots (200 μL) were injected onto a column
(1.6 cm × 90 cm) of Sepharose CL-6B (Pharmacia, Uppsala, Sweden). The eluent used
was sodium hydroxide (0.5 M) with a pumping speed of 0.5 mL/min. Fractions (0.5 mL)
were collected analyzed for carbohydrate content by the phenol-sulphuric acid method
(Dubois et al. 1956). The column was calibrated with both linear and branched dextrins of
known DP (Bertoft and Spoof 1989).
169
Production of φ,β-Limit Dextrin of Amylopectin
The procedure reported by Bertoft (2004) with modifications by Kalinga et al. (2013) was
employed for the production of φ,β-limit dextrins (φ,β-LDs) from amylopectin.
Amylopectin (100 mg) was weighed and dissolved in 90 % DMSO (3 mL) in a beaker on a
water bath for 10 min at room temperature and constantly stirred gently with a magnetic
stirrer overnight. The dissolved amylopectin was then diluted with of warm water (~80°C,
32.7 mL), phosphate buffer-pH 6.8 (3.6 mL) and EDTA solution
(1.7 mL). Freshly
prepared solution of phosphorylase a from rabbit liver
(0.9 U/mL - Sigma)
(9 mL) was added and incubated overnight at room temperature bringing the concentration
of carbohydrate in solution to 2 mg/mL. The enzyme was inactivated by heating in boiling
water bath for 5 min and thereafter, glucose 1-phosphate (G-1-P) and maltose formed in the
reaction were removed by tangential flow filtration (Minimate™ TFF System, Canton,
MA), using a filter with a molecular weight cut-off limit of 10,000 Da. After the filtration
process, phosphate buffer, EDTA solution and phosphorylase a were added to the filtrate
and incubated overnight. Additional G-1-P and maltose generated were filtered out with
the tangential flow filtration system after inactivating enzymes. β-Amylase from barley
(Megazyme 28, 400 U/mL) (2 µL) and phosphate buffer-pH 6.8 (6.6 mL) were added and
incubated overnight. G-1-P and maltose generated were filtered out with the tangential
flow filtration system after inactivating enzymes. The addition of β-amylase was repeated
and the resulting φ,β-LDs were recovered and freeze-dried.
170
6.2.6
Analysis of Unit Chain Distribution
Two mg of amylopectin or φ,β-limit dextrin was dissolved in 90% DMSO (50 μL) with
gentle and continuous stirring overnight. The solution was diluted with warm water (400
μL) (80 °C) after which 0.01 M sodium acetate buffer (50 μL) (pH 5.5) was added.
Pullulanase M1 (1 μL) and isoamylase (1 μL) (Megazyme International, Ireland) were then
added to the mixture and stirred overnight at room temperature to debranch the structure of
the amylopectin or φ,β-limit dextrin. The enzymes were inactivated by boiling for 5 min
and the volume of the mixture was adjusted to bring the carbohydrate concentration to 1
mg/mL. The sample was further passed through a 0.45 μm nylon filter after which an
aliquot (25 μL) was injected into the Dionex ICS 3000 high-performance anion-exchange
chromatography (HPAEC) system (Dionex Corporation, Sunnyvale, CA, USA) equipped
with a pulsed amperometric detector (PAD), CarboPac PA-100 ion-exchange column
(4 x 250 mm) and a similar guard column (4 x 50 mm). The samples were then eluted with
a flow rate of 1 mL/min. The eluents used were A (150 mM sodium hydroxide) and
B (150 mM sodium hydroxide containing 500 mM sodium acetate). An elution gradient
(130 min) was employed by mixing eluent B into eluent A as follows: 0-9 min, 15-36%
B; 9-18 min, 36-45% B; 18-110 min, 45-100% B; 100-112 min, 100-15% B; 112-130 min,
15% B was used. The column was equilibrated with 15% eluent B for 60 min between
runs. The areas under the chromatograms were converted to carbohydrate concentration
using the method of Koch et al. (1998).
171
6.2.7
Statistical Analysis
The statistical programme Statgraphics Centurion XV, Version 15.1.02 (StatPoint,
Warrenton, VA, USA) was employed to conduct statistical analysis. One-way analysis of
variance was performed to determine significant differences between any pair of means at
the 95% confidence level and multiple range tests were done to determine which means
were different.
6.3 Results and Discussion
6.3.1
Purity of Amylopectin
An example of the comparison of a purified amylopectin sample with its corresponding
whole starch is shown in Fig. 6.1. In the chromatograms, the short chains of amylopectin
eluted after the long chains of amylose from the whole starch and debranched amylopectin
samples. Fractions 46-104 were considered to be the contribution of residual amylose
while fractions 105-168 were attributed to the debranched amylopectin chains.
The
debranched amylopectin exhibited a major peak at fractions 134-138. In this study, the
purity of the amylopectins was >95% for all samples. The shortfall of 100% purity could
be attributed to the presence of trace amounts of amylose and/or long chains of
amylopectin designated as “super long amylopectin chains”, which are reported to exist in
some amylopectins, particularly in some rice starches (Koroteeva et al. 2007; Takeda et al.
1987).
Laohaphatanaleart et al. (2009) reported that the super long chains of rice
amylopectin constitute 3.4-8.6% of rice starches.
172
6.3.2
Unit Chain Length Distributions of Amylopectins
The profiles of the unit chain distribution of the debranched amylopectins of the rice
samples analyzed with HPAEC are presented in Figure 6.2. Two sets of profiles were
found for the samples.
The first profile, which corresponded to the two ARAs and
NERICA 4, had chain distributions showing peaks at DP 12 and DP 13, respectively. The
second profile, which corresponded to the Asian rice samples, had quite distinct patterns.
Both showed highest peaks at DP 12, just as NERICA 4 and the ARAs, but a peak at DP 14
was also present. In addition, they possessed a shoulder, which occurred at DP 19. This
was consistent with descriptions of patterns for waxy and non-waxy rice amylopectins
reported by Hanashiro et al. (1996) and Laohaphatanaleart et al. (2009). The short and
long chains of the distribution could be divided at DP 35 for all samples except NERICA 4,
for which it occurred at DP 34.
Additionally, the peak DP of the long chains for
Koshihikari, WITA 4 and IRGC 103759 occurred at DP 43, while that of NERICA 4 and
TOG 12440 occurred at 42 and 44, respectively.
6.3.3
Unit Chain Length Distributions of φ,β-Limit Dextrins
Bertoft (1989) reported that the external chains in amylopectin are hydrolyzed
consecutively by phosphorylase a and β-amylase to produce their corresponding φ,β-limit
dextrins, and as a result of the debranching action of the enzymes, the A-chains appear as a
maltose peak and the rest as B-chains if a chromatogram of the chain distribution is drawn.
Figure 6.3 shows the chromatograms of the B-chains and thus reflects the composition of
the internal chains of amylopectin. In general, all internal unit chain profiles possessed the
two major groups of short and long chains as found in the whole amylopectins. However,
173
while the short chains (S-chains) of amylopectin are mixtures of short A- and short
B-chains (BS- or B1-chains), the S-chains of the φ,β-limit dextrins represent only
BS-chains, (Bertoft et al. 2008). The A-chains in the form of maltose have been removed
by filtration during the analysis and thus it did not appear in the chromatograms in Figure
6.3. At the lowest position in the chromatograms (weight-based) a division between the
short (BS) and long chains (BL) was made at DP 22 except for Koshihikari which was at
DP 24. This was comparable to ~DP 21 that was reported for rice (Bertoft et al. 2008).
The peak-DP of the various B-chain categories was the same for all samples. All the rice
amylopectin samples possessed a clearly distinguished group of Bfp-chains (DP 3-7) with a
peak position at DP 5. It was also evident that the ARAs and NERICA 4 had the same
peak-DP (31) of the long B-chain categories. However, the Asian rice samples had slightly
different peak-DPs and it was higher (32) in WITA 4 but lower in Koshihikari (29).
The average chain length classifications of different chain categories in φ,β-limit dextrins
and their parent amylopectins as obtained by HPAEC analysis are presented in Table 6.1.
The φ,β-limit value, average chain length (CL) and external chain length (ECL) of the
original amylopectins were estimated (Bertoft 2004; Charoenkul et al. 2006). The average
chain length (CL) of the rice amylopectin samples ranged from 16.87 to 18.27 glucose
units. (Chain length is measured in glucose residues, albeit it is most often not written out).
This was predictable since cereal starches with type A X-ray diffraction pattern are known
to have amylopectins with shorter CL in relation to amylopectins from roots and tubers,
which have type B X-ray diffraction pattern and longer chains (Hizukuri 1985). The chain
length of the two ARAs and NERICA 4 were significantly (p<0.05) longer than the chain
lengths of the two Asian rice samples.
174
The short chains (SCL) of the ARAs and NERICA 4 were also significantly (p <0.05)
longer than those found in the Asian rice samples, however, the length of long chains
(LCL) in both types of rice were not significantly different. The SCL and LCL ranged
from 14.7 to 15.6 and 48.6 to 50.5, respectively.
The φ,β-limit value of the rice
amylopectins, which generally reflects the length of the external chains, ranged from
54.7% (Koshihikari) to 60.7% for IGCR 103759.
This range is consistent with that
reported for rice amylopectin (Laohaphatanaleart et al. 2009) and is also comparable to
ranges reported for other plants (Bertoft et al. 2008). The average external chain length
ECL of the rice amylopectins ranged from 10.7-12.6 and was shorter in the Asian rice than
in NERICA 4 and the ARAs. As external chain segments participate in the crystalline
lamellae of the starch granules, the ARAs by virtue of their relatively longer external chain
length may have a thicker lamella than the rest of the samples (Koroteeva et al. 2007). The
ECLs of the samples in this study were comparable to values reported for other cereals
(A-type starches) (Zhu et al. 2011; Bertoft et al. 2008).
The average internal chain length (ICL) of the samples were not significantly different
(p <0.05), albeit the ARAs and NERICA 4 tended to have shorter ICL suggesting that they
have a more tightly branched structure in their amorphous lamellae. The total internal
chain length (TICL) of the amylopectins (defined as the entire internal length of the Bchains excluding the external segment and corresponds in φ,β-LDs to the B-chain length
minus one residue) ranged from 11.38-12.43 and was comparable to the range of 12.4-13.8
for three Asian rice varieties reported by Laohaphatanaleart et al. (2009). There were no
significant differences among the samples with respect to CLLD defined as the average
chain length of whole limit dextrin (including A-chains) and the CL of long B-chains
175
(BL-CLLD). However, the CL of short B-chains (BS-CLLD) of the ARAs were shorter than
the Asian rice samples.
