Characterization and Chemical Surface Gelatinization of Yellow Pea Starch

Home , Starch gelatinization

A THESIS SUBMITTED TO THE FACULTY OF UNIVERSITY OF MINNESOTA BY

Yan Zhou

IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE

George A. Annor Leonard F. Marquart

December 2019

©Yan Zhou 2019 Acknowledgement

First of all, I would like to thank my advisers, Dr. George Amponsah Annor and

Dr. Len Marquart for their support, direction and continuous encouragement throughout my master’s program. Special thanks to Dr. Eric Bertoft for his guidance throughout this project. I also want to thank my committee member, Dr. Tonya Schoenfuss for reviewing my thesis. A sincere thanks to Nancy Toedt, Dorit Hafner and Andrew Howe for helping me along the whole master’s program.

Thank you to everyone in the Cereal lab for their support and help during this journey. I am very thankful to Akua, Bu Fan and Juan for their assistance and training during my research.

Finally, I want to thank my family and friends for their enormous support and understanding.

i Dedication

This thesis is dedicated to my parents, Yu Zhou and Ling Wu, who have been supporting my decisions and encouraging me to explore more possibilities in life.

To my grandma Yinchai Gong for always believing in me and inspiring me to follow my dreams.

To my best friend Qi He and Sui Zhang, who have been listening to me and instilling confidence in me.

ii Abstract

This study investigated the molecular characteristics and amylopectin unit and internal chain profiles of two yellow pea starches with significantly different pasting properties, before and after they were chemically gelatinized with 13M lithium chloride solution to separate the periphery and the core of their starch granules. The sample with a lower peak, final and setback viscosities but a higher pasting temperature was labeled P1, while P2 had opposite pasting properties. Prior to chemical gelatinization, large starch granules (> 20 µm) were separated from the starches to ensure uniform chemical gelatinization. After physically separating the gelatinized portion (about 45%) of the granules, amylopectin was fractionated from all three fractions i.e. native large granules

(LG), remaining granules (RG) and the gelatinized portion of the granules (GP). RG represented the core while GP represented the periphery of the granules. Granule size distribution, iodine binding, molecular size distribution, thermal properties, percent crystallinity as well as the unit and internal chain profiles as j,b-limit dextrins of fractionated amylopectin were determined. Granules from both samples had a unimodal distribution with sizes averaging 25.44 and 26.51 µm for P1 and P2 respectively. Starches exhibited the typical C-type crystal allomorphs with amylose contents ranging between

33.27 and 28.37%. The two samples exhibited similar unit chain profiles, but significant differences were observed in their internal chains. Extracted amylopectin samples were observed to have super long chains with average chain lengths of 18.75 and 19.24 glucose units for P1 and P2 respectively. P1 with lower pasting viscosities had significantly shorter average chains and internal chains (4.95 glucose units) and P2 with lower gelatinization temperature had less of crystalline A-chains (Acrystal). The two samples had significantly

iii different internal chain profiles that could explain the differences in their pasting profiles.

The LG of P1 required 300 min to gelatinize to about 45% of the granule while P2 took

190 min. The results revealed that amylose was evenly distributed in P1 but more concentrated at the periphery of granules in P2. The periphery of granules had more short chain amylose than in the core for both samples. The short chain length (SCL) and long B- chains of j,b-limit dextrins (BLLD) were longer at the core where more of B-allomorphs are known to be distributed in both samples. Fingerprint A-chains (Afp), short B-chains

(BS) and B-chains were more concentrated at the periphery than at the core.

iv Table of Contents List of Tables ...... viii

List of Figures ...... ix

Chapter 1: Literature Review ...... 1

1.1 Pea ...... 2

1.2 Pea Starch ...... 3

1.2.1 Starch Composition ...... 3

1.2.2 Granule Morphology of Pea Starch ...... 5

1.2.3 Crystallinity and Polymorphic Composition of Pea Starch ...... 6

1.2.4 Pasting Properties ...... 7

1.2.5 Thermal Properties ...... 8

1.2.6 Swelling Power and Solubility ...... 9

1.2.7 Retrogradation ...... 10

1.2.8 Enzymatic hydrolysis ...... 11

1.2.9 Acid hydrolysis ...... 12

1.3 Starch Modification ...... 12

1.3.1 Physical Modification ...... 12

1.3.2 Chemical Modification ...... 13

1.4 Chemical Gelatinization ...... 14

1.5 Rationale ...... 16

1.6 Objectives ...... 16

Chapter 2. Functional Characteristics of Yellow Pea Starches as Affected by

Amylopectin Unit and Internal Chains ...... 17

2.1 Introduction ...... 18

2.2 Materials and methods ...... 20

v 2.2.1 Materials ...... 20

2.2.2 Granule size distribution ...... 20

2.2.3 X-ray diffraction ...... 20

2.2.4 Amylose content ...... 21

2.2.5 Amylopectin fractionation ...... 21

2.2.6 Production of j,b-limit dextrins of amylopectin ...... 22

2.2.7 Unit chain length profiles of amylopectin and f,b-limit dextrins ...... 23

2.2.8 Iodine binding ...... 24

2.2.9 Thermal properties ...... 24

2.2.10 Pasting properties ...... 25

2.2.11 Statistical analysis ...... 25

2.3 Results ...... 25

2.3.1 Granule size distribution ...... 25

2.3.2 Wide angle X-ray diffraction ...... 26

2.3.3 Amylose content ...... 27

2.3.4 Unit and internal chain profile ...... 28

2.3.5 Iodine affinity ...... 39

2.3.6 Pasting properties ...... 40

2.3.7 Thermal properties ...... 41

2.4 Discussion ...... 42

2.5 Conclusion ...... 44

Chapter 3: Amylopectin Unit and Internal Chain Profiles of Periphery and Core of

Pea Starch Granules as Revealed by Chemical Gelatinization ...... 45

3.1 Introduction: ...... 46

3.2 Materials and methods: ...... 48

vi 3.2.1 Materials ...... 48

3.2.2 Fractionation of starch granules ...... 48

3.2.3 Chemical surface gelatinization ...... 48

3.2.4 Separation of the gelatinized starch from the remaining granules ...... 49

3.2.5 Amylose content ...... 49

3.2.6 Amylopectin fractionation ...... 50

3.2.7 Production of j,b-limit dextrins ...... 51

3.2.8 Unit and internal chain profiles ...... 51

3.2.9 Statistical analysis ...... 52

3.3 Results: ...... 53

3.3.1 Comparison of large granules and original granules ...... 53

3.3.2 Extent of gelatinization ...... 55

3.3.3 Amylose distribution of pea starch granules ...... 57

3.3.4 Distributions of unit and internal chain profiles of pea starch amylopectin ...... 58

3.4 Discussion ...... 72

3.5 Conclusion ...... 74

Chapter 4: Conclusion and Future Work ...... 75

Chapter 5: Reference ...... 77

Chapter 6. Appendices ...... 91

Appendix A. Pasting properties of two yellow pea starch samples...... 92

Appendix B. Thermal properties of two yellow pea starch samples...... 93

vii List of Tables Table 2.1 Amylose Content and Composition of Yellow Pea Starches...... 28

Table 2.2 Average Chain Lengths (CLs) of Different Chain Categories and φ, β-Limit

Values of Pea Amylopectin Obtained by High Performance Anion-Exchange

Chromatography...... 34

Table 2.3 Relative Molar Amounts (%) of Chain Categories in Amylopectins of Pea

Starches...... 36

Table 2.4 Selected Molar Ratios of Different Chain Categories of Pea Amylopectins and

Their j, β-Limit Dextrins...... 38

Table 2.5 Iodine affinity values of two yellow pea starch samples...... 39

Table 2.6 Pasting properties of two yellow pea starch samples...... 40

Table 2.7 Thermal properties of two yellow pea starch samples...... 41

Table 3.1 Amylose content of Large, Gelatinized and Remaining Granules...... 57

Table 3.2 Average Chain Lengths (CLs) of Different Chain Categories and φ,β-Limit

Values of Large Granules and Their Gelatinized and Remaining Fraction

Amylopectins...... 68

Table 3.3 Relative Molar Amounts (%) of Chain Categories of Large Granules and Their

Gelatinized and Remaining Fraction Amylopectins...... 70

Table 3.4 Selected Molar Ratios of Different Chain Categories of Large Granules and their

Gelatinized and Remaining Fraction Amylopectin and φ,β-Limit Dextrins...... 71

viii List of Figures

Figure 1.1. Smooth and wrinkled pea starch...... 3

Figure 1.2. Scanning electron micrographs of native wrinkled pea starch granules...... 6

Figure 2.1. Particle size distributions of two yellow pea starch samples in the dry state. 26

Figure 2.2. X-ray diffraction patterns of two yellow pea starch samples...... 27

Figure 2.3. Sepharose CL-6B gel-permeation chromatography of debranched yellow pea

starches and their extracted amylopectin. LCAM = long-chain amylose; SCAM = short-

chain amylose; and AMP = amylopectin chains...... 29

Figure 2.4. Unit chain profiles of debranched yellow pea amylopectins by high

performance anion-exchange chromatography, with arrows and numbers showing

degree of polymerization...... 31

Figure 2.5. Unit chain profiles of debranched yellow pea amylopectins φ,β-limit dextrins

of pea amylopectins. obtained by high-performance anion-exchange chromatography,

with arrows and numbers indicating degree of polymerization. Short B-chains (BS)

are subdivided into “fingerprint” B-chains (Bfp) and a major group (BSmajor),

whereas long B-chains (BL) are subdivided into B2- and B3-chains...... 32

Figure 3.1. Polarized light-micrographs of large granules of yellow pea starch in 13M LiCl

solutions at different times. A: P1, treated with 60 min; B: P1, treated with 180 min;

C: P1, treated with 300 min; D: P2, treated with 60 min; E: P2, treated with 120 min;

F: P2, treated with 190 min...... 56

Figure 3.2. Sepharose CL-6B gel-permeation chromatography of debranched fractions of

pea starches and their amylopectins (A: Large granules (LG); B: Gelatinized portion

ix (GP); C: Remaining granules (RG)). LCAM = long-chain amylose; SCAM = short-chain

amylose; and AMP = amylopectin chains...... 59

Figure 3.3. Unit chain profiles of debranched amylopectins of pea starches fractions

obtained by high performance anion-exchange chromatography, with arrows and

numbers showing degree of polymerization. LG: Large granules; GP: Gelatinized

portion; RG: Remaining granules...... 63

Figure 3.4. Chain profiles of φ,β-limit dextrins of amylopectins of pea starches fractions

obtained by high performance anion-exchange chromatography, with arrows and

numbers indicating degree of polymerization. Short B-chains (BS) are subdivided into

“fingerprint” B-chains (Bfp) and a major group (BSmajor), whereas long B-chains

(BL) are subdivided into B2- and B3-chains. LG: Large granules; GP: Gelatinized

portion; RG: Remaining granules...... 66

Figure 3.5. Polarized light-micrographs of remaining granules (RG) of yellow pea starch

after washed with cold water. A: RG of P1, treated for 300 min; B: RG of P2, treated

for 190 min...... 72

Figure 6.1. Pasting properties of two yellow pea starch samples...... 92

Figure 6.2. Thermal properties of two yellow pea starch samples...... 93

Chapter 1: Literature Review

1 1.1 Pea

Peas (Pisum sativum L.), also called common pea, field pea, garden pea and dry peas are annual cool season crops, belonging to the Leguminosae family with plants such as field peas, chickpeas, cowpeas, black grams, lentils and the common beans (Allen &

Allen, 1981; Deshpande, 1990). There are two seed phenotypes of pea (Figure 1.1), smooth pea (with a smooth seed surface) and wrinkled pea (with a smooth seed surface). It has been shown that smooth pea seed has a lower soluble sugar content than wrinkled pea starch (Cousin, 1997). Three different cotyledon colors of pea seed have been reported, yellow, green and yellow-green (Borowska, Zadernowski, Borowski, & Swiecicki, 1998).

But the color of seeds might be influenced by the extent of ripeness, storage conditions and other factors. The starch of the two pea types have different granular morphologies and functionalities. Peas are cultivated worldwide with the production from Canada, Russia,

China, India and United States significantly contributing to world production. Peas are ranked fourth to soybeans, groundnuts and dry beans in terms of production, with Canada as the leader in global pea production and export (N. Wang & Daun, 2004). Most of peas produced in United States are used for livestock feed, primarily blended with grains to increase the protein content of feed. The rest are consumed as human food, such as hulls in high-fiber bread and immature green seeds in soups. Pea is a rich source of protein, fiber and carbohydrates (N. Wang & Daun, 2004). Starch and protein content of peas are about

30-50% and 20% to 25% respectively on dry basis (Shi et al., 2014). Pea protein is considered to be a good alternative protein source to soybean due to its well-balanced amino acid profiles and higher levels of lysine and tryptophan (Nunes, Raymundo, & Sousa,

2006).

Figure 1.1. Smooth and wrinkled pea starch.

1.2 Pea Starch

1.2.1 Starch Composition

Starch is mainly composed of amylose and amylopectin. Amylose is the minor part of starch, made up of a (1→4) linked glucose units. The degree of polymerization of pea amylose ranges from 1000 to 1400 glucose units. Amylopectin, the major fraction of starch, consists of linear a (1→4) linked glyosidic units with a (1→6) branches. Amylopectin chain distribution can be analyzed by high performance anion exchange chromatography

(HPAEC) and can be grouped into A-chains (which do not carry any other chains), B- chains (which carry other chains) and C-chains (with a reducing-end group) (Peat, Whelan,

& Thomas, 1952). Hizukuri (1985) reported that the unit chains of amylopectin can also be classified as short chains (DP 6-35) and long chains (DP > 35). Depending on the branched structure of amylopectin, the chains can be categorized into external chain and internal chain. It has been considered that the internal chains are mostly in the amorphous part of the amylopectin molecules while the external chains are found in the crystalline part

3 (Pérez & Bertoft, 2010). The internal chains of amylopectin can be obtained by using enzymes such as b-amylase remove external chains. b-amylase acts on a-(1,4) glucosidic linkages from non-reducing ends groups to remove maltose units, producing b-limit dextrin.

A-chains show as maltose or maltotriose stubs depending on whether the number of unit is odd or even but B-chains generally show as glucose or maltose stubs (Peat et al., 1952).

Amylopectin j-limit dextrins can be obtained by phosphorylase a from rabbit muscle.

When j-limit dextrins are subsequently subjected to b-amylase, j,b-limit dextrins are obtained, in which all A-chains appear as maltosyl stubs (Eric Bertoft & Spoof, 1989). The internal B-chains, corresponding to DP > 3 in j,b-limit dextrins, can be divided into short

(BS) and long (BL) chains at around DP 25, depending on the sample. The BS-chains can be grouped into the fingerprint B-chains (DP 3-8) and the major B-chains (DP 9-25). BL- chains can be divided into B2 and B3 group at DP 50 (Eric Bertoft, Piyachomkwan,

Chatakanonda, & Sriroth, 2008).

