Copyright is owned by the Author of the thesis.  Permission is given for 
a copy to be downloaded by an individual for the purpose of research and 
private study only.  The thesis may not be reproduced elsewhere without 
the permission of the Author. 
 



EVALUATION OF FUNCTIONALITY OF 
COMMERCIAL RESISTANT STARCHES IN 

FOOD SYSTEMS 

. ~-~ ,,, . 
'tl , ' Massey University 

A THESIS PRESENTED IN PARTIAL FULFILLMENT OF 
THE REQUIREMENTS FOR THE DEGREE OF MASTERS 

OF TECHNOLOGY 

By 

AMIT TANEJA 

RIDDET CENTER AND 
INSTITUTE OF FOOD NUTRITION AND HUMAN HEAL TH 

MASSEY UNIVERSITY 
PALMERSTON NORTH 

2005 



TO MY MOTHER 



Abstract 

ABSTRACT 

The objective of this study was (i) to investigate the functional 

properties of commercial resistant starches in a fluid model food system, (ii) to 

determine the level of resistant starch that could replace the regular thickener 

without affecting the sensory properties of the system and (iii) to verify the 

claims made by manufacturers of resistant starches. 

In order to evaluate four commercially available resistant starches , a 

chicken soup model food system was developed. The choice of food system 

was based on the ease of rheological measurement along with relatively easy 

method of preparation. A representative soup formulation was chosen which 

contained industrial starch, wheat flour and xanthan gum as th ickening 

agents. A suitable experimental plan was developed using fractional factorial 

and central composite designs for evaluation of the soup model . The viscosity 

of the soup model was determined using Paar Physica rheometer and the 

sensory analysis was done using acceptance and simple difference testing. 

The rheological properties, i.e. the consistency index (K) and flow 

behavior index (n) , derived from the power law model, for the soup model 

were analyzed using response surface methodology, which enabled an 

evaluation of the functionality of the model and visualization of correlation 

between various factors (ingredients) and resistant starch . Results revealed 

that all resistant starches lacked any starch like functionality as none of them 

was able to replace the waxy maize starch functionality to any significant 

extent. Hence, it was necessary to allow for the replacement of waxy maize 

starch by increasing the amount of xanthan gum in the formulations. Thus, 

regression models, built to predict the optimum regions of response, were 

used in replacing waxy maize starch in soup with resistant starch by 

increasing the amount of xanthan gum. 

Comparative sensory responses obtained from paired sample testing 

determined that the optimum level at which resistant starch could be added to 

soup model was only 20%. At higher levels (40% and 60%), a difference in 



Abstract 11 

taste could be perceived . 

The claims made by manufacturers regarding the thermal stability of 

resistant starches were val idated and the in vitro assays showed no 

significant difference (P>0.05) in percent resistant starch (dry weight basis) 

level with the increase in holding time (5-20mins) at 95 °C while soup making. 



Acknowledgement 111 

ACKNOWLEDGEMENT 

I wish to express my sincere gratitude to my supervisors, Professor 

Harjinder Singh and Dr. Derek Haisman for their excellent supervision , 

understanding , and encouragement throughout the project. They showed me the 

logical way to approach problems with their patience and helpful discussion. I 

also convey my special thanks to Professor Paul Maughan for his support and 

valuable guidance. 

Special thanks to Dr. Nigel Grigg for his expert advice on experimental 

designs and helpful discussions on statistical analysis. 

I am very grateful to the staff of the Institute of Food, Nutrition and Human 

Health. In particular, I thank Mr. Steve Glasgow, Mr. Warwick Johnson, Mr. 

Christopher Hall , Ms Michelle Tamehana , Ms Susan Simms, Ms Karen Pickering 

and Ms Yvonne Parkes for their kindness and technical assistance during my 

post graduate study at Massey University. I also thank all fellow postgraduates 

and PD hutters. 

Finally, I would like to express my genuine gratitude to my father, Dr. 

Pervez Taneja , my sister Ms Nancy Taneja, my brother-in-law Mr Nitin Gautam 

and my friend Ms. Namrata Behl for their love, support and encouragement 

throughout my masters . 