The distribution profiles by weight of the B-chains and their division into subgroups are
shown in Figure 6.3 and the relative molar amounts (%) of various chain categories of the
rice amylopectins in Table 6.2. The molar amount of A-chains ranged from 49.8 to 50.6%
and hence the samples were not significantly different from each other. The samples
therefore contained almost equal-molar amounts of A- and B-chains (Table 6.2).
Koizumi et al. (1991) reported that Afp chains of amylopectins with DP 6-8 have profiles
that are characteristic to the source of plant and Bertoft et al. (2012) found that these
chains are part of clusters. Nevertheless, the Afp-chains are considered to be too short to
form double helices and for that matter do not effectively play a part in the crystalline
lamellae of the starch granules (Gidley and Bulpin, 1987). Consequently, the remainder of
the
A-chains are expected to be involved in the crystalline lamellae. Therefore, these chains
are labelled as "crystalline A-chains" (Acrystal). The Afp chains in the two ARAs and
NERICA 4 (7.16%-7.18%) were somewhat close, but statistically different (p<0.05), and
lower than that observed in the Asian rice (7.33-7.40%). However, all the samples showed
similar amounts of Acrystal as values were not significantly different from each other. The
molar amounts of Acrystal-chains reported in this study were lower than values reported by
Bertoft et al. (2008) for rice amylopectin, whereas values for Afp chains were higher.
Within the B-chain categories, the number of Bfp-chains tended to be elevated in ARAs,
which was apparently reflected in the shorter ICL of these samples. Bfp-chains have been
suggested to participate in tightly branched building blocks in clusters (Bertoft and Koch
176
2000). Long B2-chains tended to be somewhat more abundant in the African rice starches.
Otherwise, similar molar amounts of the rest of the B-chain categories (BSmajor-chains and
B3-chains) were found for all the samples and were comparable to molar amounts of the
said chains reported for rice (Bertoft et al. 2008). Molar ratios of some particular different
chain groupings of the rice amylopectins and their φ,β-limit dextrins are presented in Table
6.3. These ratios, in combination with the relative molar amounts of the different chain
categories (Table 6.2), can be used to categorize the samples on the basis of their internal
unit chain profiles into one of the four structural types of amylopectin suggested by Bertoft
et al. (2008). Going by this classification, the range of values for the molar amounts of
BS-chains, Bfp-chains, BSmajor-chains, B2-chains and B3-chains of the samples place them
under the category of type 2 amylopectins. Similarly the molar ratios of BS:BL, Acrystal:BS,
Acrystal:B, Bfp:BSmajor and S:L fit the samples into type 2. Thus, the amylopectin of African
rice belongs to the same structural type as amylopectin in Asian varieties
(Bertoft et al. 2008; Laohaphatanaleart et al. 2009).
6.4 Correlations Between Chain Length Classifications and Some Functional
Properties
Table 6.4 represents the correlation between the chain categories of the rice amylopectin
and some functional properties of the rice starch samples presented in Chapter 5. With
regards to thermal properties, gelatinization transition temperatures (To, Tp, Tc) and
enthalpy (ΔH) significantly and positively correlated with CL, SCL and ECL but
negatively with BS-CLLD. On the other hand, correlations between To, Tp, Tc and ΔH and
LCL, ICL, TICL, CLLD and BL-CLLD were not significant (p<0.05). The external chains
177
participate in the formation of double helices and ΔH mainly reflects the unwinding of
double helices during gelatinization (Cooke and Gidley 1992). Since ECL contributes to
the amount of SCL and CL, it was observed that SCL and CL also positively correlated
with ΔH. Longer external chains can also increase the crystalline stability by forming
additional hydrogen bonds between glucan strands and/or double helices (Vamadevan et al.
2013).
Since melting temperature (Tp) reflects the crystalline stability (Tester and
Morrison 1990) it was also observed that there was a positive correlation between ECL and
melting transition temperatures (To, Tp, Tc). With regards to the pasting properties, final
viscosity correlated negatively with ICL, TICL, CLLD and BS-CLLD but positively with CL,
SCL and ECL. Correlation with LCL and BL-CLLD were not significant. Similarly total
setback viscosity correlated negatively with BS-CLLD but positively with CL, SCL and
ECL. Correlations with ICL, TICL, CLLD and BL-CLLD were not significant. It has been
reported that starches with longer amylopectin chains retrograde faster than those with
short amylopectin chains (Silverio et al. 2000; Fredriksson et al. 1998; Kalichevsky et al.
1990). Generally, the external chains of amylopectin can re-associate faster than internal
chains and this recrystallization process can increase the final viscosity and setback values
hence the observed positive correlation. Figure 6.4 shows the multiple linear regression
models to describe the relationship between ECL, enthalpy, final and setback viscosities.
The equation of the fitted models are; Enthalpy = 2.70171 + 1.02258*ECL,
Final Viscosity = 38.9856 + 21.3679*ECL and Total setback = -281.426 + 35.1199*ECL
178
6.5 Conclusions
The structure of the amylopectins studied falls into the type 2 classification of
amylopectins proposed by Bertoft et al. (2008). The profiles of the unit chain distribution
of the debranched amylopectins of the ARAs and NERICA 4 were distinct from the Asian
rice. The ECL of the ARAs and NERICA 4 were longer than the Asian rice varieties.
Since six glucose units are needed per turn of double helix the data suggest that the ARAs
and NERICA 4 can possibly have two complete turns, while the Asian rice varieties can
have less than two. This may explain the higher gelatinization enthalpies observed in the
ARAs. Additionally the observations suggest that more consideration should be given to
the ECL chains categories of amylopectin to throw more light on the functional properties
of starches.
The observed correlations strongly suggest that the short chains of the
amylopectins, especially their external segments were important in influencing the thermal
and pasting properties of the starch samples. The differences observed in the amylopectin
structures of African and Asian rice in this study suggest possible differences in the fine
structure of the clusters and building blocks of their amylopectins.
6.6 Acknowledgements
This research work was undertaken with the kind assistance of the Global Rice Science
Partnership (GRiSP) and the Africa Rice Center (AfricaRice) in terms of financial support
and supply of samples.
179
6.7 Tables and Figures
#
Table 6.1. Chain Length Classifications and φ, β-Limit Values of Rice Amylopectin as obtained by HPAEC Analyses
φ, β-Limit
Sample
CL
SCL
LCL
ECL
ICL TICL
value
CLLD BS-CLLD BL-CLLD
African rice
IRGC 103759
18.27c
15.64b
50.36a
12.58b
4.69a
11.38a
60.65c
7.19a
8.41a
39.21a
TOG 12440
18.11c
15.45b
49.14a
12.14b
4.97a
11.90a
58.74bc
7.47a
8.76ab
39.07a
58.51abc
7.44a
8.69ab
38.75a
O. sativa x O. glaberrima cross
17.23c
15.47b
50.50a
12.00b
4.94a
11.90a
KOSHIHIKARI
16.87a
14.76a
50.26a
10.73a
5.15a
12.43a 54.68a
7.65a
9.26b
39.26a
WITA 4
17.23b
14.71a
48.64a
11.08a
5.15a
12.29a
7.65a
9.03ab
39.53a
NERICA 4
Asian rice
#
55.61ab
Within each column values with the same letters are not significantly (p < 0.05) different
CL = Average Chain Length of Amylopectin
SCL = CL of Short Chains
LCL = CL of Long Chains
ECL (External Chain Length) = CL x (φ, β-Limit Value/100) + 1.5
ICL (Internal Chain Length) = CL-ECL – 1
TICL (Total Internal Cain Length) = B-CLLD-1
φ, β-Limit Value = Computed from the Difference between CL of Amylopectin and its φ,β-LD
CLLD = Average Chain Length of φ, β-Limit Dextrin
BS-CLLD = CL of Short B-chains
BL-CLLD = CL of Long B-chains
180
Table 6.2. Relative Molar Amounts (%) of Chain Groupings in Amylopectins Isolated from Rice Starches
#
A-chains1
Afp2
Acrystal3
B-chains4
BS
BL
Bfp5
BSmajor6
B27
B38
IRGC 103759
49.99a
7.16a
42.84a
50.01a
42.99a
7.03a
22.69b
19.46a
6.27ab
0.65a
TOG 12440
49.80a
7.17ab
42.63a
50.21a
41.80a
8.41b
22.14ab
21.22abc
6.18ab
0.68a
50.04a
7.18b
42.86a
49.97a
41.42a
8.55b
22.45b
20.11ab
6.91b
0.67a
KOSHIHIKARI
50.61a
7.33c
43.28a
49.34a
41.71a
7.69ab
19.73a
23.41c
5.59a
0.67a
WITA 4
49.96a
7.40d
42.56a
50.05a
41.53a
8.52b
20.69ab
22.61bc
6.01a
0.71a
Sample
African rice
O. sativa x O. glaberrima cross
NERICA 4
Asian rice
#
Within each column values with the same letters are not significantly different (p < 0.05)
Detected as Maltose after Debranching of φ, β-LD
2
“Fingerprint” A-chains at DP 6-8 in the Original Amylopectin Sample
3
Acrystal -Chains Calculated as all A-chains Less Afp
4
B-Chains Correspond to DP >3 in φ, β-LDs, and were Divided into Short (BS) and Long (BL) Chains
Chains (BL) at between DP 22–25 and above Respectively Depending on the Sample
5
“Fingerprint” B-Chains at DP 3–7 in φ, β-LDs
6
The Major Group of Short B-Chains at DP 8 to 22 (this may stretch to DP 24 depending on the sample in question)
7
Long Chains between DP 22 or 25 (depending on the sample) and 55
8
Long Chains at DP >56
1
181
Table 6.3. Molar Ratios of Different Chain Groupings of Rice Amylopectins and Their φ,β-Limit Dextrins#
Sample
A:B
S:L
BS:BL
Acrystal:BS
Acrystal:B
Bfp:BSmajor
IRGC 103759
1.0a
11.6a
6.1b
1.0a
0.9a
1.1c
TOG 12440
1.0a
11.7a
5.0a
1.0a
0.9a
1.0abc
1.0a
11.8a
4.9a
1.0a
0.9a
1.1bc
KOSHIHIKARI
1.0a
14.2b
5.4ab
1.0a
0.9a
0.8a
WITA 4
1.0a
12.4a
4.9a
1.0a
0.9a
0.9ab
African rice
O. sativa x O. glaberrima cross
NERICA 4
Asian rice
#
Values with the same letters are not significantly different (p < 0.05) Within Each Column
S (Short Amylopectin Chains)
L (Long Amylopectin Chains)
The Other Abbreviations Have Been Explained in Table 6.2
182
Table 6.4. Correlation between Chain Groupings of Rice Amylopectins and Some Function Properties of Starch#
CL
SCL
LCL
ECL
ICL
TICL
CLLD
BS-CLLD BL-CLLD
To
0.8998
0.8967
NS
0.8548
NS
NS
NS
-0.7038
NS
Tp
0.9002
0.9033
NS
0.8653
NS
NS
NS
-0.7148
NS
Tc
0.8767
0.9057
NS
0.8441
NS
NS
NS
-0.6826
NS
ΔH
0.7714
0.8228
NS
0.7791
NS
NS
NS
-0.6575
NS
Final viscosity
0.8251
0.7958
NS
0.8578 -0.6889 -0.6799 -0.6889
-0.8307
NS
Total setback
0.9386
0.7997
NS
0.9060
-0.8380
NS
#
NS
NS
NS
Statistical significance of the estimated correlations was conducted at p<0.05
NS=Non Significant
183
10
9
Purified amylopectin
Carbohydrate (wt. %)
8
Whole starch
7
6
5
4
3
2
1
0
40
50
60
70
80
90
100
110
120
130
140
150
160
170
Fraction number
Figure 6.1. Gel-Permeation Chromatograms (Sepharose CL-6B) of Debranched
Whole Starch and Purified Amylopectin from Rice Starch (Koshihikari).