To understand the organization of chains in amylopectin, the structure of molecule has been described with different models over the years. The most popular of these models is the traditional cluster model first introduced independently by French and Nikuni in

((1978)) and subsequently modified by Hizukuri. Recently introduced is the backbone model proposed by Bertoft (Eric Bertoft, 2013). According to the model, long amylopectin chains (BL- chains) form a backbone in the amorphous lamella on which building blocks are distributed. Some of relatively longer BS-chains can also serve as the backbone. The weight ratio of short to long chains of smooth pea amylopectin had been reported to be similar with that of chickpea, but higher than that of cowpea due to less long chains of

4 amylopectin in smooth pea (Huang et al., 2007). However, limited research has been done on amylopectin internal structure of pea starch so far.

Amylose, the minor component of starch, also influences the properties of starches.

There are significant differences in amylose content between smooth pea and wrinkled pea, which might result from cultivar differences (Rosenthal, Espindola, Serapiao, & Silva,

1971), the amylose determination methods and the stage of seed development (Banks,

Greenwood, & Muir, 1974). The amylose content of smooth pea ranges from 33.1-49.6%, whereas that of wrinkled pea ranges from 60.5-88% (Zhou, Hoover, & Liu, 2004).

1.2.2 Granule Morphology of Pea Starch

Wrinkled pea starch have a mixture of simple and compound granules (Figure 1.2).

The latter of which are composed of 3-10 individual sub-units connected together

(Colonna, Buleon, Lemaguer, & Mercier, 1982). It has been reported that the majority of large granules in wrinkled pea are composed of 4-6 loosely associated pieces in a ring formation with a diameter ranging from 17-30 µm (Eric Bertoft, Manelius, & Qin, 1993).

It was also found that the surface of wrinkled pea starch granules is smooth and exhibits no ruptures or compound granules. The majority of smooth pea starch granules are oval, but there are still some spherical and irregular granules in large groups. Compared with smooth pea, the granules of wrinkled pea starch are smaller. In seven major cultivars from

North Dakota, the length of starch granules ranges from 26-30 µm and the width was around 16-20 µm (Simsek, Tulbek, Yao, & Schatz, 2009). The granules from four cultivars grown in Canada displayed irregular shapes with a diameter ranging from 26.4-30.5 µm

(Raghunathan, Hoover, Waduge, Liu, & Warkentin, 2017). The starch granules from seven varieties of field peas (Yarrum, Maki, Parafield, Kaspa, PRL95, PRL131 and PRL417) 5 were similar in their granular characteristics, such as the mean diameter of the granules ranges between 21.4 and 26.1µm (S. Wang, Sharp, & Copeland, 2011).

Figure 1.2. Scanning electron micrographs of native wrinkled pea starch granules.

1.2.3 Crystallinity and Polymorphic Composition of Pea Starch

It can be observed that the growth rings extend from the helium of starch granules when viewed with a light microscope and sometimes with a scanning electron microscope of broken starch granules. The granular rings are formed by the stacks of amorphous and semi-crystalline backgrounds. Within the semi-crystalline backgrounds are alternating layers of amorphous and crystalline lamella. The crystalline part is formed by the short, external chains of amylopectin, two of which can form a double helix. Wide-angle X-ray scattering indicates that the double helices in amylopectin crystallizes into two types of polymorphs, A and B-type respectively. It has been shown that the double helices of A- type polymorph are arranged as a monoclinic unit cell with 8 water molecules, whereas

6 which of B-type are tightly packed in a hexagonal unit cell carrying 36 water molecules

(Imberty & Perez, 1988; Popov et al., 2009). The starch of cereals commonly exhibits A- type and B-type appears in tubers starch. Except for wrinkled pea starch, most legume starch shows typical C-type diffraction pattern, which is an mixture of A-type and B-type

(Colonna et al., 1982). Based on the X-ray diffraction graph, it is indicated that pea starches have a composite structure of A-type and B-type polymorphs. B-type polymorphs are primarily located at the core, while A-type polymorphs are present at the periphery

(Bogracheva, Morris, Ring, & Hedley, 1998). According to the results revealed by SEM during acid hydrolysis, the amorphous regions of pea starch are mainly in the central region of the granule, surrounded by the crystalline areas (S. Wang, Yu, & Yu, 2008).

1.2.4 Pasting Properties

Pasting properties are generally measured using rapid visco analyzer (RVA) or the

Micro Visco Amylograph, and parameters, such as pasting temperature, peak, breakdown, setback and final viscosities, are recorded. Compared with rice starch and the other pulse starch, pea starch showed lower peak viscosity (de Souza Gomes et al., 2018). The pasting properties of pea starch vary from cultivar and type. Starches from Carneval, Carrera,

Grande and Keoma pea varieties exhibited uniform pasting temperatures and some differences in viscosity at 95 °C (W. Ratnayake, Hoover, Shahidi, Perera, & Jane, 2001).

It has been found that pea starch pasting characteristics are impacted by factors such as crystallinity, amylose leaching, swelling degrees and distribution of amylopectin branch chain length (Hoover & Senanayake, 1996). Low setback and viscosity of pea starches was attributed to shorter amylose chain and a lower percentage of chains at DP 6-12 (W.

7 Ratnayake et al., 2001). Other components like lipids also affect pasting characteristics of pea starch (Simsek et al., 2009).

1.2.5 Thermal Properties

The gelatinization of starch refers to the transition from an ordered to disordered phase of granule structure when starch is heated with excess water. During heating in excess water hydrogen bonds in the starch granules are broken down, more water is absorbed, resulting in swelling of granules, a reduction of crystalline part and amylose leaching (Donovan, 1979).

The gelatinization characteristics of starch are generally conducted by differential scanning calorimetry (DSC). The transition begins from the core area and turns to B- polymorphs located at the center of granules. The central area starts to swell, but the periphery of granules where A-polymorphs are situated are not destroyed until the temperature is increased (W. S. Ratnayake, Hoover, & Warkentin, 2002). A lower enthalpy was displayed in pea starch containing more A-polymorphs, which is loosely packed needing less energy to disrupt. However, pea starch with a higher content of A-polymorphs tended to gelatinize at a higher temperature. It has been reported that the gelatinization parameters can also be affected by amylopectin crystalline structure, in particular short chain in amylopectin (DP 6-12), but not by the percentage of crystalline lamella (Noda,

Takahata, Sato, Ikoma, & Mochida, 1996). Low gelatinization temperatures (To, Tp and

Tc) and enthalpies are due to a higher content of short chains in amylopectin. The increase in enthalpies might be attributed to the loss of double helices instead of the loss of crystallinity (Cooke & Gidley, 1992). However, Tester and Morrison (Tester & Morrison,

1990) supposed that the enthalpies could indicate the overall crystallinity.

8 1.2.6 Swelling Power and Solubility

When starch is treated with heat in the presence of excess water, the hydrogen bonds within the crystalline region are broken down and new hydrogen bonds between water and hydroxyl groups of amylose or amylopectin are formed, resulting in swelling of granules. Swelling power and solubility reflects the extent of interconnection between chains located in the crystalline and amorphous lamella, which can be affected by many factors, such as the amylose content, the arrangement of amylopectin structure and the distribution of amylopectin chains (Sasaki & Matsuki, 1998). It has also been reported that swelling power and solubility would be decreased by amylose-lipid complexes (Hoover &

Hadziyev, 1981).

Field pea starch has been shown a two-stage swelling pattern, which is the typical pattern of legume starch (GUJSKA, D.‐REINHARD, & KHAN, 1994). From 50 to 60

°C starch granules swell smoothly, whereas the swelling power and solubility rise sharply after that. This may be caused the melting of crystalline region, in which the increase in thermal energy, indicated by the gelatinization enthalpies, is required to break down the hydrogen bonds (Hoover & Sosulski, 1985b). Some legume starches possess higher amylose content than non-waxy cereal starch, the stronger swelling power of which might be attributed to the tightly packed amylose chains in the amorphous background (Hoover,

Hughes, Chung, & Liu, 2010). It has been found that the solubility and swelling power of smooth pea starch are much higher than wrinkled one, which can be potentially explained by lower content of amylose-lipid complexes and the dense distribution of amylose chains

(Hoover & Manuel, 1996).

9 1.2.7 Retrogradation

It is well known that gelatinization corresponds to the process where starch crystallites are transformed from an ordered to an amorphous phase when heated in the presence of excess water. The amylose and amylopectin chains in the viscous solution reorganize to form a more ordered structure when cooled down, which is named retrogradation. Retrogradation can be classified as short-term and long-term processes. The increase in crystallinity, syneresis, transformation to B-type diffraction pattern and formation of gel happen during retrogradation. The amylose matrix gel is formed in the prophase and then the short, external chains of amylopectin forms double helices, which crystallize to B-type polymorphs and then the crystallinity is increased (Zobel, 1988). As a result, starch retrogradation affect its structural, mechanical and sensory properties desired in food applications (Karim, Norziah, & Seow, 2000). Starch retrogradation can be determined using methods such as turbidity, differential scanning calorimetry (DSC), wide angle X-ray scattering (WAXS) and nuclear magnetic resonance (NMR). The retrogradation enthalpies of pea starch was higher than that of cereal but lower than potato stored at 6 °C for 4 day, possibly due to differences in the distribution of amylopectin chains (Fredriksson, Silverio, Andersson, Eliasson, & Åman, 1998). Syneresis parameters, reflecting the extent of retrogradation, showed the structure and functionalities of legume starch were susceptible to change during retrogradation due to a higher content of amylose compared to cereal and tuber starch (Aggarwal, Singh, Kamboj, & Brar, 2004; Hoover &

Ratnayake, 2002).

10 1.2.8 Enzymatic hydrolysis

It has been revealed that digestibility of starch can be affected by several factors, including starch source, granule size, crystallinity, the proportion of B-polymorphs, amylose content, interaction between starch segments, amylopectin structure as well as existence of amylose-lipid complexes (Holm et al., 1983; Ring, Gee, Whittam, Orford, &

Johnson, 1988; Sandhu & Lim, 2008; Snow & O'Dea, 1981). Hydrolysis with a-amylase happens mainly in the amorphous backgrounds as reported by Franco, Ciacco, & Tavares

(1988). It was proposed that isolated amylose was first to be attacked and then the amylose chains within the crystalline lamella were hydrolyzed (Eric Bertoft et al., 1993).

Digestibility is found to have negative correlation to granule size when hydrolyzed by a- amylase, which might be due to the lower content of amylose in large granules. Starch granules of A-type diffraction pattern are more sensitive to be attacked by a-amylase than that of B-type due to more branched points in the crystalline regions with short double helical structure (J.-l. Jane, Wong, & McPherson, 1997). Legume starch have higher extent of digestibility than that of potato but lower than cereal starch (Liu, Donner, Yin, Huang,

& Fan, 2006). It was observed that pea starch (67%) was less digestible than maize (100%) and wheat (95%) hydrolyzed for 24h (Ring et al., 1988), which might be due to the higher amylose content of pea starch and rare pores on its granular surface (Hoover & Sosulski,

1985a). Compared with wrinkled pea, smooth pea starch is less susceptible to hydrolysis

(Eric Bertoft et al., 1993).

11 1.2.9 Acid hydrolysis

Acid hydrolysis (2.2M HCl) of starches from various sources all display a two- stage pattern of solubilization. During the first 8 days, hydrolysis tends to be relatively fast occurring in the amorphous backgrounds, which usually followed by a slower hydrolysis degradation of crystalline lamella usually after 8 days. The hydrolysis rate in the first stage might be influenced by the arrangement and chain distributions of amylopectin within the amorphous part, whereas that of the second stage has been shown to have relation with the molar amount of short chains forming double helices (W. Ratnayake et al., 2001). The extent of acid hydrolysis of pea starch on the 8th day was in the range from 26.1 to 32.1%

(Raghunathan et al., 2017; W. Ratnayake et al., 2001), which was similar to that lentil starches (Hoover & Manuel, 1995).

1.3 Starch Modification

It is well known that some starches obtain limited utilization value due to their poor functionalities. Therefore, modification of starch can provide more possibilities to facilitate their applications in the food industry. Starches can be modified by physical, chemical and enzymatic techniques.

1.3.1 Physical Modification

Physical modification is commonly considered as a safe and economic method which doesn’t involve any chemicals. Annealing (ANN) and heat-moisture treatment

(HMT) both undergo hydrothermal process without loss of granule components. When treated at elevated temperature (100-120°C) and restricted moisture content (22-27%) for

16h, the physicochemical behaviors of pea starch were observed to change, including a

12 decrease in swelling power, amylose leaching and peak viscosity (Chung, Liu, & Hoover,

2010). Pre-gelatinization refers to a process that starches are dried after complete gelatinization. Some new methods such as deep freezing and thawing (Szymońska, Krok,

& Tomasik, 2000), osmotic-pressure (Pukkahuta, Shobsngob, & Varavinit, 2007), instantaneous controlled pressure drop (Maache-Rezzoug et al., 2009) and thermal inhabited treatment (Lim, Han, Lim, & BeMiller, 2002) have all been used for physical starch modification.

1.3.2 Chemical Modification

Chemical modification is another treatment involving the addition of functional groups to starches by derivatization, including cationization, crosslinking and hydroxypropylation. The dissociative hydroxyl groups of starch interact with cationic monomers, producing cationic starches. It has been reported that cationization of pea starch results in a decrease in gelatinization and pasting temperature but an increase in peak viscosity, facilitating pea starch application in the paper industry due to improvement in paper strength (Han & Sosulski, 1998). The most commonly used food grade cross-linking agents are sodium tripolyphosphate, sodium trimetaphosphate, a mixture of adipic acid and acetic anhydride, phosphoryl chloride and phosphoryl oxychloride (Wattanachant,

Muhammad, Hashim, & Rahman, 2003; Yook, PEK, & PARK, 1993). Cross-linked pea starches with phosphorus oxychloride was observed to possess a relatively higher setback but a lower viscosity (Jensen, Winters, & Hubbe, 1986). The pea starches are susceptible to retrogradation, which is not desired in the food application. However, cross-linking of hydroxypropyl ester groups to pea starches can significantly increase pasting viscosity and mitigate the extent of syneresis (Hoover, Hannouz, & Sosulski, 1988).

13 1.4 Chemical Gelatinization

Gelatinization parameters of starch can be altered by some concentrated salt solution such as 4 M calcium chloride (CaCl2) and 13 M lithium chloride (LiCl) solutions.