184
6
5
5
18
3
44
2
6
35
0
20
40
60
Degree of polymerization
6
80
100
0
20
40
60
Degree of polymerization
12
14
80
100
0
20
40
60
80
Degree of polymerization
KOSHIHIKARI
5
5
19
4
19
4
43
3
43
Weight (%)
Weight (%)
6
0
WITA 4
14
34
6
6
12
42
3
2
0
0
4
1
1
NERICA 4
18
43
2
13
5
18
3
35
6
IRGC 103759
4
Weight (%)
Weight (%)
4
1
12
TOG 12440
Weight (%)
12
6
3
35
35
2
2
1
1
6
6
0
0
0
20
40
60
80
Degree of polymerization
100
0
20
40
60
80
Degree of polymerization
100
Figure 6.2. Unit Chain Outline of Debranched Rice Amylopectins Obtained by High Performance Anion-Exchange Chromatography (HPAEC).
(Arrows and Digits Signify Degree of Polymerization)
African Rice= IRGC103759; TOG 12440
Asian Rice= Koshihikari; WITA 4
O.sativa x O.glaberrima cross=NERICA 4
185
100
BS
5
3
31
B2
22
2
Weight (%)
Weight (%)
BL
22
1
0
0
20
40
60
80
Degree of polymerization
BS
WITA 4
BS
9
Weight (%)
32
22
1
22
2
100
0
20
40
60
KOSHIHIKARI
BL
3
22
29
2
1
0
0
20
40
60
Degree of polymerization
80
100
0
0
20
40
60
80
Degree of polymerization
100
Figure 6.3. Unit Chain Outline of Debranched φ,β-Limit Dextrins of Rice Amylopectins Obtained by High-Performance Anion-Exchange Chromatography
(HPAEC).
(Arrows and Digits Signify Degree of Polymerization)
African Rice = IRGC103759; TOG 12440 Asian Rice = Koshihikari; WITA 4
O.sativa x O.glaberrima cross =NERICA 4
186
80
Degree of polymerization
5 9
4
3
2
20
40
60
80
Degree of polymerization
Bfp
5
31
0
0
100
3
1
BL
Bfp
4
31
2
B3
8
4
3
1
0
5
8
4
BL
Bfp
BL
5
8
4
weight (%)
Bfp
BSmajor
NERICA 4
IRGC 103759
Weight (%)
Bfp
BS
BS
TOG 12440
100
Plot of Fitted Model
Plot of Fitted Model
16
320
310
Final Viscosity
Enthalpy
15
14
13
300
290
280
270
12
260
10
10.5
11
11.5
ECL
12
12.5
13
Enthalpy = 2.70171 + 1.02258*ECL
r=0.7791
10
10.5
11
11.5
ECL
12
12.5
13
Final Viscosity = 38.9856 + 21.3679*ECL
r=0.8578
Plot of Fitted Model
180
Total Setback
160
140
120
100
80
10
10.5
11
11.5
ECL
12
12.5
13
Total Setback = -281.426 + 35.1199*ECL
r=0.9060
Figure 6.4. Multiple Linear Regression Models to Describe the Relationship between ECL, Enthalpy, Final and Setback Viscosities.
187
6.8 References
Bao, J. S. 2012. Toward understanding the genetic and molecular bases of the eating and
cooking qualities of rice. Cereal Foods World 57: 148-156.
Bertoft, E., and Spoof, L. 1989. Fractional precipitation of amylopectin alpha-dextrins using
methanol. Carbohydr. Res. 189: 169-180.
Bertoft, E., and Koch, K. 2000. Composition of chains in waxy-rice starch and its
structural units. Carbohydr. Polym. 41: 121-132.
Bertoft, E. 2004. On the nature of categories of chains in amylopectin and their connection
to the super helix model. Carbohydr. Polym. 57: 211-224.
Bertoft, E., Piyachomkwan, K., Chatakanonda, P., and Sriroth, K. 2008. Internal unit chain
composition in amylopectins. Carbohydr. Polym. 74: 527-543.
Bertoft, E., Koch, K., and Åman, P. 2012. Building block organisation of clusters in
amylopectin from different structural types. Int. J. Biol. Macromol. 50: 1212-1223.
Charoenkul, N., Uttapap, D., Pathipanawat, W., and Takeda, Y. 2006. Simultaneous
determination of amylose content and unit chain distribution of amylopectins of
cassava starches by fluorescent labeling/HPSEC. Carbohydr. Polym. 65: 102-108.
Cooke, D., and Gidley, M. J. 1992. Loss of crystalline and molecular order during starch
gelatinisation: origin of the enthalpic transition. Carbohydr. Res. 227: 103-112.
188
DuBois, M., Gilles, K. A., Hamilton, J. K., Rebers, P. A., and Smith, F. 1956.
Colorimetric method for determination of sugars and related substances. Anal.
Chem.28: 350-356.
Fennema, O. R. 1985. Water and ice. In: Fennema OR, editor. Food Chem. New York, N.Y.:
Marcel Dekker.: 23-67.
Fredriksson, H., Silverio, J., Andersson, R., Eliasson, A.-C., and Åman, P. 1998. The
influence of amylose and amylopectin characteristics on gelatinization and
retrogradation properties of different starches. Carbohydr. Polym. 35: 119-134.
French, D. 1972 : Fine structure of starch and its relationship to the organization of starch
granules. J. Jpn. Soc. Starch Sci. 19: 8-25.
Gidley, M. J., and Bulpin, P. V. 1987. Crystallisation of malto-oligosaccharides as
models of the crystalline forms of starch: minimum chain-length requirement for the
formation of double helices. Carbohydr. Res. 161: 291-300.
Hanashiro, I., Abe, J.I., and Hizukuri, S. 1996. A periodic distribution of chain length of
amylopectin as revealed by high-performance anion-exchange chromatography.
Carbohydr. Res. 283: 151-159.
Hizukuri, S. 1985. Relationship between the distribution of the chain length of amylopectin
and the crystalline structure of starch granules. Carbohydr. Res. 141: 295-306.
Hizukuri, S. 1986. Polymodal distribution of the chain lengths of amylopectins, and its
significance. Carbohydr. Res. 147: 342-347.
189
Juliano, B.O. 1998. Varietal impact on rice quality. Cereal Foods World 43: 207.
Kalichevsky, M. T., Orford, P. D., and Ring, S. G. 1990. The retrogradation and gelation
of amylopectins from various botanical sources. Carbohydr. Res. 198: 49-55.
Kalinga, D. N., Waduge, R., Liu, Q., Yada, R. Y., Bertoft, E., and Seetharaman, K. 2013.
On the differences in the granular architecture and starch structure between pericarp
and endosperm wheat starches. Starch/Stärke. 65: 791-800.
Klucinec, J.D., and Thompson D.B. 2002. Structure of amylopectins from ae-containing maize
starches. Cereal Chem. 79: 19-23.
Koch, K., Andersson, R., and Åman, P. 1998. Quantitative analysis of amylopectin unit
chains by means of high-performance anion-exchange chromatography with pulsed
amperometric detection. J. Chrom. A. 800:199-206.
Koizumi, K., Fukuda, M., and Hizukuri, S. 1991. Estimation of the distributions of chain
length of amylopectins by high-performance liquid chromatography with pulsed
amperometric detection. J. Chrom. A. 585:233-238.
Koroteeva, D. A., Kiseleva, V. I., Sriroth, K., Piyachomkwan, K., Bertoft, E., and Yuryev,
P.V. 2007. Structural and thermodynamic properties of rice starches with different
genetic background part 1. Differentiation of amylopectin and amylose defects.
Int. J. Biol. Macromol. 41: 391-403.
190
Laohaphatanaleart, K., Piyachomkwan, K., Sriroth, K., Santisopasri, V., and Bertoft, E. 2009.
A Study of the internal structure in cassava and rice amylopectin.
Starch/Stärke. 61:557-569.
Laohaphatanaleart, K., Piyachomkwan, K., Sriroth, K., and Bertoft, E. 2010. The fine
structure of cassava starch amylopectin: part 1: organization of clusters. Int. J. Biol.
Macromol. 47: 317-324.
Manners, D.J. 1989. Recent developments in our understanding of amylopectin structure.
Carbohydr. Polym. 11: 87-112.
Nakamura, Y. 2002. Towards a better understanding of the metabolic system for amylopectin
biosynthesis in plants: rice endosperm as a model tissue. Plant Cell Physiol. 43: 718725.
Nikuni, Z. 1978. Studies on starch granules. Starch/Stärke. 30: 105-111.
Peat, S., Whelan, W., and Thomas, G. J. 1952. Evidence of multiple branching in waxy maize
starch. J. Chem. Soc. 4546-4548.