These salt solutions cause starch gelatinization at room temperature through a process called chemical gelatinization (Evans & Haisman, 1982). DSC results showed that starch gelatinization was exothermic in CaCl2 (4M) solution but endothermic in water (Evans &

Haisman, 1982). The types and concentration of salt have various effects on the gelatinization temperature (Gough & Pybus, 1973). Salt solution with high charge density at a low concentration has negative effects on the gelatinization temperature. At a high concentration, the permeation rate of cations into granules is decreased by the high viscosity. Thus, starch can gelatinize immediately at an appropriate concentration. Initial gelatinization occurs at the periphery of starch granules with an increased concentration of salt solution. The interactions between the cations of salts and hydroxyl groups can release the heat needed to melt starch crystalline parts (Jay‐Lin Jane, 1993a). It has been reported that starch granules treated with the potassium thiocyanate (KSCN) and potassium iodide

(KI) solution start gelatinization from the hilum rather than the periphery (Jay‐Lin Jane,

1993b). Onset temperature and enthalpies were observed to increase with sulfate ions but decrease with thiocyanate (Jay‐Lin Jane, 1993b). This can be explained by differences in electrostatic interactions between ions and starch. Ions with large charges tend to have strong interactions with water, which leads to an increase in solution viscosity and stabilization of granule structure (Koch & Jane, 2000).

Chemical surface gelatinization is commonly used to study the internal structure of starch granules. It was observed that normal maize starch granules in 13M LiCl gelatinized

14 more evenly from the periphery than in 4M CaCl2 under light microscope (Pan & Jane,

2000). The analysis of remaining part (inner part) obtained by removing the gelatinized part shows that amylose is more distributed at the periphery rather than the helium of starch granules but the opposite for long B-chains was observed (Pan & Jane, 2000). A study of internal granule structure of various corn starches using different extent of chemical gelatinization showed similar distribution of amylose in HYLON V and HYLON VII corn starch, whereas a significant difference in the distribution of amylopectin long chains between high amylose starch and normal maize starch was observed (Kuakpetoon &

Wang, 2007). Based on the gelatinization characterization of 11 starches from different botanical sources treated with 4M CaCl2, starches were grouped into three categories:

Group 1 starches that can gelatinize evenly on the surface (high-amylose maize, waxy potato, sweet potato and normal potato starch) ; Group 2 starches that gelatinize at certain positions, (wheat and barley starch) ; Group 3 starches with single starch granules aggregating after gelatinization (waxy corn and waxy barley) (Koch & Jane, 2000). In

Group 1, compared with normal potato starch, the polyhedric granules of sweet potato starch show more holes and is easily attacked by salt solution similar to results reported in previous studies on enzymes for hydrolysis (Valetudie, Colonna, Bouchet, & Gallant,

1993). The holes are almost evenly distributed on the surface of round starch granules.

However, for wheat and barley starch in Group 2, their granules tend to obtain equatorial grooves like round lollipop more susceptible to be eroded than other areas of granular surface (Koch & Jane, 2000).

15 1.5 Rationale

This research can provide information on the molecular structure, especially the amylopectin internal structure and physicochemical properties of pea starch. A better understanding of the relationship between structure and properties is vital for improving its applicability in the food industry. Given the particularity of pea starch (C-type crystalline), the study used chemical surface gelatinization as a method of starch modification to separate A and B-type allomorphs in pea starch. It is of great importance to understand the differences in amylopectin chain profiles between A and B-allomorphs in order to support their applications in various food systems.

1.6 Objectives

1. Study the molecular structure, chemical and functional characteristics of pea starches

with significantly different thermal properties.

2. Determine the differences in unit and internal chain profiles between inner part of

granules where B-polymorphs are situated and periphery where A-polymorphs are

located prepared by chemical gelatinization.

Chapter 2. Functional Characteristics of Yellow Pea Starches as Affected by

Amylopectin Unit and Internal Chains

17 2.1 Introduction

Pea (Pisum sativum L.), also named common pea, field pea, garden pea and dry pea is an annual cool season crop. It is ranked forth to soybeans, groundnuts and other pulses in terms of production. Peas are cultivated worldwide, with Canada as the leader in global pea production and exportation (N. Wang & Daun, 2004). Peas are primarily used for livestock feed in North America and Europe, but for human food in Asia. Pea is a rich source of protein, fiber and carbohydrates (N. Wang & Daun, 2004). Its starch and protein contents are about 30-50% and 20-25% on dry matter basis respectively (Shi et al., 2014).

Compared with other crops, pea has higher levels of the essential amino acids lysine and tryptophan and thus believed to be a good alternative protein source to soybean.

Pea starch is the primary by-product of protein isolation. However, due to its inferior functionality after the pea protein extraction process, pea starch is scarcely used for food applications. Studying the structure and physiochemical characteristics of pea starch, particularly their amylopectin structure can provide useful information for further research into improving their functionalities.

Amylopectin chain profiles of starch influence their molecular structure and physiochemical properties (Vermeylen, Goderis, Reynaers, & Delcour, 2004).

Amylopectin, a highly branched starch constituent, consists of linear chains of a-1,4 linked glucose residues connected through α-1,6 linkages. To better understand the structure of amylopectin, there have been models proposed since 1940 (Meyer & Bernfeld, 1940) to explain the organization of chains in the amylopectin molecule. The most recent one proposed in the 2013 is the backbone model (Eric Bertoft, 2013). Based on the backbone model, long chains in the amylopectin molecule form a backbone, along which small and

18 branched units known as building blocks are distributed. It is important to state here that the amylopectin structural analysis conducted in this study was based on the backbone model. The branched building blocks and long chains form the internal structure of amylopectin. It is believed that the organization of the chains in the amylopectin structure, particularly the building blocks will significantly influence the functional properties of starches.

Over the years the amylopectin unit and internal chain profiles of starches from different botanical sources have been investigated and reported (Annor, Marcone, Bertoft,

& Seetharaman, 2014; Eric Bertoft et al., 2008; Gayin, Abdel-Aal, Manful, & Bertoft,

2016). However limited reports on the amylopectin unit and internal chain profiles of yellow pea starches exists. In this study, two commercially available yellow pea starches with significantly different pasting profiles were gifted by Roquette Frères in France. One of the pea starches had lower peak, final and setback viscosities but a higher pasting temperature and was labeled as P1. P2 on the other hand had higher peak, final and setback viscosities but a lower pasting temperature. To understand the differences in the pasting properties of the two yellow pea starches, it was critical to investigate differences in the unit and internal chain profiles of their amylopectins.

This study aimed at investigating the relationship between the pasting properties of yellow pea starches and the unit and internal structure of their amylopectin. The study also determined the physical, molecular and thermal properties of these starches.

19 2.2 Materials and methods

2.2.1 Materials

Two commercially available yellow pea starches with distinct pasting properties were used for this study. One sample had lower peak, final and setback viscosities but a higher pasting temperature (P1). Sample P2 had higher peak, final and setback viscosities but a lower pasting temperature Barley b-amylase (435 U/mg), isoamylase (260 U/mg), and pullulanase M1 (34 U/mg) were purchased from Megazyme International, Bray,

Ireland. Rabbit muscle phosphorylase a (22 U/mg) was from Sigma-Aldrich, St. Louis,

MO, USA.

2.2.2 Granule size distribution

Particle size distribution of the starch granules was carried out using a Laser

Scattering Particle Sizer Distribution Analyzer LA-960 (HORIBA, Kyoto, Japan). Starch powders were dispersed and transported to the optics by the injector-driven forced dispersion system in compressed air at a flow rate of 0.2 m/s and pressure of 0.30 MPa.

The refractive index used was 1.5745.

2.2.3 X-ray diffraction

The starch samples were equilibrated in the desiccator for 36 h before X-ray powder diffraction analysis. X-ray diffraction patterns were obtained with a Bruker D8 Advance

X-ray diffractometer (Bruker AXS, Germany). The parameters were as follows: A Cu Kα radiation source (λ = 1.5418Å) operated at 40 kv and 40 mA. The relative intensities were recorded in a scattering angle range (2θ) of 2.0–40.0º at a scanning speed of 2º min-1. The

20 relative crystallinity can be calculated according to the method described by Nara &

Komiya (Nara & Komiya, 1983).

2.2.4 Amylose content

The amylose content of starch samples was determined by gel-permeation chromatography on a Sepharose CL-6B column (1cm ´ 90 cm) after debranching. The starch sample (4.0 mg) was dissolved in 90% DMSO (100 µL) by heating for 10 min in boiling water bath and stirring constantly with a magnetic stirrer overnight. Hot water (900

µL) was then to the sample added to dilute the solution. After cooling down to room temperature, 100 µL sodium acetate buffer (0.01 M, pH 5.5) was added together with 1 µL isoamylase and 1 µL pullulanase M1 (Megazyme International Ireland). The mixture was stirred overnight at room temperature and then boiled for 5 min to inactivate the enzyme reaction. The solution was finally diluted to a concentration of 2 mg/mL by adding 900 uL double distilled water. 1 mL of sample was injected into the Sepharose CL-6B column and eluted with sodium hydroxide solution (0.5 M) at 1 mL/min. The carbohydrate content of fractions (0.5 mL) was measured by the phenol-sulfuric acid method (Dubois, Gilles,

Hamilton, Rebers, & Smith, 1956).

2.2.5 Amylopectin fractionation

Amylopectin fractionation was carried out according to the description (Genkina,

Wikman, Bertoft, & Yuryev, 2007; Waigh et al., 2000; Zhu, Corke, & Bertoft, 2011) with some minor modifications. The starch sample (2 g) was dissolved in 90% DMSO (40 mL) in boiling water bath for 10 min till a clear solution was obtained and then stirred overnight.

The solution was heated in water bath to 95°C after which it was cooled down to 65°C and

21 four volumes of ethanol was added dropwise to precipitate the starch using a peristaltic pump. The mixture was cooled down to room temperature and then centrifuged at 8,000 ´ g for 10 min at 4°C to separate the supernatant from the precipitated starch. The dissolution in 90% DMSO and precipitation with ethanol was repeated to obtain non-granular starch.

The precipitated starch was dispersed in 90% DMSO (56 mL) and then heated in boiling bath for 3 h under nitrogen. A mixture consisting of isoamyl alcohol (23.5 mL), 1-butanol

(23.5 mL) and water (324 mL) was added slowly to the solution at 85°C. The whole mixture was then put in a water bath (28°C) for 20 h to slowly cool it down, and then centrifuged at 22,000 ´ g for 30 min at 4°C. The volume of the supernatant (the amylopectin portion) was then reduced to about 50 mL by rotary evaporation. The amylopectin was purified a second time by repeating the precipitation of residual amylose with the isoamyl alcohol-1-butanol mixture and then the supernatant was concentrated to about 50 mL again.

Three volumes of methanol were added to the extracted amylopectin and the mixture left overnight at room temperature. The solution was centrifuged at 22,000 ´ g for 20 min at

4°C and the precipitate was dissolved in hot water (20 mL) and re-precipitated with three volumes of ethanol at room temperature for 1 hour. The final precipitate was re-dissolved in hot water (20 mL) and then freeze-dried. The purity of extracted amylopectin was determined on Sepharose CL-6B after debranching with isoamylase and pullulanase M1.

To ensure the complete removal of amylose from the extracted amylopectin, two extra purification steps with butanol and isoamyl alcohol was done.

2.2.6 Production of j,b-limit dextrins of amylopectin

The method of obtaining j,b-limit dextrins was based on a previous description

(Eric Bertoft, 2004) with some modifications as reported by (Kalinga et al., 2013). To 22 produce j,b-limit dextrins, the external chains of amylopectin were removed by the successive treatment of phosphorylase a and β-amylase (Eric Bertoft & Spoof, 1989).

2.2.7 Unit chain length profiles of amylopectin and f,b-limit dextrins

Amylopectin or j,b-limit dextrins (2.0 mg) was dispersed in 90% DMSO (50 µL) and then stirred overnight to be completely dissolved. The solution was diluted with warm water (80°C, 400 µL) and 0.01M sodium acetate buffer (pH 5.5, 50 µL). 1 µL isoamylase and 1 µL pullulanase M1 were added to the mixture. The reaction lasted for 12 hours at room temperature with constant stirring and ended by heating in boiling water bath for 5 mins. The final concentration reached 1 mg/mL by adding 1500 µL water. The sample (25

µL) was injected into high-performance anion-exchange chromatography (HPAEC) after being filtered through 0.45 um nylon mesh.

The Dionex ICS 5000+ high performance anion-exchange chromatography system, obtained from Dionex Corporation (Sunnyvale, CA, USA), consisted of a pulsed amperometric detector (HPAEC-PAD), a gradient pump, and a CarboPac PA-100 ion- exchange column (4 × 250 mm) combined with a similar guard column (4 × 50 mm). The samples were eluted at a flow speed of 1 mL/min with two eluents, 150 mM sodium hydroxide (Eluent A) and 150 mM sodium hydroxide containing 500 mM sodium acetate

(Eluent B), respectively. A gradient was obtained by mixing eluent A and B 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; and 112-130 min, 15% B. The areas under the chromatograms were then adjusted to carbohydrate concentration for quantitative analysis (Koch, Andersson, & Åman, 1998).

23 2.2.8 Iodine binding

Isolated amylopectin samples (2 mg) were dissolved in 90% DMSO (100 µL) by heating in hot water bath for 5min and stirring overnight at room temperature. 1900 µL warm water was then added to obtain a final concentration of 1mg/mL. To adjust to the effective range of absorbance, samples were appropriately diluted. The carbohydrate content of the sample (1 mL) was determined using phenol-sulfuric acid method described by Dubois et al. (Dubois et al., 1956). After added with 0.1mL of 0.01 M I2/0.1 M KI, the lmax of the starch–iodine complex of the samples (1 mL) was measured with spectrophotometer. The iodine affinity was indicated by the peak value, which was obtained as the absorbance at lmax divided by the carbohydrate content.

2.2.9 Thermal properties

Thermal properties of the starch samples were measured using Differential

Scanning Calorimetry (DSC) (Mettler Toledo, Ohio, USA) equipped with thermal analysis data and data recording software (Universal Analysis 2000). was then added to obtain A starch: water ratio of 1:3 (w/w) was obtained by weighing 4 mg starch and adding 12µL deionized water in aluminum pans. The pans were sealed and allowed to equilibrate for 1 hr at room temperature before scanned from 25 to 105 °C at a heating rate of 5 °C/min.

The test used an empty aluminum pan as the reference. The onset (To), peak (Tp), conclusion (Tc) temperatures and the enthalpy change (DH) were recorded. All the measurements were conducted in duplicate.

24 2.2.10 Pasting properties

Micro Visco-Amylo-Graph (Brabender GmbH & Co KG, Duisburg, Germany) was used to determine the pasting properties of starches. Starch slurries with 5% w/w starch

(dry basis) in a total weight of 105 g were heated from 30 to 95 °C at a rate of 5 °C/min, held for 20 min at 95 °C and then cooled to 50 °C at the same rate of heating. Pasting temperature, peak viscosity, final viscosity, breakdown and setback viscosity were reported.

2.2.11 Statistical analysis

All the determinations were made in duplicate. One-way analyses of variance

(ANOVA) by Duncan’s multiple range test was used to determine significant differences between sample means at p < 0.05. All the results were analyzed using Statgraphics

Centurion XVI, version 16.1.0 (Stat Point, Warrenton, VA, U.S.A.).

2.3 Results

2.3.1 Granule size distribution

The granule size distributions of the pea starch granules showed only one peak (Fig.

2.1). The mean diameter of starch granules of P1 (25.44µm) was smaller than that of P2

(26.51µm). The results are within the granule size range of smooth pea (14-32µm) reported by Hoover & Ratnayake (Eric Bertoft, 2004; Laohaphatanaleart, Piyachomkwan, Sriroth,

Santisopasri, & Bertoft, 2009).

25 20

10 olume(%) V

0 0 10 20 30 40 50 60 70 80 90 100 Particle size (µm) P1 P2

Figure 2.1. Particle size distributions of two yellow pea starch samples in the dry state.

2.3.2 Wide angle X-ray diffraction

The yellow pea starches exhibited a typical C-type diffraction pattern. The X-ray pattern (Fig. 2.2) showed a weak peak at 5.6° 2q (characteristic peak for B-allomorph), a single peak at 15.3° 2q and strong peaks at 17.3° and 23.5° 2q (characteristic peak for A- allomorph), suggesting the samples contained both A- and B-type crystalline. The relative crystallinity of P1 and P2 were 29.83% and 37.53%, respectively.

26 P2

Figure 2.2. X-ray diffraction patterns of two yellow pea starch samples.

2.3.3 Amylose content

The amylose content of pea starches was analyzed by gel permeation chromatography with Sepharose CL-6B after debranching (Fig. 2.3). The amylose content of the starches was subdivided into long-chain (LCAM) and short-chain amylose (SCAM).

The relative amounts of amylose, LCAM, SCAM and the ratio LCAM:SCAM is shown in Table

2.1. The amylose content of P1 was 33.27%, which was more than that of P2 (28.27%). P1 had a higher value of LCAM:SCAM due to a higher amount of LCAM.

27 Table 2.1 Amylose Content and Composition of Yellow Pea Starches. P1 P2

Amylose (%) 33.27a 28.27a

a a LCAM:SCAM 1.11 0.80

a a LCAM (%) 17.46 12.49

a a SCAM (%) 15.81 15.88

Values with different letters are significantly different (P < 0.05) within each row. LCAM = long-chain amylose, and SCAM = short-chain amylose.

2.3.4 Unit and internal chain profile

To estimate the efficiency of the amylopectin fractionation process, debranched extracted amylopectin was run on gel permeation chromatography (Sepharose CL-6B) after debranching (Fig. 2.3). The fractionated amylopectin from both samples had about

7% amylose. It is worth noting that repeated purifications of the extracted amylopectin did not reduce their amylose contents from the 7% reported. This lead us to believe that the chains that are eluting in the amylose region of the column may likely be super long chains of amylopectin as reported by (Kong, Corke, & Bertoft, 2009).

28 14 P1

LCAM SCAM AMP 8

6 Carbohydrate()% 4

0 20 40 60 80 100 120 Fraction Number yellow pea starch yellow pea amylopectin

14 P2

LCAM SCAM AMP 8

6 Carbohydrate(%) 4

0 20 40 60 80 100 120 Fraction Number yellow pea starch yellow pea amylopectin

Figure 2.3. Sepharose CL-6B gel-permeation chromatography of debranched yellow pea starches and their extracted amylopectin. LCAM = long-chain amylose; SCAM = short-chain amylose; and AMP = amylopectin chains.

29 The unit chain distributions of the extracted amylopectin determined by HPAEC-

PAD are shown in Fig. 2.4. The two samples had similar profiles. The border between the short and long chains of the samples was at a degree of polymerization (DP) of 38, whereas the peak DP of long chains in P1 was slightly higher than that of P2. The unit chain profiles of the j,b-limit dextrins of the two samples are displayed in Fig. 2.5. The border between short and long B-chains of P1 was higher (DP 26) than that of P2 (DP 25). The peak DP of short and long B-chains of P1 was slightly lower than that of P2.

30 P1

6 13

5 45 4 6 (%) 3 38 eight

W 2

0 0 20 40 60 80 100 120 Degree of Polymerization

6 13 5

4 44

3 6 38 eight eight (%)

W 2

0 0 20 40 60 80 100 120 Degree of Polymerization

Figure 2.4. Unit chain profiles of debranched yellow pea amylopectins by high performance anion-exchange chromatography, with arrows and numbers showing degree of polymerization.

31 P1

5 BS Bfp BL 4 5 9 32 3 26

eight eight (%) 2 W

0 0 20 40 60 80 100 120 Degree of Polymerization

5 BS Bfp BL 4 5 10 34 3 25

eight (%) eight 2 W

0 0 20 40 60 80 100 120 Degree of Polymerization

Figure 2.5. Unit chain profiles of debranched yellow pea amylopectins φ,β-limit dextrins of pea amylopectins. obtained by high-performance anion-exchange chromatography, with arrows and numbers indicating degree of polymerization. Short B-chains (BS) are subdivided into “fingerprint” B-chains (Bfp) and a major group (BSmajor), whereas long B-chains (BL) are subdivided into B2- and B3-chains.

32 The average chain lengths of various chain categories and j,b-limit values are shown in Table 2.2. The average chain length (CL) and short chain length (SCL) of P1 was shorter than that of P2, with no significant difference observed in long chains between two samples. The chain length of j,b-limit dextrins (CLLD) in P1 was shorter than in P2. The j,b-limit value describes the relative proportion of the external segments in amylopectin and is used to estimate the average length of the external chains (ECL) (Annor et al., 2014).

The j,b-limit value of P1 was higher than P2, albeit the ECL values of the two samples were similar. This observation suggested that the thickness of the crystalline lamellae in the granules of the two samples were similar. The internal chain length and the average number of glucosyl units between two branch points, was significantly shorter in P1 than

P2 due to the apparently shorter BS- and BL-chains in the j,b-limit dextrins of P1, suggesting that P1 had a more compactly branched structure in the amorphous lamella.

33 Table 2.2 Average Chain Lengths (CLs) of Different Chain Categories and φ, β-Limit Values of Pea Amylopectin Obtained by High Performance Anion-Exchange Chromatography. P1 P2

CL 18.75a 19.24b

SCL 15.96a 16.21b

LCL 51.42a 51.86a

ECL 12.81a 12.84a

ICL 4.95a 5.40b

TICL 12.29a 12.73b

φ, β-Limit value 60.30b 58.95a

a b CLLD 7.45 7.90

a a BS-CLLD 9.05 8.99

a b BL-CLLD 41.25 41.79

Values with different letters are significantly (P < 0.05) different within each row. SCL = CL of short chains; LCL = CL of long chains; ECL (external CL) = CL × (φ, β-limit value/100) + 1.5; ICL (internal CL) = CL – ECL – 1; TICL (total internal CL) = B-CLLD – 1; j, β -limit value was calculated from the difference in CL of amylopectin and its φ, β- limit dextrin; CLLD = average CL of φ, β-limit dextrin; BS-CLLD = CL of short B-chains; and BL-CLLD = CL of long B-chains.

34 Table 2.3 shows the relative molar amounts of different chain categories. A-chains, the external chains that don't carry any other chains, constituted 51.75 and 49.68% of the chains in P1 and P2, respectively. A-chains can also be grouped into fingerprint A-chains

(Afp : DP 6-8) and crystalline A-chains (Acrystal) (DP>8) (Koch et al., 1998). Acrystal-chains form double helices and contribute to the crystalline structure, whereas Afp-chains are too short to form double helices (Hoover & Ratnayake, 2002). The Acrystal-chains were more abundant in P1 (43.47%) than in P2 (42.29%). The internal structure of amylopectin is revealed by the B-chains (Table 2.3) with DP ≥ 3 in f,b-limit dextrins. P2 had more long

B-chains (BL; 7.26%) than P1 (6.35%). The BL-chains can be divided into B2 (DP 26-50) and B3-chains (DP ≥ 50). The molar amount of B2-chains ranged from 5.23 to 5.89% in pea starches, which was much more than in cereal starches (Takeda, Hizukuri, & Juliano,

1987). The mole percent of B3 chains in the two pea starches were similar to that reported in other C-type starch such as Sago Palm and Kudzu starches (Eric Bertoft et al., 2008).

The molar amount of short B-chains (BS) was significantly more in P2, and this was due to a significantly higher amount of Bfp-chains. It has been reported that Bfp-chains contribute to the formation of tightly branched structures (Gidley & Bulpin, 1987) and

BSmajor-chains connect building blocks to the backbone through short branches (Eric

Bertoft et al., 2008).

35 Table 2.3 Relative Molar Amounts (%) of Chain Categories in Amylopectins of Pea Starches. P1 P2

A-chains 51.75b 49.68a

b a Afp 8.28 7.39

b a Acrystal 43.47 42.29

B-chains 48.25a 50.32b

BS 41.90a 43.05b

BL 6.35a 7.26b

a b Bfp 20.04 21.01

a a BSmajor 21.86 22.04

B2 5.23a 5.89b

B3 1.12a 1.37b

Values with different letters are significantly different (P < 0.05) within each row. DP = degree of polymerization. Detected as maltose after debranching of φ, β-limit dextrins. “Fingerprint” A-chains at DP 6-8 in the original amylopectin sample; Acrystal-chains calculated as all A-chains – Afp; B-chains correspond to DP > 3 in φ, β-limit dextrins and were divided into short (BS) and long (BL) chains at between DP 22–25 and above, respectively, depending on the sample; “Fingerprint” B-chains at DP 3–7 in φ, β -limit dextrins; The major group of short B-chains at DP 8 to 22–24, depending on the sample; Long chains between DP 22 or 25 and 55, depending on the sample; Long chains at DP > 56.

36 Some molar ratios of different chain categories in amylopectin and its j,b-limit dextrin are presented in Table 2.4. The molar ratio of A to B-chains of the pea amylopectins was close to one, which represented the similar proportion of Staudinger and Haworth conformations in internal structure of amylopectin (Haworth, Hirst, & Isherwood, 1937;

Staudinger & Husemann, 1937). The molar ratio of Sap to Lap for P1 was significantly higher than P2 and the ratio BS:BL of P1 was also much higher than P2. The ratio Acrystal:

BS of two samples were both close to 1. P1 had much higher ratio of Acrystal: BS as well as

Acrystal: B, whereas the ratio Bfp:BSmajor was much higher in P2. Amylopectin of various samples has been divided into four structural types relying on their internal structure (Eric

Bertoft et al., 2008). Sap:Lap of P1 was 14.75, which was in the range of group 1 amylopectins. But P1 had a BS:BL value of 6.60 and a Bfp:BSmajor value of 0.92, which typically fitted in group 2 amylopectins. P2 can also be classified as group 2 according to its molar ratios. One exception was that the ratio Acrystal: B (0.84) was extremely low in P2, even below the lower limit of group 1.

37 Table 2.4 Selected Molar Ratios of Different Chain Categories of Pea Amylopectins and Their j, β-Limit Dextrins. P1 P2

A:B 1.07b 0.99a

b a Sap:Lap 14.75 12.77

BS:BL 6.60b 5.93a

b a Acrystal:BS 1.04 0.98

b a Acrystal:B 0.90 0.84

a b Bfp:BSmajor 0.92 0.95

Values with different letters are significantly different (P < 0.05) within each row. A = A- chains; B = B-chains; Sap = short amylopectin chains; Lap = long amylopectin chains; BSmajor = the major group of short B-chains at DP 8 to 22–24, depending on the sample; and Bfp = “fingerprint” B-chains at DP 3–7 in φ, β-limit dextrins. Acrystal-chains were calculated as all A-chains – Afp. B-chains correspond to DP > 3 in φ, β-limit dextrins and were divided into short (BS) and long (BL) chains at between DP 22–25 and above, respectively, depending on the sample.

38 2.3.5 Iodine affinity

Helical complexes can be formed by iodine and long chains of amylose, which leads to a blue color absorbing maximum light at lmax > 600 nm (Banks, Greenwood, &

Khan, 1971). Iodine turns amylopectin purple-blue with maximum absorption at a slower wavelength (lmax < 600 nm) because of the shorter chains within amylopectin (Bailey &

Whelan, 1961). The lmax of the two samples of pea amylopectin was 554 and 558 nm respectively (Table 2.5). The peak value (PV) indicated the amount of iodine binding with amylopectin, which for P2 (0.508) was more than that of P1 (0.433).

Table 2.5 Iodine affinity values of two yellow pea starch samples. P1 P2

a a λmax of amylopectin 554 558

Peak value 0.433a 0.508a

Values with different letters are significantly different (P < 0.05) within each row.

39 2.3.6 Pasting properties

The pasting properties of pea starches are shown in Table 2.6 and Appendix A Fig.

6.1. The overall shape of MVAG graphs of the two samples was similar. As seen in Fig.

6.1, it was nearly flat when held at 95 °C. The pasting profiles both exhibited no remarkable pasting peak, a low breakdown viscosity and a high setback. However, there were some significant differences in pasting parameters between the two samples. The pasting temperature of P1 (74.4 ℃) was higher than P2 (69.5 ℃), but the peak, final and setback viscosity were lower in P1.

Table 2.6 Pasting properties of two yellow pea starch samples. P1 P2

Pasting Temp (℃) 74.4a 69.5b

Peak Viscosity (BU) 53.5a 75.5b

Breakdown Viscosity (BU) 0.5a 0.0a

Setback Viscosity (BU) 40.0a 57.0b

Final Viscosity (BU) 93.0a 132.5b

Values with different letters are significantly different (P < 0.05) within each row. BU: Brabender Units.

40 2.3.7 Thermal properties

The gelatinization temperature (To: onset temperature, Tp: peak temperature, Tc: conclusion temperature) and enthalpies (DH) are displayed in Table 2.7. There was significant difference in thermal properties between the two samples (Appendix B Fig.

6.2). The onset, peak and conclusion temperature was significantly higher for P1 than for

P2, whereas the enthalpy of P1 (4.32 J/g) was much lower than that of P2 (7.75 J/g). The gelatinization temperature range (Tc-To) of P1 was broader than that of P2. The results were within the range documented before for pea starch (E Bertoft & Koch, 2000).