Silverio, J., Fredriksson, H., Andersson, R., Eliasson, A.-C., and Åman, P. 2000. The effect
of temperature cycling on the amylopectin retrogradation of starches with different
amylopectin unit-chain length distribution. Carbohydr. Polym. 42: 175-184.
Takeda, Y., Hizukuri, S., and Juliano, B. O. 1987. Structures of rice amylopectins with low
and high affinities for iodine. Carbohydr. Res. 168: 79-88.
191
Takeda, Y., Hizukuri, S., and Juliano, B.O. 1986. Purification and structure of amylose from
rice starch. Carbohyd. Res. 148: 299-308.
Takeda, Y., Tomooka, S., and Hizukuri, S. 1993. Structures of branched and linear
molecules of rice amylose. Carbohyd. Res. 246: 267-72.
Tester, R. F., and Morrison, W. R., 1990. Swelling and gelatinization of cereal starches I.
effects of amylopectin, amylose and lipids. Cereal Chem. 67: 551-557.
Vamadevan, V., Bertoft, E., and Seetharaman, K. 2013. On the importance of organization
of glucan chains on thermal properties of starch. Carbohydr. Polym. 92: 1653-1659.
Waduge, R., Xu, S., and Seetharaman, K. 2010. Iodine absorption properties and its effect
on the crystallinity of developing wheat starch granules. Carbohydr. Polym. 82: 786794.
Wang, L. Z., and White, P. J. 1994. Structure and properties of amylose, amylopectin, and
inter-mediate materials of oat starches. Cereal Chem. 71: 254-262
Wang,Y-J., White, P., Pollak, L., and Jane, J. 1993. Amylopectin and intermediate
materials in starches from mutant genotypes of the Oh43 inbred line. Cereal Chem.
70: 521-525.
Whistler, R. L, and BeMiller, J. N. 1997. Carbohydrate chemistry for food scientists. St. Paul,
Minnesota: Eagan Press.
Zhu, F., Corke, H., and Bertoft, E. 2011. Amylopectin internal molecular structure in
relation to physical properties of sweetpotato starch. Carbohydr. Polym. 84: 907-918.
192
CHAPTER 7: STRUCTURE OF CLUSTERS AND BUILDING BLOCKS IN
AMYLOPECTIN FROM AFRICAN RICE ACCESSIONS
193
Abstract
This study aimed at determining the fine structure of amylopectin from Africa rice accessions
(Oryza glaberrima) and Asian rice (Oryza sativa). The clusters in the amylopectins, which
were produced by hydrolysis using α-amylase from Bacillus amyloliquefaciens and isolated by
methanol precipitation, were treated with phosphorylase a and then β-amylase to remove all
external chains and obtain φ,β-limit dextrins. The clusters were further hydrolyzed extensively
with α-amylase to characterize their building block composition. The clusters, building blocks,
and their unit chain length profiles were analyzed by gel-permeation chromatography and/or
high-performance anion-exchange chromatography with pulsed amperometric detector. While
NERICA 4, which represents a cross between African and Asian rices, possessed the largest
clusters (DP 89.6), the average DPs of the clusters from the African rice samples were slightly
larger (DP ~83) than in the Asian variety (DP ~81). On the basis of number of chains (NC), the
African rice samples represented both the smallest (11.8; IRGC 103759) and the largest clusters
(14.1; TOG 12440). The short CL of TOG 12440 accounted for the large NC of its clusters and
similarly, the clusters of NERICA 4, which had high DP, had intermediate NC due to long CL.
Overall, the result showed that the structure of the clusters in TOG 12440 was dense with short
chains and high degree of branching, whereas in NERICA 4 the situation was the opposite.
IRGC 103759 and the Asian rice (WITA 4) possessed clusters with intermediate characteristics.
The major type of building blocks in all samples was group 2 (representing between 40.349.4%) followed by group 3 (27.0-29.3%). Similar amounts of groups 4 and 5 (9.3-14.4%)
were present in the clusters, whereas group 6 represented only between 1.0 and 4.4% of the
building blocks. The average number of building blocks (NBbl) in the clusters was significantly
194
larger in NERICA 4 (5.8) and WITA 4 (5.7) than in the two African rice accessions (5.1 and
5.3). Key words: African rice, starch, amylopectin, clusters, building blocks
195
7.1 Introduction
Oryza glaberrima (African rice), is one of the two main domesticated rice that is cultivated in
the West Africa sub region mainly due to its ability to withstand the harsh environmental
conditions and Oryza sativa (Asian rice) is the other distinct type of domesticated rice having a
global distribution particularly in Asia. The predominant type of rice consumed worldwide is
the polished or white rice (Nanri et al. 2010) of which the major component is starch (90% dry
basis).
Starch contains two major types of biomacromolecules namely: amylose and
amylopectin, of which amylopectin is considered as the major component in storage starch and
accounts for up to 85% by weight (Nakamura 2002). Amylopectin is reported to be responsible
for the semi-crystalline growth ring structure in the starch granules, which consist of stacks of
alternating amorphous and crystalline lamellae (Jenkins et al. 1993). In the crystalline lamellae,
short A-chains (un-substituted chains) and external sections of B-chains (substituted chains)
form double helices that crystallize into the so-called A-allomorph crystal type (Bertoft et al
2008; Imberty et al. 1991).
When amylopectin is treated with exo-acting enzymes, such as phosphorylase or β-amylase, the
internal part of amylopectin can be isolated by the removal of its external chains (Manners
1989).
When phosphorylase and β-amylase are used successively, φ,β-limit dextrins are
produced and debranching of these dextrins leads to the exposure of the internal chains, which
can be analysed by anion-exchange chromatography (Bertoft 2004). The internal unit chain
distributions in amylopectins from different botanical sources have been analyzed by Bertoft et
al. (2008). They reported that the internal chains were all B-chains and could be divided into
short (BS) and long chains (BL). BS-chains were divided into very short “fingerprint” Bfpchains with a degree of polymerisation (DP) 3-7, and BSmajor, with DP ∼8-25 (the exact range
196
depends on the botanical origin of starch).
The BL-chains were divided into B2-chains
(DP∼26-50) and B3-chains (DP > 50). Based on the internal unit chain profile, Bertoft et al.
(2008) categorized amylopectins from different plants into four structural types. Cereals belong
to either type 1 or type 2 amylopectins, of which rice is type 2. This type of structure typically
contains considerably more BL-chains and a narrower size distribution of BS-chains compared
to type 1.
Branches of amylopectin combine to form clusters, which are defined as groups of branches
located within a length of nine glucose residues along the backbone (Bertoft 2007). Chains of
clusters that constitute the crystalline lamellae are external chains, while the chain segments
forming the amorphous lamellae are internal chains and defined as segments between branches
(Pérez and Bertoft 2010). In the backbone model of amylopectin, the branches of the clusters
are outspread along a backbone, which is mostly composed of longer amylopectin chains. In
the amorphous lamellae, the branches form tightly branched regions inside the clusters known
as building blocks, which have internal chain lengths of ~1-3 glucose residues. Building blocks
occur in a range of sizes; the smallest consists of two chains and have a degree of
polymerization (DP) of 5 to 9, whereas the largest blocks have more than 10 chains (DP > 45).
The average numbers of building blocks (NBbl) in clusters have been reported to vary from
about 3 in the small clusters of potato to 9 in clusters of cassava (Bertoft 2007; Bertoft et al.
2010). Within the cluster, the building blocks are interconnected by inter-block chain lengths
(IB-CL) of ~5 to 8 glucose residues (Bertoft et al. 2012b).
α-Amylase of Bacillus amyloliquefaciens has been used in the past to isolate and characterise
clusters and building blocks from a range of different plant sources (Kong et al. 2009; Zhu et al.
2011; Bertoft et al. 2012a).
Clusters are formed through a carefully controlled limited
197
hydrolysis (Bertoft 1999: Zhu et al. 2011), whereas building blocks are isolated by an extensive
hydrolysis of clusters into dextrins which are in practice considered to be α-LDs (Kong et al.
2009: Bertoft 2007a). The objective of this study is to investigate the structure of clusters and
building blocks in amylopectin from two African rice accessions. They are compared with one
Asian rice variety and a cross between African and Asian rice.
7.2 Materials and Methods
7.2.1
Materials
The Africa Rice Center based in Cotonou, Benin, provided four dehusked and polished rice
samples for this study. Two of the samples were African rice accessions (ARAs) – IRGC
103759 and TOG 12440. The others were one Asian rice variety – cv. WITA 4 – and an
interspecific cross between African rice x Asian rice – NERICA 4. All chemicals and solvents
used were of ACS certified grade.
7.2.2
Enzymes
The enzymes used for the analysis were β-amylase from barley (EC 3.2.1.2, specific activity
705 U/mg), isoamylase from Pseudomonas amyloderamosa (EC 3.2.1.68, specific activity 210
U/mg), pullulanase from Klebsiella pneumoniae (EC 3.2.1.41, specific activity 699 U/mg), and
α-amylase from B. amyloliquefaciens (EC 3.2.1.1), supplied by Megazyme International,
Wicklow, Ireland. The method described by Bertoft et al. (1993) was used to determine the
activity of the α-amylase at pH 6.5, 25°C. Phosphorylase a from rabbit muscle (EC 2.4.1.1, ~25
U/mg protein or 9 U/mg solid) was from Sigma–Aldrich, Deisenhofen, Germany.
198
7.2.3
Methods
7.2.4
Starch Isolation
Starch was isolated from the rice samples as described in chapter 4.
7.2.5
Isolation of Amylopectin from Rice Starch and Determination of Purity
The protocols outlined in chapter 6 were used to isolate amylopectin from the rice starch
samples and test for purity.
7.2.6
Cluster Preparation - Time Course for α-Amylolysis of Amylopectin
Time course analysis of α-amylolysis for the production of clusters from amylopectin was
executed as per the method described by Zhu et al. (2013) and Kong et al. (2009). Amylopectin
(20 mg) was dissolved in 400 μL of 90 % DMSO in a boiling water bath and thereafter stirred
constantly overnight at room temperature. The dissolved sample was then diluted with hot
double-distilled water (1.4 mL) and cooled to 25 °C in a water bath. 200 μL of α-amylase (0.9
U/mL) in sodium acetate (NaOAc) buffer (0.01 M; pH 6.5) equilibrated to 25 °C was added to
hydrolyse the sample for more than 2 h with constant magnetic stirring. The enzyme and
substrate concentrations were 0.09 U/mL and 10 mg/mL, respectively, during the incubation.