Table 2.7 Thermal properties of two yellow pea starch samples. P1 P2

b a To (°C) 60.0 54.3

b a Tp (°C) 67.7 60.3

b a Tc (°C) 73.2 67.1

ΔH (J/g) 4.3a 7.8b

Values with different letters are significantly different (P < 0.05) within each row. To: onset temperature, Tp: peak temperature, Tc: conclusion temperature, ΔH: enthalpy of gelatinization.

41 2.4 Discussion

The molar ratios of various chain categories can show some features of amylopectin internal structure. Given the assumption that the building blocks were formed by short chains and interconnected by long chains, the ratio of Sap:Lap could potentially reflect the number of chains per building block. P1 had more chains per building block than P2. The ratio of BS:BL indicates how the building blocks are arranged on the backbone of amylopectin since the backbone is mainly composed of BL-chains according to the backbone model (Eric Bertoft, Koch, & Åman, 2012). More building blocks were distributed on the backbone in P1. Imberty and Pérez (Imberty & Pérez, 1989) hypothesized that one A-chain unites with one B-chains to form a double helix, the ratio

Acrystal:B (0.90 and 0.84 for P1 and P2 respectively) would be 1 if all the B-chains were involved in the formation of crystalline lamella (W. S. Ratnayake et al., 2002). Another assumption suggested that only BS-chains participated in the formation of double helices

(Eric Bertoft, 2013), so Acrystal:BS should be close to 1 (1.04 and 0.98 for P1 and P2 respectively). The ratio Bfp:BSmajor might show the structural arrangement of building blocks, but more details still await to be explored.

It has been found that iodine forms inclusion complexes with internal chain segments of amylopectin (Imberty & Pérez, 1989), which shows that these segments can form helical structures with iodine. The PV values of the two pea starch samples (Table

2.5) were considerably higher compared to that of waxy rice starches previously reported

(Eric Bertoft et al., 2016), suggesting that more iodine was attached to the internal segments of pea amylopectin than to rice amylopectin. Indeed, the higher iodine affinity of P2 was in accordance with the longer internal chain segments in P2 compared to P1.

42 The higher pasting temperature of P1 (Table 2.6) may be due to its higher content of amylose (Laohaphatanaleart, Piyachomkwan, Sriroth, & Bertoft, 2010). The results were consistent with the gelatinization temperature results of DSC. The starch (P2) with longer chain length (CL) and more B3-chains showed higher setback viscosity, which was accordant with a previous study on amaranth starch (Kong, Bertoft, Bao, & Corke, 2008).

It was observed that P1 with higher content of Afp had lower peak viscosity. This result was the opposite of what was reported for sweet potato starches (Zhu et al., 2011).

According Kozlov et al (2006), Afp causes structural defects in starch granules and thus facilitating the penetration of water molecules resulting in a higher peak viscosity. But in this case, it may be due to the fact that pasting behaviors of starches are also impacted by factors such as crystallinity, amylose leaching and swelling degree of starch granules (Eric

Bertoft, 2004). P2 with longer internal chain lengths (ICL) had higher pasting viscosity, adding more evidence to previous studies that pasting properties might be influenced by the amorphous lamella of starch granules (Hoover & Senanayake, 1996). This might be due to more water absorbed by the amorphous region.

Gelatinization characteristics as measured by DSC has been shown to be mostly influenced by amylopectin short chains (Kong et al., 2008; Zhu et al., 2011). P1 with more

Acrystal displayed higher gelatinization temperature, because Acrystal-chains were involved in the formation of crystalline structure in the starch granules (Annor et al., 2014). P2 with more BL-chains and lower ratio of Sap:Lap had lower peak temperature (Tp). This trend agreed with the previous study on rice starch (Okuda, Aramaki, Koseki, Satoh, &

Hashizume, 2005). The BL-chains act as the backbones which might promote the process of gelatinization in a side-chain liquid-crystalline polymeric model for starch reported by

43 Waigh et al. (2000). It was observed that the starch (P2) with longer ICL exhibited higher

DH, which was consistent with previous study conducted by Kong et al. (2008). DH has been shown to be influenced by other factors such as the crystallinity and the proportion of

B-type allomorph (Zhu et al., 2011).

2.5 Conclusion

The study showed that the two yellow pea starch samples were significantly different in their physiochemical properties, and amylopectin chain profiles. It was observed that variation in functionalities of the pea starches were not only affected by their amylose profiles and crystallinity, but also their amylopectin structures. The presence of more Acrystal in the samples contributed to higher gelatinization temperature. Peak viscosity and enthalpy of gelatinization of the samples was significantly affected by internal segments of amylopectin.

Chapter 3: Amylopectin Unit and Internal Chain Profiles of Periphery and Core of

Pea Starch Granules as Revealed by Chemical Gelatinization

45 3.1 Introduction:

Starch is produced as storage carbohydrate in plants and serves as the primary energy supply for human beings. There are two components in starch; amylose and amylopectin. Amylose, the minor component, consists mainly of a-1,4 linked glucose residues. Amylopectin is composed of linear chains of a-1,4 linked glucose residues connected through a-1,6 linkages.

A cross section of starch granules shows growth rings consisting of alternating amorphous and semi-crystalline backgrounds. Within the semi-crystalline backgrounds are stacks of amorphous and crystalline lamella. The highly branched molecules of amylopectin constitute the semi-crystalline structure. It was initially considered that amylose was more likely distributed in the amorphous background without association with amylopectin (Nikuni, 1969). Recent reports have shown that amylose molecules are randomly dispersed in the amylopectin matrix (J. Jane, Xu, Radosavljevic, & Seib, 1992).

The chains of amylopectin have been divided into external and internal chains, the former of which form double helical structures in the crystalline lamella of amylopectin, whereas internal chains are mainly found in the amorphous lamella (Pérez & Bertoft,

2010). The unit chains of amylopectin can also be categorized into A- and B-chains. A- chains are defined as chains that don’t carry other chains and B-chains carry other chains

(Peat et al., 1952). All the A-chains are external, but B-chains contain both external and internal segments. The internal segments of amylopectin can be separated through the removal of external chains using exo-acting enzymes, such as phosphorylase a and b- amylase. b-amylase produces maltose unit from non-reducing end and phosphorylase a can

46 remove glucose unit. j,b-limit dextrins are produced by the successive use of b-amylase and phosphorylase a, in which A-chains appear as maltosyl stubs whereas the external segment of B-chains remains as glucosyl stubs. j,b-limit dextrins after being debranched are analyzed by high performance anion exchange chromatography to determine the internal chain profiles of amylopectin. The internal chains are grouped into short (BS) and long B-chains (BL). According to the building block backbone model proposed by Eric

Bertoft (2013), long chains of amylopectin serve as the backbone in the amorphous lamella where building blocks (small, branched units) are distributed.

Starches can be chemically gelatinized with certain salts at room temperature, usually starting at the periphery of the granules. Controlled chemical starch gelatinization can be employed to investigate properties at the core or periphery of starch granules. The core structure of potato and maize starches has been investigated after treatment with 4M

CaCl2 or 13M LiCl solution (J.-l. Jane & Shen, 1993; Pan & Jane, 2000). They observed amylose was more concentrated at the periphery, whereas long B-chains were more distributed at the core area in potato and maize starch granules. There is little or no information on the amylopectin chain profile of the core and periphery of pea starches. Pea starch shows typical C-type diffraction pattern, which is a composite structure of A-type and B-type polymorphs. It has been reported that A-polymorphs are located at the periphery of the granules with B-polymorphs situated at the central part (Bogracheva et al.,

1998). Therefore, chemical surface gelatinization can be potentially used to separate A and

B-type polymorphs in pea starches and their characteristics investigated.

47 The objective of this study is to investigate the differences in unit and internal chain profiles of amylopectin of large pea starches chemically gelatinized with 13M LiCl and separated into its core and peripheral portions.

3.2 Materials and methods:

3.2.1 Materials

Ireland. Rabbit muscle phosphorylase a (22 U/mg) was from Sigma-Aldrich, St. Louis,

MO, USA.

3.2.2 Fractionation of starch granules

A nylon mesh with a size of 20 µm was used to separate of pea starch into large and small granules in water with large granules being >20 µm. After separation, granules precipitated with ethanol and acetone, and dried at room temperature overnight.

3.2.3 Chemical surface gelatinization

The chemical gelatinization was done according to the method described by Pan &

Jane (2000). Large granules (10 g) were treated with 13M LiCl solution (75 mL) and stirred at room temperature for various lengths of time. A desired extent of gelatinization (50%) was determined by observing granules under normal and polarized light microscope. Cold

48 water (600 mL, 4 °C) was added to stop the chemical gelatinization process. The mixture was centrifuged at 3,200 ´ g for 15min and then washed twice with 900 mL of distilled water.

3.2.4 Separation of the gelatinized starch from the remaining granules

The partially gelatinized starch granules (ca. 10 g) were resuspended in distilled water (60 mL) and the portion of gelatinized starch was separated from the ungelatinized starch granules using a blender (Hamilton Beach, Inc.) for 10 min and then centrifuged at

3,200 ´ g for 15min. The separation process was repeated each time with 60 mL of distilled water until the supernatant was clear. The precipitate (RG) was washed with ethanol and acetone and dried at room temperature overnight. The supernatant containing the gelatinized starch was concentrated to 100 mL by rotary evaporation at 40 °C. Two volumes of 95% ethanol was added to precipitate the gelatinized starch (GP). The precipitate was then washed with pure ethanol and acetone, and dried at room temperature overnight.

The extent of surface gelatinization was calculated as:

gelatinized starch (g) % of surface gelatinization = gelatinized starch g + remaining granules (g)

3.2.5 Amylose content

49 boiling water bath and stirring constantly with a magnetic stirrer overnight. Hot water (900

3.2.6 Amylopectin fractionation

Amylopectin fractionation was carried out according to the description (Genkina et al., 2007; Waigh et al., 2000; Zhu et al., 2011) with some minor modifications. The starch sample (2 g) was dissolved in 90% DMSO (40 mL) in boiling water bath for 10 min till a clear solution was obtained and then stirred overnight. The solution was heated in water bath to 95°C after which it was cooled down to 65°C and four volumes of ethanol was added dropwise to precipitate the starch using a peristaltic pump. The mixture was cooled down to room temperature and then centrifuged at 8,000 ´ g for 10 min at 4°C to separate the supernatant from the precipitated starch. The dissolution in 90% DMSO and precipitation with ethanol was repeated to obtain non-granular starch. The precipitated starch was dispersed in 90% DMSO (56 mL) and then heated in boiling bath for 3 h under nitrogen. A mixture consisting of isoamyl alcohol (23.5 mL), 1-butanol (23.5 mL) and water (324 mL) was added slowly to the solution at 85°C. The whole mixture was then put in a water bath (28°C) for 20 h to slowly cool it down, and then centrifuged at 22,000 ´ g

50 for 30 min at 4°C. The volume of the supernatant (the amylopectin portion) was then reduced to about 50 mL by rotary evaporation. The amylopectin was purified a second time by repeating the precipitation of residual amylose with the isoamyl alcohol-1-butanol mixture and then the supernatant was concentrated to about 50 mL again. Three volumes of methanol were added to the extracted amylopectin and the mixture left overnight at room temperature. The solution was centrifuged at 22,000 ´ g for 20 min at 4°C and the precipitate was dissolved in hot water (20 mL) and re-precipitated with three volumes of ethanol at room temperature for 1 hour. The final precipitate was re-dissolved in hot water

(20 mL) and then freeze-dried. The purity of extracted amylopectin was determined on

Sepharose CL-6B after debranching with isoamylase and pullulanase M1. To ensure the complete removal of amylose from the extracted amylopectin, two extra purification steps with butanol and isoamyl alcohol was done.

3.2.7 Production of j,b-limit dextrins

The method of obtaining j,b-limit dextrins was based on a previous description

(Eric Bertoft, 2004) with some modifications as reported by (Kalinga et al., 2013). To produce j,b-limit dextrins, the external chains of amylopectin were removed by the successive treatment of phosphorylase a and β-amylase (Eric Bertoft & Spoof, 1989).

3.2.8 Unit and internal chain profiles

51 room temperature with constant stirring and ended by heating in boiling water bath for 5 mins. The final concentration reached 1 mg/ml by adding 1500 µL water. The sample (25

µL) was injected into high-performance anion-exchange chromatography (HPAEC) after being filtered through 0.45 um nylon mesh.

The Dionex ICS 5000+ high performance anion-exchange chromatography system, obtained from Dionex Corporation (Sunnyvale, CA, USA), consisted of a pulsed amperometric detector (HPAEC-PAD), a gradient pump, and a CarboPac PA-100 ion- exchange column (4 × 250 mm) combined with a similar guard column (4 × 50 mm). The samples were eluted at a flow speed of 1 mL/min with two eluents, 150 mM sodium hydroxide (eluent A) and 150 mM sodium hydroxide containing 500 mM sodium acetate

(eluent B), respectively. A gradient was obtained by mixing eluent A and B 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; and 112-130 min, 15% B. The areas under the chromatograms were then adjusted to carbohydrate concentration for quantitative analysis (Koch et al., 1998).

3.2.9 Statistical analysis

All the determinations were made in duplicate. One-way analyses of variance

(ANOVA) by Duncan’s multiple range test was used to determine significant differences between sample means at p < 0.05. All the results were analyzed using Statgraphics

Centurion XVI, version 16.1.0 (Stat Point, Warrenton, VA, U.S.A.).

52 3.3 Results:

3.3.1 Comparison of large granules and original granules

3.3.1.1 Amylose content

In order to achieve more uniform chemical gelatinization, starch granules were fractionated into large granules (LG) and small granules. The yield of LG was about 80-

90% of the native starch for both samples. The amylose content and composition of LG are shown in Table 3.1. There was no significant difference in amylose content between LG

(33.61%) and its native starch (33.27%) of P1. However, amylose was more concentrated in LG of P2. The amylose content of LG between two samples were similar.

3.3.1.2 Unit and internal chain profiles of amylopectin

The purity of fractionated amylopectin was determined by gel permeation chromatography (Sepharose CL-6B) after debranching (Fig. 3.2). The amylose content of the LG amylopectin of both samples reached up to 10% even after 2 extra purification steps. This results can be potentially explained by fact that the chains that are eluting in the amylose region of the column may likely be super long chains of amylopectin as reported by Kong et al. (2009).

Table 3.2 reveals the average chain lengths of different chain categories and j,b- limit value of amylopectin of LG. The chains of amylopectin can be grouped into short and long chains at DP 36 (Eric Bertoft, 2004). In P1, LG possessed longer chain length (CL) compared to its native starch due to longer short chain length (SCL) and long chain length

(LCL). However, CL and SCL were longer in original samples than in LG of P2. The j,b- limit value used to describe the relative percentage of external segment of amylopectin was

53 higher in native starches than in LG for both pea starches, which was accordant with the results of external chain length (ECL). The internal chain lengths (ICL) of the two native starches were both shorter than the corresponding LG, showing that the native starches were more compactly branched structures. This can be potentially explained by the shorter

CL of BS- and BL-chains in the j,b-limit dextrins of native starches.

The relative molar amounts of chain categories in amylopectin of the LG are displayed in Table 3.3. A-chains, which are external chains that don’t carry any other chains, can be divided into fingerprint (Afp) and crystal A-chains (Acrystal). It has been reported that Afp-chains (DP 6-8) are too short to be involved in the formation of double helices but Acrystal-chains (DP > 8) mainly contribute to the structure of crystallite (Gidley

& Bulpin, 1987). The native starches had more of Afp-chains than the LG in P1, but LG had a more Afp-chains than the native starches in P2. B-chains, DP > 3 in j,b-limit dextrins, can be grouped into BS (DP 3-25) and BL (DP > 25) chains. In P1, there was no significant difference in B- and BS-chains between native and LG, whereas BL-chains were more in the LG due to higher amounts of B2 and B3. The native starch of P2 contained more B- and BS-chains than their LG, but the amount of BL-chains estimated as the differences between B- and BS-chains were similar. BS-chains profiles are composed of fingerprint group (Bfp, DP 3-7) and major group (BSmajor, DP 8-25) of BS-chains. The native starches had more Bfp-chains than the LG for both samples. It has been reported that the Bfp chains are found in the building blocks of amylopectin (E Bertoft & Koch, 2000). A significantly higher amount of BSmajor-chains was observed in the LG of P1 but there was no significant difference between the native starch and the LG in P2.

54 Based on selected molar ratios of the different chain categories (Table 3.4), amylopectin of different samples can be divided into four structural types (Eric Bertoft et al., 2008). Amylopectin from the LG in both samples can also be considered as Group 2.

Though the ratio Acrystal:B of both large groups was slightly lower than the value of type 2, the ratio S:L of the large granules amylopectin in both samples ranged from 12.06 to 12.62, which fits into the range of type 2 amylopectins. Also, the ratio BS:BL, Acrystal:BS and

Bfp:BSmajor were typically within the range reported for type 2 amylopectins.

One of the most significant differences in LG between P1 and P2 was the average

CL. The CL of the LG in P1 was longer than P2 due to longer SCL and LCL. Longer ECL was also shown in the LG of P1. The LG of P2 had more Afp-chains than that of P1.

3.3.2 Extent of gelatinization

The large granules of two yellow pea starch were treated with 13M LiCl solutions at room temperature. During the gelatinization, it was observed that a layer started to appear from the periphery of granules and expanded gradually (Fig. 3.1). The gelatinization process was checked under light microscopy to estimate whether the extent reach 40%-

60%. P1 obtained 46.03% of chemical surface gelatinization treated with 300 min, but it only took 190 min for P2 to reach 45.03%. The differences in gelatinization time might be caused by distributions of amylose and amylopectin chains.

A (60 min) D (60 min)

B (180 min) E (120 min)

C (300 min) F (190 min) Figure 3.1. Polarized light-micrographs of large granules of yellow pea starch in 13M LiCl solutions at different times. A: P1, treated with 60 min; B: P1, treated with 180 min; C: P1, treated with 300 min; D: P2, treated with 60 min; E: P2, treated with 120 min; F: P2, treated with 190 min.

56 3.3.3 Amylose distribution of pea starch granules

The chain profiles of the debranched LG are shown in Fig. 3.2. Table 3.1 also shows the amylose content of GP and RG of samples. There was no significant difference in amylose content between GP and RG in P1, whereas the amylose content of the GP

(35.78%) was significantly higher than that of RG (32.01%) in P2. More long chains of amylose (LCAM) and a higher ratio of LCAM: SCAM were found in the GP than the RG for both samples. The RG had more short chains of amylose (SCAM) than the GP in P1, but the amount of SCAM between GP and RG were similar in P2.

Table 3.1 Amylose content of Large, Gelatinized and Remaining Granules. P1 P2

LG GP RG LG GP RG

Amylose (%) 33.61 32.32A 32.50 34.06b 35.78bB 32.01a

a b aA a c bB LCAM:SCAM 1.30 1.74 1.26 1.27 1.69 1.54

a bA aA a bB aB LCAM (%) 18.99 20.53 18.09 19.08 22.46 19.41

b aA bB b aB aA SCAM (%) 14.62 11.79 14.41 14.98 13.32 12.60

Values with different superscript lowercase letters are significantly (P < 0.05) different within each sample. Values with different superscript uppercase letters are significantly (P < 0.05) different between samples. LCAM = long-chain amylose, and SCAM = short-chain amylose. LG: Large granules; GP: Gelatinized portion; RG: Remaining granules.

57 3.3.4 Distributions of unit and internal chain profiles of pea starch amylopectin

The amylose content of the extracted fraction of GP and RG from both samples was similar with that of LG up to 10% even after further purifications (Fig. 3.2). The high content of amylose in the extracted amylopectin might be due to the presence of long amylopectin chains (Kong et al., 2009).

8 LCAM SCAM AMP

6 arbohydrate (%)

C 4

0 20 30 40 50 60 70 80 90 100 110 120 Fraction Number

LG of P1 LG of P2 LG amylopectin of P1 LG amylopectin of P2

58 14

LCAM SCAM AMP 6 arbohydrate (%)

C 4

0 20 30 40 50 60 70 80 90 100 110 120 Fraction Number

GP of P1 GP of P2 GP amylopectin of P1 GP amylopectin of P2

LC 8 AM SCAM AMP

6 arbohydrate (%)

C 4

0 20 30 40 50 60 70 80 90 100 110 120 Fraction Number

RG of P1 RG of P2 RG amylopectin of P1 RG amylopectin of P2

Figure 3.2. Sepharose CL-6B gel-permeation chromatography of debranched fractions of pea starches and their amylopectins (A: Large granules (LG); B: Gelatinized portion (GP); C: Remaining granules (RG)). LCAM = long-chain amylose; SCAM = short-chain amylose; and AMP = amylopectin chains. 59 Both samples obtained similar unit profiles of amylopectin between the core and periphery of starch granules (Fig. 3.3). The border between the short and long chains of the samples were all at DP 38, whereas the DP of long chains peak of RG in P1 was slightly lower than the others. Similar unit chain profiles of j,b-limit dextrins within the starch granules of two samples are shown in Fig 3.4. The border between short and long B-chains of GP and RG are both at DP 26 in P1, but at 25 in P2. The peak-DP of long B-chains of

GP was slightly higher than RG in P1 and was the opposite in P2.

60 LG of P1

6 14

4 44 6 3 38 eight eight (%)

W 2

0 0 20 40 60 80 100 120 Degree of Polymerization

GP of P1

6 14

4 43 6 3 38 eight eight (%)

W 2

0 0 20 40 60 80 100 120 Degree of Polymerization

61 RG of P1

6 14

5 42 4 6 3 38 eight eight (%)

W 2

0 0 20 40 60 80 100 120 Degree of Polymerization

LG of P2 6 14

5 43 4 6 38 3 eight eight (%)

W 2

0 0 20 40 60 80 100 120 Degree of Polymerization

62 GP of P2 6 14

5 43 4 6 38 3 eight eight (%)

W 2

0 0 20 40 60 80 100 120 Degree of Polymerization

RG of P2 6 14

5 43 4 38 6 3 eight eight (%)

W 2

0 0 20 40 60 80 100 120 Degree of Polymerization

Figure 3.3. Unit chain profiles of debranched amylopectins of pea starches fractions obtained by high performance anion-exchange chromatography, with arrows and numbers showing degree of polymerization. LG: Large granules; GP: Gelatinized portion; RG: Remaining granules.

63 LG of P1

5 BS Bfp BL 4 5 9 32 3 25 eight eight (%) 2 W

0 0 20 40 60 80 100 120 Degree of Polymerization

GP of P1 BS 5 Bfp BL 4 5 9 33 3 26

eight (%) eight 2 W

0 0 20 40 60 80 100 120 Degree of Polymerization

64 RG of P1 BS 5 Bfp BL 4 5 9 32 3 26

eight eight (%) 2 W

0 0 20 40 60 80 100 120 Degree of Polymerization

LG of P2 BS 5 Bfp BL 4 5 9 32 3 26

eight eight (%) 2 W

0 0 20 40 60 80 100 120 Degree of Polymerization

65 GP of P2 BS 5 Bfp BL 4 5 9 31 3 25

eight eight (%) 2 W

0 0 20 40 60 80 100 120 Degree of Polymerization

RG of P2 BS 5 Bfp BL 4 5 9 33 3 25

eight eight (%) 2 W

0 0 20 40 60 80 100 120 Degree of Polymerization

Figure 3.4. Chain profiles of φ,β-limit dextrins of amylopectins of pea starches fractions obtained by high performance anion-exchange chromatography, with arrows and numbers indicating degree of polymerization. Short B-chains (BS) are subdivided into “fingerprint” B-chains (Bfp) and a major group (BSmajor), whereas long B-chains (BL) are subdivided into B2- and B3-chains. LG: Large granules; GP: Gelatinized portion; RG: Remaining granules.

66 The average chain lengths of different chain categories and j,b-limit value in amylopectin of pea starches after chemical gelatinization is displayed in Table 3.2. RG

(16.49) contained longer SCL than GP (16.40) and the large granules (16.41) in P1, but there was no statistical difference in their CL and LCL. The total internal chain lengths

(TICL) of RG (14.10) were significantly longer than that of GP (13.51) in P1, which seemed to be the results of longer BL-chains in the j,b-limit dextrins of RG. In P2, the

SCL were shorter in GP than RG of P2, whereas the LCL were shorter in RG. When comparing the GP and RG between two samples, CL, SCL and ECL of P1 were significantly longer than P2. The trend was similar to the large granules of two samples.

67 Table 3.2 Average Chain Lengths (CLs) of Different Chain Categories and φ,β-Limit Values of Large Granules and Their Gelatinized and Remaining Fraction Amylopectins. P1 P2

LG GP RG LG GP RG

CL 19.57B 19.59B 19.74B 19.01A 19.05A 18.97A

SCL 16.41aB 16.40aB 16.49bB 16.16bA 16.12aA 16.20cA

LCL 52.40B 52.48 52.54B 51.57bA 52.02b 50.89aA

ECL 12.79B 12.91B 13.07B 12.44A 12.32A 12.31A

ICL 5.78 5.68 5.67 5.57 5.73 5.66

TICL 13.98bB 13.51a 14.10bB 13.67A 13.49 13.59A

φ,β-Limit value 57.70 58.23B 58.60B 57.54 56.80A 56.98A

CLLD 8.28 8.18 8.17 8.07 8.23 8.16

aB b ab aA b b BS-CLLD 9.73 9.84 9.78 9.64 9.80 9.79

b a bB b a aA BL-CLLD 42.91 40.81 42.80 42.41 41.13 41.17

Values with different superscript lowercase letters are significantly (P < 0.05) different within each sample. Values with different superscript uppercase letters are significantly (P < 0.05) different between samples. SCL = CL of short chains; LCL = CL of long chains; ECL (external CL) = CL × (φ, β-limit value/100) + 1.5; ICL (internal CL) = CL – ECL – 1; TICL (total internal CL) = B-CLLD – 1; φ, β -limit value was calculated from the difference in CL of amylopectin and its j, β-limit dextrin; CLLD = average CL of φ, β-limit dextrin; BS-CLLD = CL of short B-chains; and BL-CLLD = CL of long B-chains. LG: Large granules; GP: Gelatinized portion; RG: Remaining granules.

68 Table 3.3 shows the relative molar amounts of chain categories. One of the most significant differences was the higher content of A-chains in RG of P1 due to more of

Acrystal-chains, but GP (7.03%) had more Afp-chains than RG (6.79%). B-chains were more abundant in GP than in RG of P1, which can be explained by higher content of BS-chains in GP. The GP of P1 starch granules had a higher content of Bfp-chains. BSmajor chains were more abundant in GP (24.59%) than in RG (22.76%) (Eric Bertoft et al., 2012). However,

B3 chains (DP>50) of RG were much more than that of GP in P1. There was no significant difference in the unit chain profiles of amylopectin between the GP and RG in P2, except for Afp-chains. The amount of Afp in GP and RG of P2 was more than that of P1, which was consistent with the results in large granules of two samples. There was significant difference in the relative molar amount of other chain categories in the RG between two samples. The A-chains of RG in P1 was more than that in P2 due to a higher content of

Acrystal chains. The higher amount of Bfp and BSmajor-chains led to more of BS-chains of RG of P2. More B-chains were shown in RG of P2 due to its more BS-chains.

69 Table 3.3 Relative Molar Amounts (%) of Chain Categories of Large Granules and Their Gelatinized and Remaining Fraction Amylopectins. P1 P2

LG GP RG LG GP RG

A-chains 51.64ab 50.57a 52.87bB 52.08b 50.11a 51.05abA

bA bA aA bB cB aB Afp 6.95 7.03 6.79 7.83 8.04 7.48

ab a bB b a abA Acrystal 44.69 43.54 46.08 44.25 42.06 43.57

B-chains 48.36ab 49.43b 47.13aA 47.92a 49.89b 48.95abB

BS 40.70ab 41.97b 39.54aA 40.58a 42.43b 41.47abB

BL 7.66 7.46 7.59 7.34 7.47 7.48

b b aA B Bfp 17.53 17.38 16.78 17.77 17.86 17.37

a b aA a b bB BSmajor 23.18 24.59 22.76 22.80 24.57 24.10

B2 6.09 6.18 6.03 5.88a 6.17b 6.16b

B3 1.57bA 1.28a 1.56bB 1.46bB 1.30a 1.32aA

Values with different lowercase letters are significantly (P < 0.05) different within each sample. Values with different uppercase letters are significantly (P < 0.05) different between samples. DP = degree of polymerization. Detected as maltose after debranching of φ, β-limit dextrins. “Fingerprint” A-chains at DP 6-8 in the original amylopectin sample; Acrystal-chains calculated as all A-chains – Afp; B-chains correspond to DP > 3 in φ, β-limit dextrins and were divided into short (BS) and long (BL) chains at between DP 22–25 and above, respectively, depending on the sample; “Fingerprint” B-chains at DP 3–7 in φ, β- limit dextrins; The major group of short B-chains at DP 8 to 22–24, depending on the sample; Long chains between DP 22 or 25 and 55, depending on the sample; Long chains at DP > 56. LG: Large granules; GP: Gelatinized portion; RG: Remaining granules.