Aliquots (200 μL) were drawn from the reaction mixture at 20 min interval up to 160 min into
Eppendorf tubes containing 40 μL of NaOH (5 M) to halt the reaction by destroying the αamylase. The samples were stored at –18 °C if not analyzed immediately. To determine the
molecular size distribution, the samples were diluted with 200 μL of water and analyzed by gel-
199
permeation chromatography (GPC) on the column of Sepharose CL-6B as described previously
(chapter 6).
7.2.7
Production of Clusters and Their Corresponding φ,β-Limit Dextrins
Clusters from amylopectin and their corresponding φ,β-limit dextrins were produced by the
method reported by Kong et al. (2009). Amylopectin (100 mg) was dissolved in 1.6 mL of 90%
DMSO in centrifuge tubes by gently heating and then constantly stirring overnight at room
temperature. Hot double-distilled water (5.6 mL) was added and the sample was cooled to 25
°C in a water bath. 800 µL of α-amylase (0.9 U/mL) in NaOAc buffer (0.01 M, pH 6.5) was
added to start the reaction in the water bath with constant magnetic stirring. The reaction was
continued for 90 min (based on the time course experiment) before halting it by adding 160 µL
NaOH (5 M). The mixture was allowed to stand at room temperature for 1 h to ensure the
destruction of the α-amylase. Methanol (40 mL) was added to precipitate the clusters in the
form of α-dextrins. The sample was allowed to precipitate for 3 h at 4 °C before recovering the
clusters by centrifugation (4000 x g, 20 min) at 4 °C. The precipitate was washed twice with
pure methanol (10 mL) and allowed to dry overnight in a fume hood.
For the production of φ,β limit dextrins, samples of clusters were dissolved in hot MilliQ water
(30 mL) and the carbohydrate content was adjusted to ~3 mg/mL. For 1 volume of the solution,
0.1 volume of 1.1 M sodium phosphate buffer (pH 6.8), 0.05 volumes of 2.8 mM EDTA, and
0.25 volumes of freshly prepared phosphorylase a solution (25 U/mg 2.5 mg dissolved in 25
mL MilliQ water) were added. The solution was stirred overnight at room temperature, after
which the reaction was terminated in a boiling water bath for 5 min.
The solution was
concentrated with a rotary evaporator to 8 mL. The carbohydrate on the walls of the flask was
200
recovered by washing with 2 mL of water. Removal of glucose 1-phosphate was performed on
two PD-10 columns (Sephadex G-25, Pharmacia, Uppsala, Sweden) coupled in series: 2 mL
sample was applied on the columns followed by 3 mL MilliQ water. The eluate was discarded
and then 5 mL MilliQ water was added and the sample was collected. The carbohydrate content
was adjusted to ~3 mg/mL and the phosphorolysis and purification procedures were repeated to
produce φ-limit dextrins of clusters (φ-LD).
The carbohydrate concentration of purified φ-LD was adjusted to ~3 mg/mL. For 1 volume of
the purified φ-LD solution, 1/3 volume of 0.01 M NaOAc buffer (pH 6.0) and β-amylase (~4
U/mg substrate) were added and the solution was stirred overnight at room temperature. After
termination of the reaction in a boiling water bath for 5 min, the solution was concentrated by
rotary evaporation and purified on PD-10 columns as described above. The β-amylolysis step
was repeated and the φ,β-limit dextrins of clusters (φ,β-LD) produced were finally purified two
times on the PD-10 columns to completely remove maltose.
The purified φ,β-LD was
lyophilized or directly analyzed on the Sepharose CL-6B column for size distribution of the
clusters as described above.
7.2.8
Chain Length Distribution of Clusters
φ,β-LD (1 mg) was weighed and dissolved in warm water (900 μL). 0.01 M NaOAc buffer
(100 μL; pH 5.5) was added and debranching of φ,β-LD was done overnight at room
temperature with 2 μL pullulanase (36.3 U/mL) and 2 μL isoamylase (1000 U/mL). The unit
chain distribution was analyzed with high-performance anion-exchange chromatography with
pulsed amperometric detection (HPAEC-PAD) (Dionex ICS 3000, Synnyvale, CA, USA) on a
201
CarboPac PA-100 column (4 x 250 mm) (Bertoft et al. 2012a). PAD signals were converted to
carbohydrate content (Koch et al. 1998).
7.2.9
Preparation of Building Blocks and Determination of Structure
The production of building blocks from the φ,β-LDs of clusters was done as reported by Bertoft
et al. (2012a). φ,β-LD (5 mg) was dissolved in warm water (540 μL) and thereafter hydrolysed
with α-amylase (60 μL, 60 U/mL in 0.01M NaOAc buffer, pH 6.5) for 3 h at 35 °C. The
reaction was terminated by boiling in a water bath for 5 min. Then NaOAc buffer (60 µL; 0.01
M, pH 6.0) and β-amylase (3 µL) were added to remove any remaining external chain
segments. After 18 h of β-amylolysis the reaction was stopped by boiling in a water bath for 5
min. The size distribution of the building blocks was analyzed on a column (1.6 x 90 cm) of
Superdex 30 (Pharmacia, Uppsala, Sweden) with NaCl (0.05 M) as an eluent set at a flow rate
of 1 mL/min. The column was calibrated using α-dextrins as described by Bertoft and Spoof
(1989). The carbohydrate content of the fractions (1 mL) collected was estimated by using the
phenol-sulphuric acid method (Dubois et al. 1956).
7.2.10 Statistical Analysis
The statistical programme Statgraphics Centurion XV, Version 15.1.02 (StatPoint, Warrenton,
VA, USA) was employed to conduct statistical analysis. One-way analysis of variance was
performed to determine significant differences between any pair of means at the 95.0%
confidence level and multiple range tests were done to determine which means were different.
202
7.3 Results and Discussion
7.3.1
Time Course Analysis of α-Amylolysis of Amylopectin from Rice Starch
Figure 7.1 shows the degradation by time of amylopectin from rice starch hydrolyzed with αamylase. The amylopectin isolated from WITA 4 starch sample was used as a model for the
assessment of a suitable time of α-amylolysis for the production of clusters. The kinetics of the
hydrolysis as shown by the change in molecular size distribution with time was consistent with
earlier reports on amylopectin from sweet potato (Zhu et al. 2011), grain amaranth (Kong et al.
2009) and maize (Zhu et al. 2013). The size distribution of the dextrins produced was bimodal
and fraction A (Figure 7.1) characteristically represented single clusters or groups of clusters
(domains), while fraction B represented small fragments mainly derived from external chains of
amylopectin (Bertoft 1989).
The hydrolysis process was initially very rapid resulting in
increased number of dextrins, but the rate slowed down after approximately 90 min.
α-
Amylase of B. amyloliquefaciens has been reported to have nine subsites, which bind the Dglucosyl units of its substrate. α-Amylase is known to react at a very fast rate when all of its
nine subsites are filled. However, when all the subsites are not filled, the reaction slows down
(Robyt and French, 1963). α-Amylase attacks the external chains of amylopectin leading to the
formation of maltohexaose. Simultaneously, however, it also initially attacks the long internal
chains between the clusters of amylopectin, which results in the release of clusters (Bertoft
1989). In this study, the activity of α-amylase used was 0.09 U/mL and when the relative molar
amount of branched dextrins (fraction A) was plotted against time (Figure 7.2), it was obvious
that the branched dextrins became less susceptible to further α-amylolysis after approximately
203
90 min (i.e., at this time a high proportion of clusters existed in the reaction mixture). Hence 90
min of α-amylolysis was chosen for cluster production from all four amylopectin samples.
7.3.2
Characterization of φ,β-Limit Dextrins Produced from Clusters
Following the choice of 90 min as the time for hydrolysis, the four amylopectin samples were
subjected to α-amylolysis in order to release clusters. The clusters produced in the form of
branched α-dextrins (represented by fraction A in Figure 7.1) were precipitated with methanol.
These sets of clusters were further hydrolyzed with phosphorylase a and β-amylase to cleave
external chains that remained after α-amylolysis. The molecular size distribution patterns of the
resulting φ,β-limit dextrins of clusters (φ,β-LDs) were rather similar and are shown in Figure
7.3 and Table 7.1. However, the average DP of the clusters from the African rice was slightly
larger (DP ~83) than in the Asian variety (WITA 4; DP 81). The largest clusters were,
however, possessed by the variety representing a cross between African and Asian rices
(NERICA 4; DP 89.6). In a previous report, disparities were observed in the sizes of the
clusters isolated from different botanical sources.
Generally, starches that exhibit B-type
crystallinity (e.g., potato and yam starch) have relatively small clusters compared to those with
A-type crystallinity such as cereals (Bertoft et al. 2012a). The sizes of the clusters in this study
were larger than those previously reported for clusters isolated from several plant sources
(Bertoft et al. 2012a). Only the clusters from Andean yam bean starch, with an average size of
DP 82.7, were comparable to the sizes reported for rice in this study.
The φ,β-LDs of clusters were debranched by isoamylase and pullulanase before further analysis
with HPAEC to obtain the unit chain distribution (Figure 7.4), from which the average chain
length (CL) was estimated. CL ranged from 5.9 to 7.3 (Table 7.1). A number of structural
204
parameters for the characterization of cluster structures based on DP and CL were also
calculated and are presented in Table 7.1. Generally, these data compared well with a previous
report on Asian rice and fit with the structural features of clusters from group 2 amylopectins
(Bertoft 2012). The average number of chains per cluster (NC), which is another parameter that
can be used to express the size of the clusters (Bertoft et al. 2012a), ranged from 11.8 to 14.1.
On this basis the African rice samples represented both the smallest (IRGC 103759) and the
largest clusters (TOG 12440). The large NC of the clusters from TOG 12440 was due to their
short CL. Similarly, the clusters of NERICA 4, which had high DP, had intermediate NC due
to long CL. The internal chain length (ICL) ranged from 3.8 to 5.3 and was reflected in the
degree of branching (DB), which ranged from 12.7 to 15.8% and corresponded to NERICA 4
and TOG 12440, respectively. The total internal chain length (TICL) of the clusters in TOG
12440 was also short (7.9), whereas NERICA 4 possessed the longest TICL (8.8). Overall, the
result showed that the structure of the clusters in TOG 12440 was dense with short chains and
high DB, whereas in NERICA 4 the situation was the opposite. IRGC 103759 and WITA 4
possessed clusters with intermediate characteristics (Table 7.1).
7.3.3
Unit Chains in Clusters
In the structure of amylopectin, A-chains refer to chains that do not carry any other chains by α(1,6)-linkages while B-chains carry other chains (Peat et al. 1952). To differentiate the new
chains produced in a cluster from those that originally existed in amylopectin, the comparable
chains in clusters are designed by lowercase letters (Bertoft et al. 2012a). The b-chains in
isolated clusters have two origins; some are produced by α-amylase action, while some arise
from pre-existing B-chains in the parent amylopectin (Bertoft, et al. 2012a). The b-chains were
205
further sub-divided into different categories (Figure 7.5) to describe their possible role in the
structure of the clusters (Bertoft and Koch 2000; Bertoft et al. 2012a). Chains of the categories
b1a, b1b, b2, and b3 are thought to play a role in interconnecting the building blocks in a
cluster. The major types of inter-block chains are b1- and b2-chains, which for the most part
correspond to the BSmajor -chains in the original amylopectin. The b1-chains contain one interblock segment and are divided into b1a and b1b subgroups: b1a-chains have a DP from 7 to 10
and b1b-chains a DP from 11 to 18. The b2-chains (DP 19-27) and b3-chains (DP > 28) are
involved in the interconnection of three and four building blocks and contain two and three
inter-block segments, respectively (Bertoft et al. 2012a). b0-Chains, which are the smallest
type of b-chains with DP 4-6, are fully embedded in the building blocks and have no inter-block
segments.
Substantial amounts of b0-chains are produced during α-amylolysis and are
therefore remnants of inter-cluster segments of the parent amylopectin molecule (Bertoft et al.
2012a). a-Chains are also to a substantial degree, a result of the action of the α-amylase. These
chains are represented by maltose in the debranched ϕ,β-LD chromatogram (Fig. 7.4). In
addition, clusters possess typically a lot of chains with DP 3 (Bertoft et al. 2008; Bertoft et al.
2012b), and this was also the case with the rice samples in this study. These short chains
represent a mixture of a-chains and the shortest possible b-chains (Bertoft et al. 2012a).
Table 7.2 shows the molar distribution of the chain categories in the clusters from the four rice
samples. Significant differences were found in the amount of a-chains (DP 2) in the samples.
While sample IRGC 103759 had most a-chains (49.5%), TOG 12440 had the least of them
(43.5%).
Since chains indicated by maltotriose are a mixture of a- and b-chains, it is a
challenge to precisely quantify the proportion of each category. The values are, however,
interesting to compare with the corresponding A-chains in the parent amylopectin molecules,
206
which all possessed close to 50% A-chains (i.e., the ratio of A:B chains was 1.0) (Table 6.2). In
the clusters, IRGC 103759 and NERICA 4 had similar levels of a-chains with DP 2, but bchains with DP ≥ 4 were found in rather small number. If the ratio of a:b chains in the clusters
were similar to that in their parent amylopectins, this suggests that the chains with DP 3 were
almost entirely composed of b-chains. Chains with DP 2 in TOG 12440 and WITA 4 were well
below 50%. However, in these samples chains with DP 3 were considerably more abundant
than in the other two samples. Possibly, therefore, chains with DP 3 in TOG 12440 and WITA
4 contained a considerable number of a-chains. Another possibility is that the number of achains, and thus the ratio of a:b chains indeed did change in some of the samples but not in
others. In any case, the differences in the composition of the short chains suggested that there
existed specific differences in the length of the segments involved in the interconnection of
clusters in the amylopectin samples.
Table 7.2 also shows the composition of the sub-categories of the b-chains. In all samples, b0chains predominated and the other b-chains were present in reverse number to their length.
Compared to the number of long B2- and B3-chains in the original amylopectin samples (Table
6.2), the number of long chains in clusters (b3-chains) was low, which shows that they had been
attacked by the α-amylase. There were no special difference in the chain composition between
the African samples and the Asian sample. In fact, all of the chain categories presented in
Table 7.2 possessed the most significant differences between the two African rice samples. The
most probable explanation for this is that these two samples were chosen based on their large
differences in pasting properties: Despite having about the same amylose:amylopectin ratio,
their pasting profile differed remarkably.
While peak time was similar and the pasting
temperature of IRGC 103759 (78.1°C) differed only slightly from that of TOG 12440 (78.9 °C),
207
the peak and trough viscosities of IRGC 103759 (253.6; 145.4 RVU) were significantly higher
than the corresponding values for TOG 12440 (235.6; 139.3 RVU, respectively). IRGC 103759
also possessed higher values of final, breakdown and setback viscosities. Specific differences in
the length of the segments involved in the interconnection of clusters in the amylopectin
samples, leading to differences in the packing of the crystallites in the starch granules
(Vamadevan et al. 2013), may have accounted for these disparities in pasting characteristics.
7.3.4
Characterization of Building Blocks in Clusters
Building blocks are defined as very small and tightly branched units in clusters and they can be
produced by extensive α-amylolysis of the clusters at high enzyme concentrations (Bertoft et al.
1999; Bertoft 2007b; Bertoft et al. 2011). In addition, small quantities of linear dextrins with
DP 1-6 are formed (Bertoft 2007b; Bertoft et al. 2011) and of these dextrins, those with DP 4-6
are hydrolysed into DP 1-3 when β-amylase is added. This improves the separation of these
small dextrins from the branched building blocks in GPC. Figure 7.6 shows the size
distributions of building blocks in the clusters of the rice amylopectins analyzed on Superdex
30. The profile of building blocks shows that group 2 (the smallest building blocks which
consist of two chains) was most common and group 3 (which consist of three chains) was the
second most common type. Groups 4, 5 and 6 (i.e. building blocks with increasing number of
chains) were found in sequentially decreasing amounts in all the rice amylopectin samples.
Group 1, which was comprised of glucose, maltose and maltotriose, was found in specific
amounts in the different samples.
Figure 7.7a shows an example of the building blocks
distribution when analysed by HPAEC-PAD. This technique gives a much higher resolution of
building blocks with individual DP. In addition, the peaks of glucose, maltose and maltotriose
208
included in group 1 are completely resolved. However, group 5 and 6 cannot be resolved by
HPAEC.
One example of the unit chain distribution of building blocks (from TOG 12440) is shown in
Fig. 7.7b. After subtraction of the pre-existing amount of glucose, maltose and maltotriose
(Fig. 7.7a) the unit chain composition of the branched building blocks can be estimated and a
typical unit chain profile is shown in Fig. 7.7c as bar graph. The peak of maltose represented achains and, as in the clusters, the peak of maltotriose contained a mixture of a- and b-chains of
unknown proportions. The rest of the chains with DP ≥ 4 were all b-chains and they typically
possessed a peak at DP 5. High amounts of chains at DP 4, 6 and 7 were also generally present.
Chains longer than DP 10 were mostly only represented in trace amounts.
The structural characterisation of the branched building blocks in clusters is summarised in
Table 7.3. The branched building blocks denoted the major part by weight of the samples
(73.5-83.9%), and by mole they were between 30.0-41.9%. The average DP of these branched
blocks ranged between 10.6-13.4 with the two African rice samples and NERICA 4 having
significantly higher values than the Asian rice sample. As expected, the NC, CL and ICL were
much lower and DB was higher in the building blocks compared to the clusters (Table 7.1).
The two African rice samples had building blocks with higher NC, longer CL and higher DB
than the Asian rice sample (Table 7.3). The same can be said for NERICA 4, except that it had
a somewhat lower DB than the Asian rice sample.
The composition of building blocks in the clusters is summarised in Table 7.4. The density of
building blocks (DBbl), which is a parameter analogous to the density of branches (Bertoft et
al. 2010), ranged from 6.1 to 7.0, of which the latter corresponded to WITA 4. Thus, African
rice accessions and NERICA 4 possessed lower density of building blocks than WITA 4.
209
IB-CL corresponds to the average length of the chains between the branches in two adjacent
building blocks in the cluster (Bertoft et al. 2010) and was similar in all samples. The average
number of building blocks (NBbl) in the clusters was significantly larger in NERICA 4 and
WITA 4 than in the two African rice accessions.
The relative distribution of the different groups of building blocks is also shown in Table 7.4.
Group 2 was the major type in all samples and it represented between 40.3-49.4% of the blocks.
Group 3 was the second largest group representing 27.0-29.3%, while group 4 was represented
by roughly 11.0-12.5% of the blocks in all cluster samples. Similar amounts of group 5 as of
group 4 were present in the clusters, whereas group 6 represented only between 1.0 and 4.4% of
the building blocks. The most significant difference between the African and the Asian rice
samples was a higher representation of the large building blocks of groups 5 and 6 in the former
samples. Instead, group 2 was present in lower amounts in African rice samples. Indeed, the
abundant presence of large blocks of groups 4, 5 and 6 in TOG 12440 resulted in the large
average DP (13.4) of the branched blocks.
Generally, the size-distribution of the building blocks in the rice samples was similar to that
found for most other starches previously (Bertoft et al. 2012; Zhu et al. 2013; Kong et al.
2008)). However, the size-distribution of TOG 12440 was rather unique, as it possessed the
lowest number of group 2 blocks so far recorded. In addition, building blocks of group 5 were
present in higher amounts than group 4, which also was a unique feature. More African rice
samples should be analysed for their building block compositions before it can be concluded if
these starches in general tend to possess larger building blocks than other starches.
210
7.3.5
Correlations
Despite the small sample size of the study, correlations between IB-CL and NBbl and some
thermal and pasting properties of starch from the rice samples were performed to check the
existence of any relationships. NBbl correlated negatively (-0.7522; p<0.05) with enthalpy but
all other correlations were not significant (p<0.5).
7.3.6
Conclusions
The average DPs of the clusters from the African rice samples were slightly larger (DP ~83)
than in the Asian variety (DP ~81) while NERICA 4 (DP 89.6), which represents a cross
between African and Asian rices, possessed the largest clusters. The short CL of TOG 12440
accounted for its large NC and similarly, the clusters of NERICA 4, had intermediate NC due to
its long CL. Generally, the results showed that the structure of the clusters in TOG 12440 were
dense with short chains and high DB while clusters in NERICA 4 were less dense with long
chains and a low DB.
IRGC 103759 and WITA 4, however, possessed clusters with
intermediate characteristics. The most significant difference between African and Asian rice
appeared to be found in the composition of building blocks. The two African rice samples and
NERICA 4 had significantly higher average DP of the branched blocks than the Asian rice
sample due to the abundance of large building blocks of group 6. This also resulted in higher
average NC, longer CL and higher DB of the building blocks in the African rice samples. The
major type of building blocks in all samples was, however, group 2 followed by group 3.