70 The selected molar ratios of different chain categories are shown in Table 3.4. The ratio A:B, Acrystal:BS and Acrystal:B in RG was higher than that in the GP for both samples.

There was no significant difference in the ratio Sap:Lap among LG, RG and GP of the two samples. The ratio BS:BL of GP was much more than the RG of P1 but was not statistically different in P2. A higher ratio Bfp:BSmajor was shown in RG of P1 whereas the ratios between two parts were similar in P2.

Table 3.4 Selected Molar Ratios of Different Chain Categories of Large Granules and their Gelatinized and Remaining Fraction Amylopectin and φ,β-Limit Dextrins. P1 P2

LG GP RG LG GP RG

A:B 1.07ab 1.02a 1.12bB 1.09b 1.00a 1.04abA

Sap:Lap 12.06 12.41 12.18 12.62 12.39 12.36

BS:BL 5.32a 5.63b 5.21aA 5.53 5.68 5.54B

ab a bB b a abA Acrystal:BS 1.10 1.04 1.17 1.09 0.99 1.05

ab a bB b a abA Acrystal:B 0.92 0.88 0.98 0.92 0.84 0.89

ab aA bB b aB aA Bfp:BSmajor 0.76 0.71 0.74 0.78 0.73 0.72

Values with different lowercase letters are significantly (P < 0.05) different within each sample. Values with different uppercase letters are significantly (P < 0.05) different between samples. A = A-chains; B = B-chains; Sap = short amylopectin chains; Lap = long amylopectin chains; BSmajor = the major group of short B-chains at DP 8 to 22–24, depending on the sample; and Bfp = “fingerprint” B-chains at DP 3–7 in φ, β-limit dextrins. Acrystal-chains were calculated as all A-chains – Afp. B-chains correspond to DP > 3 in φ, β-limit dextrins and were divided into short (BS) and long (BL) chains at between DP 22– 25 and above, respectively, depending on the sample. LG: Large granules; GP: Gelatinized portion; RG: Remaining granules.

71 3.4 Discussion

Since Maltese crosses were still shown in the RG of the two samples when observed under polarized light microscope (Fig. 3.5), it indicated that the crystalline structure of the

RG appeared to remain intact. The results also indicated that amylose was evenly distributed in the granules of P1 but more concentrated at the periphery of the granules in

P2. The LCAM was more distributed at the periphery for both samples, but the core area had more of SCAM in P1.

A B

Figure 3.5. Polarized light-micrographs of remaining granules (RG) of yellow pea starch after washed with cold water. A: RG of P1, treated for 300 min; B: RG of P2, treated for 190 min.

72 Although the CL of RG and GP were similar in P2, the CL of the RG where B- allomorph are known to be concentrated had longer CL than GP which contain more of A- allomorph in P1. It has been reported that the amylopectin of B-crystalline type starches had longer chain length (CL) than A- and C-types (J l Jane et al., 1999). This indicates long chains of amylopectin are more likely to crystallize into B-crystalline and short chains into

A-crystalline (Hizukuri, Kaneko, & Takeda, 1983). The SCL of RG was longer than the

GP in both two samples, which suggested that the SCL of the B-type starch was longer than that of A-type samples (Eric Bertoft et al., 2008). The LCLs of GP and RG in both samples, the difference between CL and SCL of amylopectin, fit into the range of C-types.

There was an extremely slight shift of amylopectin unit chain peak profiles between GP and RG of P1. But no shift was observed in P2. This result was in accordance with previous studies (Ao & Jane, 2007; Liu, Gu, Donner, Tetlow, & Emes, 2007). The external chain length of RG was higher than that GP in P1, consistent with that of B-types starches

(Hizukuri, 1985). Thicker lamella generally appeared in B-type starches due to longer external chains forming the crystalline structure of amylopectin compared with A-types.

However, P2 with similar ECL in its RG and GP was an exception. There was no significant difference in the j,b-limit value between RG and GP of both samples. It was also shown that the j,b-limit value had no correlation with the type of crystalline allomorph. The chain length of long B-chains (BL-CLLD) is longer in B-type starch than in A-types. As we can see, the BL-CLLD of the RG was longer than that of the GP of the two samples because more B-allomorph is expected in the in RG. However, there was no correlation between

BS-CLLD and the type of crystalline form.

73 In two samples, the RG contained more of A-chains than the GP due to the higher content of Acrystal-chains. But the content of Afp-chains, the difference of A-chains and

Acrystal-chains, was higher in the GP of both two samples. The higher molar amount of B- chains in the GP of both two samples seemed to be due to more of BS-chains. The BS- chains has been reported to be higher in A-type starches than in B-types (Hizukuri, 1985).

The RG had more B3-chains than the GP, which was accordant with the previous report that the B-type starches generally had more of long B-chains (Eric Bertoft et al., 2008).

3.5 Conclusion

The study indicated that chemical surface gelatinization can be used to investigate the core and periphery of pea starch granules. We found that amylose was evenly distributed in large granules of P1 but was more distributed at the periphery of the large granules in P2. More SCAM was found at the center where B-allomorphs are known to be concentrated than at the periphery where more of A-allomorphs are distributed for both samples. From the unit and internal chain profiles of amylopectin, longer SCL and BL-

CLLD were observed at the core of both samples granules. Afp-, B- and BS-chains were more concentrated at the periphery than at the core of the samples.

Chapter 4: Conclusion and Future Work

75 The two yellow pea samples used in this study were significantly different in their granule size distribution, amylose content, physicochemical properties and amylopectin unit and internal chain profiles. The mean diameter of starch granules of P1 was smaller than that of P2. The amylose content of P1 was higher than that of P2. P2 had longer chain length (CL) and internal chain length (ICL) but more of Acrystal were found in P1. It was observed that the differences in pasting and thermal properties were affected by amylopectin chain profiles. P1 with higher gelatinization temperature had more of Acrystal chains. Peak viscosity and enthalpy of gelatinization was shown to be significantly influenced by internal segments of amylopectin found in the amorphous lamella.

The differences in amylopectin unit and internal chains of periphery and core of granules as revealed by chemical surface gelatinization showed that longer SCL and BL-

CLLD were observed at the core of the large granules where B-allomorphs are known to be concentrated than at the periphery where more of A-allomorphs are distributed for both samples. Afp-, B- and BS-chains were more concentrated at the periphery than at the core.

More short amylose chains were observed at the core than at the periphery of large granules for both samples. Amylose was evenly distributed in the large granules of P1 but was more concentrated at the periphery of the large granules in P2.

The study showed the functional characteristics were affected by amylopectin chain profiles. Chemical gelatinization was used to investigate the differences in amylopectin unit and internal chains of periphery and core of yellow pea starch granules.

Chapter 5: Reference

77 Aggarwal, V., Singh, N., Kamboj, S. S., & Brar, P. S. (2004). Some properties of seeds

and starches separated from different Indian pea cultivars. Food chemistry, 85(4),

585-590.

Allen, O. N., & Allen, E. K. (1981). The Leguminosae, a source book of characteristics,

uses, and nodulation: Univ of Wisconsin Press.

Annor, G. A., Marcone, M., Bertoft, E., & Seetharaman, K. (2014). Unit and internal chain

profile of millet amylopectin. Cereal Chemistry, 91(1), 29-34.

Ao, Z., & Jane, J.-l. (2007). Characterization and modeling of the A-and B-granule starches

of wheat, triticale, and barley. Carbohydrate Polymers, 67(1), 46-55.

Bailey, J., & Whelan, W. J. (1961). Physical properties of starch I. Relationship between

iodine stain and chain length. Journal of Biological Chemistry, 236(4), 969-973.

Banks, W., Greenwood, C., & Khan, K. (1971). The interaction of linear, amylose

oligomers with iodine. Carbohydrate research, 17(1), 25-33.

Banks, W., Greenwood, C., & Muir, D. (1974). Studies on the Biosynthesis of Starch

Granules. Part 9. The Starch Granules from Wrinkled‐Seeded Peas. Starch‐

Stärke, 26(2), 46-50.

Bertoft, E. (2004). On the nature of categories of chains in amylopectin and their

connection to the super helix model. Carbohydrate Polymers, 57(2), 211-224.

Bertoft, E. (2013). On the building block and backbone concepts of amylopectin structure.

Cereal Chemistry, 90(4), 294-311.

Bertoft, E., Annor, G. A., Shen, X., Rumpagaporn, P., Seetharaman, K., & Hamaker, B. R.

(2016). Small differences in amylopectin fine structure may explain large

functional differences of starch. Carbohydrate Polymers, 140, 113-121.

78 Bertoft, E., & Koch, K. (2000). Composition of chains in waxy-rice starch and its structural

units. Carbohydrate Polymers, 41(2), 121-132.

Bertoft, E., Koch, K., & Åman, P. (2012). Building block organisation of clusters in

amylopectin from different structural types. International journal of biological

macromolecules, 50(5), 1212-1223.

Bertoft, E., Manelius, R., & Qin, Z. (1993). Studies on the Structure of Pea Starches. Part

2: α‐Amylolysis of Granular Wrinkled Pea Starch. Starch‐Stärke, 45(8), 258-

263.

Bertoft, E., Piyachomkwan, K., Chatakanonda, P., & Sriroth, K. (2008). Internal unit chain

composition in amylopectins. Carbohydrate Polymers, 74(3), 527-543.

Bertoft, E., & Spoof, L. (1989). Fractional precipitation of amylopectin alpha-dextrins

using methanol. Carbohydrate research, 189, 169-180.

Bogracheva, T. Y., Morris, V., Ring, S., & Hedley, C. (1998). The granular structure of

C‐type pea starch and its role in gelatinization. Biopolymers: Original Research

on Biomolecules, 45(4), 323-332.

Borowska, J., Zadernowski, R., Borowski, J., & Swiecicki, W. K. (1998). Intravarietal

differentiation of pea (Pisum sativum L. subsp. sativum) seeds–Their chemical

composition and physical properties. Plant Breed. Seed Sci, 42, 75-85.

Chung, H.-J., Liu, Q., & Hoover, R. (2010). Effect of single and dual hydrothermal

treatments on the crystalline structure, thermal properties, and nutritional fractions

of pea, lentil, and navy bean starches. Food Research International, 43(2), 501-

508.

79 Colonna, P., Buleon, A., Lemaguer, M., & Mercier, C. (1982). Pisum sativum and vicia

faba carbohydrates: Part IV — Granular structure of wrinkled pea starch.

Carbohydrate Polymers, 2(1), 43-59.

Cooke, D., & Gidley, M. J. (1992). Loss of crystalline and molecular order during starch

gelatinisation: origin of the enthalpic transition. Carbohydrate research, 227, 103-

112.

Cousin, R. (1997). Peas (Pisum sativum L.). Field Crops Research, 53(1-3), 111-130. de Souza Gomes, D., do Prado Cordoba, L., Rosa, L. S., Spier, M. R., Schnitzler, E., &

Waszczynskyj, N. (2018). Thermal, pasting properties and morphological

characterization of pea starch (Pisum sativum L.), rice starch (Oryza sativa) and

arracacha starch (Arracacia xanthorrhiza) blends, established by simplex-centroid

design. Thermochimica acta, 662, 90-99.

Deshpande, S. (1990). Food legumes: chemistry and technology. Advances in cereal

science and technology, 147-226.

Donovan, J. W. (1979). Phase transitions of the starch–water system. Biopolymers:

Original Research on Biomolecules, 18(2), 263-275.

Dubois, M., Gilles, K. A., Hamilton, J. K., Rebers, P. t., & Smith, F. (1956). Colorimetric

method for determination of sugars and related substances. Analytical chemistry,

28(3), 350-356.

Evans, I., & Haisman, D. (1982). The effect of solutes on the gelatinization temperature

range of potato starch. Starch‐Stärke, 34(7), 224-231.

80 Franco, C. M., Ciacco, C., & Tavares, D. Q. (1988). Studies on the susceptibility of

granular cassava and corn starches to enzymatic attack. Part 2: Study of the granular

structure of starch. Starch‐Stärke, 40(1), 29-32.

Fredriksson, H., Silverio, J., Andersson, R., Eliasson, A.-C., & Åman, P. (1998). The

influence of amylose and amylopectin characteristics on gelatinization and

retrogradation properties of different starches. Carbohydrate Polymers, 35(3-4),

119-134.

Gayin, J., Abdel-Aal, E.-S. M., Manful, J., & Bertoft, E. (2016). Unit and internal chain

profile of African rice (Oryza glaberrima) amylopectin. Carbohydrate Polymers,

137, 466-472.

Genkina, N. K., Wikman, J., Bertoft, E., & Yuryev, V. P. (2007). Effects of structural

imperfection on gelatinization characteristics of amylopectin starches with A-and

B-type crystallinity. Biomacromolecules, 8(7), 2329-2335.

Gidley, M. J., & 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. Carbohydrate research, 161(2), 291-300.

Gough, B., & Pybus, J. (1973). Effect of metal cations on the swelling and gelatinization

behaviour of large wheat starch granules. Starch‐Stärke, 25(4), 123-130.

GUJSKA, E., D.‐REINHARD, W., & KHAN, K. (1994). Physicochemical properties of

field pea, pinto and navy bean starches. Journal of Food Science, 59(3), 634-636.

Han, H., & Sosulski, F. (1998). Cationization of potato and tapioca starches using an

aqueous alcoholic‐alkaline process. Starch‐Stärke, 50(11‐12), 487-492.

81 Haworth, W., Hirst, E., & Isherwood, F. (1937). 129. Polysaccharides. Part XXIII.

Determination of the chain length of glycogen. Journal of the Chemical Society

(Resumed), 577-581.

Hizukuri, S. (1985). Relationship between the distribution of the chain length of

amylopectin and the crystalline structure of starch granules. Carbohydrate

research, 141(2), 295-306.

Hizukuri, S., Kaneko, T., & Takeda, Y. (1983). Measurement of the chain length of

amylopectin and its relevance to the origin of crystalline polymorphism of starch

granules. Biochimica et Biophysica Acta (BBA)-General Subjects, 760(1), 188-191.

Holm, J., Björck, I., Ostrowska, S., Eliasson, A. C., Asp, N. G., Larsson, K., & Lundquist,

I. (1983). Digestibility of amylose‐lipid complexes in‐vitro and in‐vivo.

Starch‐Stärke, 35(9), 294-297.

Hoover, R., & Hadziyev, D. (1981). The effect of monoglycerides on amylose complexing

during a potato granule process. Starch‐Stärke, 33(10), 346-355.

Hoover, R., Hannouz, D., & Sosulski, F. (1988). Effects of Hydroxypropylation on Themal

Properties, Starch Digestibility and Freeze‐Thaw Stability of Field Pea (Pisum

sativum cv Trapper) Starch. Starch‐Stärke, 40(10), 383-387.