211
7.3.7
Acknowledgements
This research work was undertaken with the kind assistance of the Global Rice Science
Partnership (GRiSP) and the Africa Rice Center (AfricaRice) in terms of financial support and
supply of samples.
212
7.3.8
Tables and Figures
Table 7.1. Characterization of φ,β Limit Dextrins of Clusters from Amylopectin#
TICL6
Sample
DP1
NC2
DB (%)3
CL4
ICL5
African rice
IRGC 103759
82.8b
11.8a
13.0a
7.1c
5.1c
8.5b
TOG 12440
83.3c
14.1c
15.8c
5.9a
3.8a
7.9a
89.6d
12.4b
12.7a
7.3d
5.3c
8.8b
81.0a
12.0ab
13.7b
6.8b
4.7b
7.9a
O. sativa x O. glaberrima cross
NERICA 4
Asian rice
WITA 4
#
Values followed by the same letters in each column are not significantly different (p < 0.05)
1
Average Size of the Cluster (DP) as Measured by GPC (Fractions containing < 1 % Carbohydrate
were excluded)
2
Number of Chains = DP/CL
3
Degree of Branching = (NC – 1)/DP x 100
4
Average Chain Length Estimated by HPAEC
5
Internal Chain Length = (CL – ECL) x NC/(NC – 1) – 1. (ECL = 1.5)
6
Total Internal Chain Length = CL of b-Chains – 1
213
Table 7.2. Relative Molar Distribution (%) of Different Chain Categories in Clusters Isolated from Amylopectin#
a
(DP2)1
a and/or b
(DP3)2
b
(DP≥4)3
b0
(DP4-DP6)4
b1a
(DP7-DP10)4
b1b
(DP11-DP18)4
b2
(DP19-DP27)4
b3
(DP>28)4
IRGC 103759
49.5c
9.8a
40.7a
15.1a
11.3a
8.6ab
3.5b
2.3c
TOG 12440
43.5a
12.8b
43.7b
20.6c
12.4c
8.2a
1.9a
0.6a
49.1c
7.7a
43.1b
16.6b
11.9b
8.6ab
3.5b
2.6d
46.0b
13.3b
40.7a
14.9a
11.8b
9.1c
3.3b
1.7b
Sample
African rice
O. sativa x O. glaberrima cross
NERICA 4
Asian rice
WITA 4
#
values followed by the same letters in each column are not significantly different (p < 0.05)
1
a-Chains
2
Mixture of a- and b-Chains
3
b-Chains
4
Categories of b-Chains of Clusters (DP ≥ 4)
214
Table 7.3. Structure of Branched Building Blocks in Clusters of Amylopectin from African Rice and Asian Rice#
NC1
DB2 (%)
CL3
Sample
Weight (%)
Mole (%)
DP
African rice
ICL4
IRGC 103759
75.6b
32.1b
12.8b
3.2c
17.2ab
4.0ab
1.9ab
TOG 12440
83.9c
41.9d
13.4c
3.5d
18.3b
3.9a
1.7a
73.5a
30.0a
12.5b
3.0b
16.1a
4.2b
2.3b
76.4b
36.3c
10.6a
2.8a
17.1ab
3.8a
1.8a
O. sativa x O. glaberrima cross
NERICA 4
Asian rice
WITA 4
#
Values followed by a different superscript in each row are significantly different (p< 0.05)
1
Number of Chains = DP/CL
2
Degree of Branching = (NC – 1)/DP x 100
3
Average Chain Length Estimated by HPAEC
4
Internal Chain Length = (CL – ECL) x NC/(NC – 1) – 1. (ECL = 1.5)
215
Table 7.4. Building Block Structures of Clusters of Amylopectin from African Rice Accessions and Asian Rice Varieties#
Molar distribution of groups of building blocks in clusters4
Sample
DBbl1
IB-CL2
NBbl3
Group 2
Group 3
Group 4
Group 5
Group 6
African rice
IRGC 103759
6.4a
6.9a
5.3a
46.5b
27.0a
11.0a
11.8c
3.8b
TOG 12440
6.1a
6.5a
5.1a
40.3a
28.7b
12.5b
14.4d
4.4b
6.4a
6.8a
5.8b
46.6b
27.4a
11.2a
11.2b
3.8b
7.0b
6.9a
5.7b
49.4c
29.3b
11.2a
9.3a
1.0a
O. sativa x O. glaberrima cross
NERICA 4
Asian rice
WITA 4
#
Values followed by a different superscript in each column are significantly different (P < 0.05)
1
Density of Building Blocks (DBbl) = (NBbl/DPcluster) x 100
2
Inter Block Chain Length (IB-CL) = Mole% Linear Dextrins x DPlinear dextrins / Mole% Branched Building Blocks + 4
3
Average Number of Branched Building Blocks (NBbl) = Wt % Bbl/100 x DPcluster/DPBbl
4
Group of Branched Building Blocks Based on Their Branches
216
Degree of polymerization
103
102
1
1
5
A
20 min
Carbohydrate (wt. %)
4
40 min
B
60 min
3
80 min
90 min
2
100 min
1
120 min
140 min
0
40
60
80
100
120
Fraction number
140
160
180
Figure 7.1. Time Course for α-Amylolysis of Rice Amylopectin by Gel-Permeation
Chromatography on Sepharose CL-6B.
217
Rice amylopectin
0.80
Branched dextrins (relative number)
0.70
0.60
0.50
0.40
0.30
0.20
0.10
0.00
20
40
60
80
Time (min)
100
120
140
Figure 7.2. Development of Relative Molar Amounts of Branched Dextrins in the
Time Course for α-Amylolysis of Rice Amylopectin for the Production of Clusters.
218
Degree of polymerization
500
1000
100
1
10
9
Carbohydrate content (wt%)
8
IRGC 103759
7
TOG 12440
6
NERICA 4
WITA 4
5
4
3
2
1
0
80
90
100
110
120
130
140
150
Fraction number
Figure 7.3. Size Distribution of φ,β-limit Dextrins of Clusters Isolated from Rice
Amylopectin by GPC on Sepharose CL-6B.
219
160
3
100
80
PAD signal
5
4
60
6
7
8
40
9
20
0
0
10
20
Time (min)
30
Figure 7.4. HPAEC Chromatogram of Unit Chain Distribution of Cluster from
TOG 12440 (African rice).
220
40
10
b3
b2
b1
b0
8
Carbohydrate content (%)
b1b
b1a
6
4
2
0
0
10
20
30
Degree of polymerization
40
50
Figure 7.5. Internal Chain Length Distribution of Cluster from TOG 12440 (African rice).
Chains with DP ≥ 4 are shown and different b-chain categories are indicated. Peak areas in Fig.
7.4 were corrected to relative carbohydrate content.
221
60
Degree of polymerization
10
25
50
7
5
2
2
3
6
Carbohydrate (wt %)
IRGC 103759
4
TOG 12440
5
1
NERICA 4
WITA 4
4
5
3
6
2
1
0
50
60
70
80
90
100
110
120
130
140
Fraction number
Figure 7.6. Size Distribution Obtained by GPC on Superdex 30 of Building Blocks in
Clusters of Rice Amylopectin. Numbers Indicate Different Groups of Building Blocks.
222
Impurity in enzyme preparation
PAD Signal
600
400
1 2
200
3
5
6
7
8 9 10 11 12
0
0
10
20
30
40
50
Time (min)
Figure 7.7a. Building blocks distribution of debranched TOG 12440 (African rice)
when analysed by HPAEC-PAD.
DPs 1-12 are indicated
1 2
600
PAD signal
Impurity in enzyme
preparation
400
3
4
200
5
6
7
8
9
10 11 12
0
0
10
20
Time (min)
30
40
Figure 7.7b. Unit chain distribution of building blocks (from TOG 12440).
DPs 1-12 are indicated
223
50
Carbohydrates (wt. %)
20
15
10
5
0
1
2
3
4
5
6
7
8
9
10 11 12 13 14 15 16 17 18 19 20 21
Chain Length
Figure 7.7c. Example of Unit Chain Compositions in Building Blocks of TOG 12440
(African rice) on Weight Basis.
224
7.3.9
References
Bertoft, E. 1989. Investigation of the fine structure of amylopectin using alpha- and betaamylase. Carbohydr. Res. 189: 195-207.
Bertoft, E. and Spoof, L. 1989. Fractional precipitation of amylopectin alpha dextrins
using methanol. Carbohydr. Res. 189: 169-180.
Bertoft, E. 1991. Investigation of the fine structure of alpha-dextrins derived from
amylopectin and their relation to the structure of waxy-maize starch. Carbohydr.
Res. 212: 229-244.
Bertoft, E., Manelius, R., and Qin, Z. 1993. Studies on the structure of pea starches. part 1:
initial stages in -amylolysis of granular smooth pea starch. Starch/Stärke. 45: 215220.
Bertoft, E., Zhu, Q., Andtfolk, H. and Jungner, M. 1999. Structural heterogeneity in waxyrice starch. Carbohydr.Polym. 38: 349-359.
Bertoft, E., and Koch, K. 2000. Composition of chains in waxy-rice starch and its
structural units. Carbohydr. Polym. 41: 121-132.
Bertoft, E. 2004. On the nature of categories of chains in amylopectin and their
connection to the super helix model. Carbohydr. Polym. 57:211-224.
Bertoft, E. 2007a. Composition of clusters and their arrangement in potato amylopectin.
Carbohydr. Polym. 68: 433-446.
225
Bertoft, E. 2007b. Composition of building blocks in clusters from potato amylopectin.
Carbohydr. Polym. 70: 123-136.
Bertoft, E., Piyachomkwan, K., Chatakanonda, P., and Sriroth, K. 2008. Internal
unit chain composition in amylopectins. Carbohydr. Polym. 74: 527-543.
Bertoft, E., Laohaphatanalert, K., Piyachomkwan, K., and Sriroth, K. 2010. The fine
structure of cassava starch amylopectin. Part 2: Building block structure of clusters.
Int. J. Biol. Macromol. 47: 325-335.
Bertoft, E., Källman, A., Koch, K., Andersson, R., and Åman, P. 2011. The cluster
Structure of barley amylopectins of different genetic backgrounds.
Int. J. Biol. Macromol. 49: 441-453.
Bertoft, E., Koch, K. and Åman, P. 2012a. Building block organisation of clusters
in amylopectin from different structural types. Int. J. Biol. Macromol. 50: 12121223.