Hoover, R., Hughes, T., Chung, H., & Liu, Q. (2010). Composition, molecular structure,

properties, and modification of pulse starches: A review. Food Research

International, 43(2), 399-413.

Hoover, R., & Manuel, H. (1995). A comparative study of the physicochemical properties

of starches from two lentil cultivars. Food chemistry, 53(3), 275-284.

82 Hoover, R., & Manuel, H. (1996). The effect of annealing on the physicochemical

properties of legume starches. SPECIAL PUBLICATION-ROYAL SOCIETY OF

CHEMISTRY, 179(1), 157-161.

Hoover, R., & Ratnayake, W. (2002). Starch characteristics of black bean, chick pea, lentil,

navy bean and pinto bean cultivars grown in Canada. Food chemistry, 78(4), 489-

498.

Hoover, R., & Senanayake, S. (1996). Composition and physicochemical properties of oat

starches. Food Research International, 29(1), 15-26.

Hoover, R., & Sosulski, F. (1985a). A comparative study of the effect of acetylation on

starches of Phaseolus vulgaris biotypes. Starch‐Stärke, 37(12), 397-404.

Hoover, R., & Sosulski, F. (1985b). Studies on the functional characteristics and

digestibility of starches from Phaseolus vulgaris biotypes. Starch‐Stärke, 37(6),

181-191.

Huang, J., Schols, H. A., van Soest, J. J., Jin, Z., Sulmann, E., & Voragen, A. G. (2007).

Physicochemical properties and amylopectin chain profiles of cowpea, chickpea

and yellow pea starches. Food chemistry, 101(4), 1338-1345.

Imberty, A., & Perez, S. (1988). A revisit to the three‐dimensional structure of B‐type

starch. Biopolymers: Original Research on Biomolecules, 27(8), 1205-1221.

Imberty, A., & Pérez, S. (1989). Conformational analysis and molecular modelling of the

branching point of amylopectin. International journal of biological

macromolecules, 11(3), 177-185.

83 Jane, J., Xu, A., Radosavljevic, M., & Seib, P. (1992). Location of amylose in normal starch

granules. I. Susceptibility of Amylose and Amylopectin to Cross-Linking Reagents,

69, 405-409.

Jane, J.-l., & Shen, J. J. (1993). Internal structure of the potato starch granule revealed by

chemical gelatinization. Carbohydrate research, 247, 279-290.

Jane, J.-l., Wong, K.-s., & McPherson, A. E. (1997). Branch-structure difference in

starches of A-and B-type X-ray patterns revealed by their Naegeli dextrins.

Carbohydrate research, 300(3), 219-227.

Jane, J. L. (1993a). 13C‐NMR study of interactions between amylodextrin and neutral

salts. Starch‐Stärke, 45(5), 172-175.

Jane, J. L. (1993b). Mechanism of starch gelatinization in neutral salt solutions. Starch‐

Stärke, 45(5), 161-166.

Jane, J. l., Chen, Y., Lee, L., McPherson, A., Wong, K., Radosavljevic, M., &

Kasemsuwan, T. (1999). Effects of amylopectin branch chain length and amylose

content on the gelatinization and pasting properties of starch. Cereal Chemistry,

76(5), 629-637.

Jensen, B., Winters, J., & Hubbe, M. (1986). Cationic potato starches proven effective as

wet-end additives. Pulp Pap, 60(4), 83.

Kalinga, D. N., Waduge, R., Liu, Q., Yada, R. Y., Bertoft, E., & Seetharaman, K. (2013).

On the differences in the granular architecture and starch structure between pericarp

and endosperm wheat starches. Starch‐Stärke, 65(9‐10), 791-800.

Karim, A. A., Norziah, M., & Seow, C. (2000). Methods for the study of starch

retrogradation. Food chemistry, 71(1), 9-36.

84 Koch, K., Andersson, R., & Åman, P. (1998). Quantitative analysis of amylopectin unit

chains by means of high-performance anion-exchange chromatography with pulsed

amperometric detection. Journal of Chromatography A, 800(2), 199-206.

Koch, K., & Jane, J. L. (2000). Morphological changes of granules of different starches by

surface gelatinization with calcium chloride. Cereal Chemistry, 77(2), 115-120.

Kong, X., Bertoft, E., Bao, J., & Corke, H. (2008). Molecular structure of amylopectin

from amaranth starch and its effect on physicochemical properties. International

journal of biological macromolecules, 43(4), 377-382.

Kong, X., Corke, H., & Bertoft, E. (2009). Fine structure characterization of amylopectins

from grain amaranth starch. Carbohydrate research, 344(13), 1701-1708.

Kozlov, S., Krivandin, A., Shatalova, O. V., Noda, T., Bertoft, E., Fornal, J., & Yuryev, V.

(2006). Structure of starches extracted from near-isogenic wheat lines: Part 2.

Molecular organization of amylopectin clusters. Journal of Thermal Analysis and

Calorimetry, 87(2), 575-584.

Kuakpetoon, D., & Wang, Y.-J. (2007). Internal structure and physicochemical properties

of corn starches as revealed by chemical surface gelatinization. Carbohydrate

research, 342(15), 2253-2263.

Laohaphatanaleart, K., Piyachomkwan, K., Sriroth, K., & Bertoft, E. (2010). The fine

structure of cassava starch amylopectin: Part 1: Organization of clusters.

International journal of biological macromolecules, 47(3), 317-324.

Laohaphatanaleart, K., Piyachomkwan, K., Sriroth, K., Santisopasri, V., & Bertoft, E.

(2009). A study of the internal structure in cassava and rice amylopectin. Starch‐

Stärke, 61(10), 557-569.

85 Lim, S. T., Han, J. A., Lim, H., & BeMiller, J. (2002). Modification of starch by dry heating

with ionic gums. Cereal Chemistry, 79(5), 601-606.

Liu, Q., Donner, E., Yin, Y., Huang, R., & Fan, M. (2006). The physicochemical properties

and in vitro digestibility of selected cereals, tubers and legumes grown in China.

Food chemistry, 99(3), 470-477.

Liu, Q., Gu, Z., Donner, E., Tetlow, I., & Emes, M. (2007). Investigation of digestibility

in vitro and physicochemical properties of A‐and B‐type starch from soft and

hard wheat flour. Cereal Chemistry, 84(1), 15-21.

Maache-Rezzoug, Z., Maugard, T., Zarguili, I., Bezzine, E., El Marzouki, M.-N., & Loisel,

C. (2009). Effect of instantaneous controlled pressure drop (DIC) on

physicochemical properties of wheat, waxy and standard maize starches. Journal

of cereal science, 49(3), 346-353.

Meyer, K. H., & Bernfeld, P. (1940). Recherches sur l'amidon V. L'amylopectine.

Helvetica Chimica Acta, 23(1), 875-885.

Nara, S., & Komiya, T. (1983). Studies on the relationship between water‐satured state

and crystallinity by the diffraction method for moistened potato starch. Starch‐

Stärke, 35(12), 407-410.

Nikuni, Z. (1969). Starch and cooking. Science of Cookery, 2, 6-14.

Nikuni, Z. (1978). Studies on starch granules. Starch‐Stärke, 30(4), 105-111.

Noda, T., Takahata, Y., Sato, T., Ikoma, H., & Mochida, H. (1996). Physicochemical

properties of starches from purple and orange fleshed sweet potato roots at two

levels of fertilizer. Starch‐Stärke, 48(11‐12), 395-399.

86 Nunes, M., Raymundo, A., & Sousa, I. (2006). Rheological behaviour and microstructure

of pea protein/κ-carrageenan/starch gels with different setting conditions. Food

Hydrocolloids, 20(1), 106-113.

Okuda, M., Aramaki, I., Koseki, T., Satoh, H., & Hashizume, K. (2005). Structural

characteristics, properties, and in vitro digestibility of rice. Cereal Chemistry,

82(4), 361-368.

Pan, D. D., & Jane, J.-l. (2000). Internal structure of normal maize starch granules revealed

by chemical surface gelatinization. Biomacromolecules, 1(1), 126-132.

Peat, S., Whelan, W. J., & Thomas, G. J. (1952). Evidence of multiple branching in waxy

maize starch. Journal of the Chemical Society (Resumed), 4536-4538.

Pérez, S., & Bertoft, E. (2010). The molecular structures of starch components and their

contribution to the architecture of starch granules: A comprehensive review.

Starch‐Stärke, 62(8), 389-420.

Popov, D., Buléon, A., Burghammer, M., Chanzy, H., Montesanti, N., Putaux, J.-L., . . .

Riekel, C. (2009). Crystal structure of A-amylose: A revisit from synchrotron

microdiffraction analysis of single crystals. Macromolecules, 42(4), 1167-1174.

Pukkahuta, C., Shobsngob, S., & Varavinit, S. (2007). Effect of osmotic pressure on starch:

new method of physical modification of starch. Starch‐Stärke, 59(2), 78-90.

Raghunathan, R., Hoover, R., Waduge, R., Liu, Q., & Warkentin, T. (2017). Impact of

molecular structure on the physicochemical properties of starches isolated from

different field pea (Pisum sativum L.) cultivars grown in Saskatchewan, Canada.

Food chemistry, 221, 1514-1521.

87 Ratnayake, W., Hoover, R., Shahidi, F., Perera, C., & Jane, J. (2001). Composition,

molecular structure, and physicochemical properties of starches from four field pea

(Pisum sativum L.) cultivars. Food chemistry, 74(2), 189-202.

Ratnayake, W. S., Hoover, R., & Warkentin, T. (2002). Pea starch: composition, structure

and properties—a review. Starch‐Stärke, 54(6), 217-234.

Ring, S. G., Gee, J. M., Whittam, M., Orford, P., & Johnson, I. T. (1988). Resistant starch:

its chemical form in foodstuffs and effect on digestibility in vitro. Food chemistry,

28(2), 97-109.

Rosenthal, F., Espindola, L., Serapiao, M., & Silva, S. (1971). Lablab bean starch. I.

Properties of its granules and pastes. Starch‐Stärke, 23(1), 18-23.

Sandhu, K. S., & Lim, S.-T. (2008). Digestibility of legume starches as influenced by their

physical and structural properties. Carbohydrate Polymers, 71(2), 245-252.

Sasaki, T., & Matsuki, J. (1998). Effect of wheat starch structure on swelling power. Cereal

Chemistry, 75(4), 525-529.

Shi, M., Zhang, Z., Yu, S., Wang, K., Gilbert, R. G., & Gao, Q. (2014). Pea starch (Pisum

sativum L.) with slow digestion property produced using β -amylase and

transglucosidase. Food chemistry, 164, 317-323.

Simsek, S., Tulbek, M. C., Yao, Y., & Schatz, B. (2009). Starch characteristics of dry peas

(Pisum sativum L.) grown in the USA. Food chemistry, 115(3), 832-838.

Snow, P., & O'Dea, K. (1981). Factors affecting the rate of hydrolysis of starch in food.

The American journal of clinical nutrition, 34(12), 2721-2727.

88 Staudinger, H., & Husemann, E. (1937). Über hochpolymere Verbindungen. 150.

Mitteilung. Über die Konstitution der Stärke. Justus Liebigs Annalen der Chemie,

527(1), 195-236.

Szymo ń ska, J., Krok, F., & Tomasik, P. (2000). Deep-freezing of potato starch.

International journal of biological macromolecules, 27(4), 307-314.

Takeda, Y., Hizukuri, S., & Juliano, B. O. (1987). Structures of rice amylopectins with low

and high affinities for iodine. Carbohydrate research, 168(1), 79-88.

Tester, R. F., & Morrison, W. R. (1990). Swelling and gelatinization of cereal starches. I.

Effects of amylopectin, amylose, and lipids. Cereal chem, 67(6), 551-557.

Valetudie, J. C., Colonna, P., Bouchet, B., & Gallant, D. J. (1993). Hydrolysis of tropical

tuber starches by bacterial and pancreatic α‐amylases. Starch‐Stärke, 45(8),

270-276.

Vermeylen, R., Goderis, B., Reynaers, H., & Delcour, J. A. (2004). Amylopectin molecular

structure reflected in macromolecular organization of granular starch.

Biomacromolecules, 5(5), 1775-1786.

Waigh, T. A., Kato, K. L., Donald, A. M., Gidley, M. J., Clarke, C. J., & Riekel, C. (2000).

Side‐chain liquid‐crystalline model for starch. Starch‐Stärke, 52(12), 450-460.

Wang, N., & Daun, J. K. (2004). Effect of variety and crude protein content on nutrients

and certain antinutrients in field peas (Pisum sativum). Journal of the Science of

Food and Agriculture, 84(9), 1021-1029.

Wang, S., Sharp, P., & Copeland, L. (2011). Structural and functional properties of starches

from field peas. Food chemistry, 126(4), 1546-1552.

89 Wang, S., Yu, J., & Yu, J. (2008). The semi-crystalline growth rings of C-type pea starch

granule revealed by SEM and HR-TEM during acid hydrolysis. Carbohydrate

Polymers, 74(3), 731-739.

Wattanachant, S., Muhammad, K., Hashim, D. M., & Rahman, R. A. (2003). Effect of

crosslinking reagents and hydroxypropylation levels on dual-modified sago starch

properties. Food chemistry, 80(4), 463-471.

Yook, C., PEK, U. H., & PARK, K. H. (1993). Gelatinization and retrogradation

characteristics of hydroxypropylated and cross‐linked rices. Journal of Food

Science, 58(2), 405-407.

Zhou, Y., Hoover, R., & Liu, Q. (2004). Relationship between α-amylase degradation and

the structure and physicochemical properties of legume starches. Carbohydrate

Polymers, 57(3), 299-317.

Zhu, F., Corke, H., & Bertoft, E. (2011). Amylopectin internal molecular structure in

relation to physical properties of sweetpotato starch. Carbohydrate Polymers,

84(3), 907-918.

Zobel, H. (1988). Starch crystal transformations and their industrial importance. Starch‐

Stärke, 40(1), 1-7.

Chapter 6. Appendices

91 Appendix A. Pasting properties of two yellow pea starch samples.

140 100

120 80 100 Temp/ 80 60 60 40 Torque/mPas 40 20 20

0 0 0 5 10 15 20 25 30 Time/min

140 100

120 80 100 Temp/ 80 60 60 40 Torque/mPas 40 20 20

0 0 0 5 10 15 20 25 30 Time/min

Figure 6.1. Pasting properties of two yellow pea starch samples.

92 Appendix B. Thermal properties of two yellow pea starch samples.

To Tp Tc a a a

P1 ΔH = 4.32a

b b b

ΔH = 7.75b P2

45 50 55 60 65 70 75 Temperature(°C)

Figure 6.2. Thermal properties of two yellow pea starch samples.