Bertoft, E., Koch, K. and Åman, P. 2012b. Structure of building blocks in amylopectins.
Carbohydr. Res. 361: 105-113.
Buléon, A., Colonna, P., Planchot, V., and Ball, S. 1998. Starch granules: structure
and biosynthesis. Int. J. Biol. Macromol. 23: 85-112.
Dubois, M., Gilles, K. A., Hamilton, J. K., Rebers, P. A., and Smith, F. 1956.
Colorimetric method for determination of sugars and related substances.
Anal.Chem. 28: 350-356.
226
Imberty, A., Buléon, A., Tran, V., and Péerez, S. 1991. Recent advances in knowledge of
starch structure. Starch/Stärke. 43: 375-384.
Jenkins, P. J., Cameron, R. E., and Donald, A. M. 1993. A universal feature in the
structure of starch granules from different botanical sources. Starch/Stärke. 45:
417-420.
Klucinec, J. D., and Thompson, D. B. 1998. Fractionation of high-amylose maize starches
by differential alcohol precipitation and chromatography of the fractions.
Cereal Chem. 75: 887-896.
Kong, X., Corke, H., and Bertoft, E. 2009. Fine structure characterization of amylopectins
from grain amaranth starch. Carbohydr. Res. 344: 1701-1708.
Koch, K., Andersson, R. and Åman, P. 1998. Quantitative analysis of amylopectin unit
chains by means of high-performance anion-exchange chromatography with pulsed
amperometric detection. J. Chromatogr. A. 800: 199-206.
Laohaphatanaleart, K., Piyachomkwan, K., Sriroth, K., Santisopasri, V., and Bertoft, E.
2009. A study of the internal structure in cassava and rice amylopectin.
Starch/Stärke. 61: 557-569.
Manners, D. J. 1989. Recent developments in our understanding of amylopectin structure.
Carbohydr. Polym. 11: 87-112.
Nakamura, Y. 2002. Towards a better understanding of the metabolic system for
amylopectin biosynthesis in plants: rice endosperm as a model tissue. Plant and
Cell Physiology 43: 718-725.
227
Nanri, A., Mizoue, T., Noda, M., Takahashi, Y., Kato, M., and Inoue, M. 2010. Rice
intake and type 2 diabetes in Japanese men and women: the Japan public health
center-based prospective study. Am. J. Clin. Nutr. 92: 1468-1477.
Peat, S., Whelan, W., and Thomas, G. J. 1952. Evidence of multiple branching in waxy
maize starch. J. Chem. Soc. 4546- 4548.
Pérez, S., and Bertoft, E. 2010. The molecular structures of starch components and their
contribution to the architecture of starch granules: A comprehensive review.
Starch/Stärke. 62: 389-420.
Robyt, J., and French, D. 1963. Action pattern and specificity of an amylase from Bacillus
subtilis. Arch. Biochem. Biophys. 100: 451-467.
Takeda, Y., Hizukuri, S., and Juliano, B. O. 1987. Structures of rice amylopectins with
low and high affinities for iodine. Carbohydr. Res. 168: 79-88.
Vamadevan, V., Bertoft, E., Soldatov, D. V., and Seetharaman, K. 2013. Impact of
molecular organization of amylopectin in starch granules upon annealing.
Carbohydr. Polym. 98: 1045-1055.
Zhu, F., Corke, H., Åman, P. and Bertoft, E. 2011. Structures of clusters in sweet potato
amylopectin. Carbohydr. Res. 346: 1112-1121.
Zhu, F., Bertoft, E and Seetharaman, K. 2013. Characterization of internal structure of
maize starch without amylose and amylopectin separation. Carbohydr. Polym. 97:
475-481.
228
CHAPTER 8: CONCLUSIONS AND FUTURE WORK
Due to the importance of starch in rice quality, cooking and processing, this study focused
on the structural and functional characteristics of flour and starch from African rice (Oryza
glaberimma). Samples of African rice accessions (ARAs) were objectively selected from a
collection of accessions with a diverse range of apparent amylose concentration and pasting
characteristics. The hierarchical cluster analysis of the 1,020 rice accessions using Ward
method which resulted in five main clusters and twenty-five sub clusters, revealed that
some of the groupings had similar apparent amylose content but interestingly, their pasting
properties were very different; a situation that necessitated further investigations. No waxy
types of ARAs were analyzed in this study and it is also worthy to note that in comparison
to the O. sativas (Asian rice) the ARAs seem to have a narrower range of apparent amylose
content (AAC). This observation is important since AAC is related to the cooking quality
of rice and other physico-chemical properties such as gel consistency, pasting viscosity and
gelatinization temperature. Additionally, AAC is also considered to be one of the most
important predictors of various rice products and the eating quality of cooked rice.
The African rice population of 1,020 accessions showed pasting temperatures that were far
higher than those reported on Asian rice in previous studies. In addition, their peak
viscosity range was narrower, while final viscosity range was broader than earlier reports.
The peak viscosity highly correlated with trough, breakdown and final viscosities (p<0.01)
in all the five clusters while breakdown viscosity correlated negatively (p<0.01) with both
peak time and pasting temperature in all the five clusters. Further to these observations,
final viscosity also correlated highly (p<0.01) with setback viscosity in all the five clusters
while correlation between peak viscosity and apparent amylose content was not significant
229
(p<0.01) within clusters II and IV. This categorization could serve as a useful tool for
selecting materials with promising pasting properties that can be employed in the
development of new rice cultivars. Based on grain dimensions and apparent amylose
contents, the seven ARAs (IRGC 86741, IRGC 80787, IRGC 103759, RAM 123, TOG
7682, TOG 12440 and TOG 14365) selected from the population of 1,020 through cluster
analysis were medium shaped with intermediate to high amylose rice types and relatively
high protein content. TOG 12440 was also a high amylose rice type however, it had bold
shape (L/W=1.91). Their relatively higher amylose concentration was a major factor in
influencing some of the cooking quality and textural parameters measured. The cooked
kernels of the selected ARAs had relatively lower cooking solids loss and harder texture
compared to the Asian rice varieties which were softer and more adhesive. The Asian rice
flours were generally more viscous than the ARAs but showed a higher breakdown
tendency.
The in vitro digestibility of the cooked kernel flours showed significant
differences in the expected glycemic indices of the samples but they were all >70.
Gelatinization transition temperatures (To, Tp, Tc) and enthalpy were significantly higher in
the ARAs flours as compared with Asian rice. However, melting transition temperatures of
retrograded flour gels were similar. The high enthalpy of gelatinization (ΔH) of the ARAs
is an indication of a more stable structure that was relatively more resistant to granule
swelling. The shear stability exhibited by the ARAs may have resulted in the negligible
breakdown. The gelatinization transition temperatures (To, Tp, Tc) of the starches and
flours were comparable but the ΔH of starch granules were significantly higher than the
flours.
A similar trend was observed for the melting transition temperatures of the
retrograded starch and flour gels.
230
Scanning electron micrographs of the rice starch granules showed polyhedral shaped
granules that were tightly packed and the granule size distribution of the starches ranged
from 2-22 μm. Additionally, all the starches displayed the typical 'A' type x-ray diffraction
pattern.
The starches showed significant (p < 0.05) differences in the ratio of the
absorbance to scattering coefficient (K/S) values when exposed to iodine vapor with the
ARAs starches recording higher K/S values than the Asian rice samples. This suggested
that the glucan chains of the ARAs were less rigid and more available to form inclusion
complexes with compared to the Asian rice. The Asian rice starches, apart from having
similar physical and morphological characteristics as ARAs starches, differed in their
molecular composition (amylose content) and structure, which might affect their functional
properties.
Based on a system of classification of amylopectins proposed by Bertoft et al. (2008), the
structure of the amylopectins of two ARAs samples (TOG 12440 and IRGC 103759) in
comparison with two Asian rice samples (Koshihikari and WITA 4) and NERICA 4 (which
is an inter specific cross between sativa x glaberrima) falls into type 2. The profiles of the
unit chain distribution of the amylopectins of the ARAs samples and NERICA 4 were
distinct from the Asian rice. The external chain length (ECL) of the ARAs and NERICA 4
were longer than the Asian rice varieties. The data suggest that the ARAs and NERICA 4
can possibly have two complete turns of glucosyl units in their double helices, since six
glucosyl units are needed per turn, while the Asian rice varieties generally have less than
two turns. This may explain the higher gelatinization enthalpies observed in the ARAs.
The positive correlations between CL, SCL, ECL and gelatinization transition temperatures
(To, Tp, Tc) and enthalpy (ΔH) strongly suggest that the short chains of the amylopectins,
231
especially their external segments were important in influencing the thermal and pasting
properties of the starch samples.
The average DPs of the clusters from the African rice samples were significantly (p<0.05)
larger than in the Asian rice sample while NERICA 4 possessed the largest clusters.
Generally, the results showed that TOG 12440 clusters were dense with short chains and
high DB while clusters in NERICA 4 were less dense with long chains and a low DB.
However, clusters of IRGC 103759 and WITA 4 possessed with intermediate
characteristics.
The average DP of the branched building blocks of the African rice
samples and NERICA 4 were significantly (p<0.05) higher than the Asian rice sample and
the major type of building blocks in all samples was group 2 followed by group 3. Similar
amounts of groups 4 and 5 were present in the clusters, whereas group 6 represented only
up to 4.4% of the building blocks. Additionally, the African rice samples had building
blocks with higher number of chains and degree of branching than the Asian rice sample.
The hierarchical cluster analysis of the 1,020 rice accessions showed a high level of genetic
diversity in the accessions of O. glaberrima from the West Africa sub region as evidenced
by the variation in both apparent amylose content and pasting properties in and within the
clusters and sub clusters. The cooking quality, in vitro digestibility, physical, thermal and
pasting characteristics of the selected African rice starches and flour were established.
Additionally, the positive correlation between the chain lengths of amylopectin (CL, SCL,
ECL) and gelatinization transition temperatures strongly suggests that the chains influenced
the thermal and pasting properties of the starch samples. These findings were based on
seven samples selected through the cluster analysis. It will therefore be expedient to work
232
on more African rice samples to affirm the findings in this study especially in the area of
starch structure in relation to its functionality in cooking, processing and nutritional aspects
(digestibility).
Based on the information generated through this study, it will also be interesting to develop
some products from African rice and test their acceptability in a bid to promote more
patronage of African rice.
233