Table of contents 

ABSTRACT 

ACKNOWLEDGEMENT 

TABLE OF CONTENTS 

TABLE OF CONTENTS 

IV 

iii 

iv 

Chapter 1 INTRODUCTION 1 

Chapter 2 LITERATURE REVIEW 3 

2 .1 Starch - chemistry, properties , various sources and uses in 

foods 3 

2 .2 Gelatinization 6 

2.3 Retrog radation 9 

2.4 Rheological properties of starch dispersions 15 

2.5 Resistant starch 

2.5.1 History 20 

2 5.2 Relationship with glycaemic index 20 

2.5.3 Resistant starch as a functional food 21 

2.5.4 Classification 22 

2.5.5 Digestion and fermentation of resistant starch 23 

2.5.6 Analysis of resistant starch in foods 27 

2.5.7 Production of resistant starch 36 

2.5.8 Thermal analysis of resistant starches using differential 

scanning calorimetry (DSC) 43 

2.5.9 Application of resistant starches 45 

2.6 Conclusions 47 

Chapter 3 MATERIALS AND METHODS 50 

3.1 Materials 50 

3.2 Approach to development of model food system for 

functionality testing of resistant starch 51 

3.3 Method of soup preparation 52 

3.4 Experimental design 52 



Table of contents 

3.4.1 The 2k factorial design 

3.4.2 Response surface methodology 

3.5 Viscosity measurement of model food system (soup) 

3.6 Sensory evaluation 

3.6.1 Acceptance test 

3.6.2 Simple difference test 

3.7 Resistant starch assay 

3.8 DSC thermal analysis 

Chapter 4 RESULTS AND DISCUSSION 

4 .1 Introduction - scheme of research 

4 .2 Screening 

4.2.1 Experimental design 

4.2.2 Results and discussion 

4.2.3 Conclusions 

4 .3 A second-order model to predict the effect of WMS and 

WF on the model food system 

4.3.1 Experimental design 

4.3.2 Results and discussion 

4.3.3 Conclusions 

4.4 The effect of wheat flour on the sensory perception of the 

model food system 

4.4.1 Test objective 

4.4.2 Experimental design 

4.4.3 Results and discussion 

4.4.4 Conclusions 

4.5 Functionality of resistant starch 

4.5.1 Results and discussion 

4.5.2 Conclusions 

4.6 A second-order model to predict the effect of waxy maize 

starch and xanthan gum on the model food system 

4. 6. 1 Experimental design 

V 

52 

54 

56 

58 

58 

58 

60 

64 

66 

66 

67 

68 

68 

75 

76 

76 

78 

82 

82 

82 

82 

83 

83 

85 

85 

85 

86 

86 



Table of contents 

4.6.2 Results and discussion 

4.6.3 Conclusions 

4.7 Thermal behavior of resistant starches 

4. 7. 1 Experimental design 

4.7.2 Results and discussion 

4.7.3 Conclusions 

4.8 The effect of holding time on %RS content of model food 

system 

4.8.1 Experimental design 

4.8.2 Results and discussion 

4.8.3 Conclusions 

4.9 Replacing waxy maize starch by RS in model food system 

4. 9. 1 Test format 

4. 9. 2 Results and discussion 

VI 

86 

89 

91 

91 

91 

94 

95 

95 

95 

97 

97 

98 

4. 9. 3 Conclusions 98 

Chapter 5 OVERALL CONCLUSIONS AND RECOMMENDATIONS 

FOR FURTHER WORK 

BIBLIOGRAPHY 

APPENDIX 

A1 Products with hi maize as an ingredient marketed in 

Australia and New Zealand 

A2 Basic soup recipes sourced to formulate cream of 

chicken soup for the research 

A3 Final soup formulation used in present research 

A4 A sample questionnaire for acceptance test 

A5 A typical worksheet for simple difference test 

A6 Box Cox transformation plot and lambda table for 

the second order model of WMS and XG 

A7 Moisture content determination (AOAC official 

method 925.10) 

104 

102 

110 

110 

111 

113 

114 

120 

121 

122 



Table of contents 

A8 

A9 

The analysis of variance table for %RS versus 

holding time for model soups 

Work sheet and score sheet for same/different test 

--------------------·-----

Vil 

123 

124 



Chapter 1: Introduction 

CHAPTER 1 

INTRODUCTION 

Until recently , starch was thought of as a source of carbohydrate that was 

completely digested and assimilated in the small intestine . However, it is well 

known that there exists a small but variable starch fraction that is resistant to 

hydrolysis by enzymes of the pancreas. Passing through the small intestine, this 

fraction reaches the large bowel where it is fermented by the colonic microflora to 

a variable extent. This fraction is called resistant starch (RS) and on that account 

is defined as the sum of starch and the products of starch degradation not 

absorbed in the small intestine of a healthy human being (Niba et al., 2002) . 

The main factors that influence rate and extent of starch digestion and 

absorption are both intrinsic and extrinsic. Processing conditions adopted and the 

resulting retrogradation steps may affect digestibility characteristics of starch­

based foods . Also the duration and conditions of storage greatly influence the 

digestion of starch (Namratha et al., 2002). 

The rates of obesity continue to increase in most western countries 

despite the efforts made by health organization and state departments . For the 

last 20 years , reducing fat intake has been the primary focus of dietary 

prevention and treatment of obesity . Animal studies and human epidemiological 

studies have shown a relationship between dietary fat and body weight and a 

reduction in its intake can produce significant weight loss in obese subjects but 

weight regain often occurs (Brand-Miller et al., 2002) . 

However, studies on the postprandial effects of carbohydrate dense, high­

glycaemic index (GI) diets have helped to explain why low-fat diets have failed to 

inhibit weight gain in obese subjects . Rapid digestion and absorption 

accompanied by higher insulin response after high-GI foods dictates energy 

partitioning and satiety wh ich favors expansion of fat stores over long term. 

Therefore , low GI diets are clearly justifiable for the prevention and treatment of 



Chapter 1: Introduction 2 

obesity in overweight subjects . Numerous studies have shown that food products 

high in resistant starches frequently result in low glycaemic and insulinaemic 

responses mainly due their resistance to digestion, which is the key determinant 

of the glycaemic index (Akerberg et al. , 1998; Brand-Miller et al., 2002) . 

RS is a natural component of many foods but its quantity in our diets still 

needs to be increased . This can be achieved by careful selection of foods and 

changes in culinary practices . This phenomenon of the resistance of starch to 

human digestion needs to be exploited by the food industry to achieve high levels 

of RS in familiar food products such as breakfast cereals, breads, bars and 

biscuits. Such functional foods need to be manufactured in greater variety and 

with higher palatability than many of the familiar high fibre products currently 

available to consumers (Johnson and Gee, 1996). 

New starch ingredients can be evaluated in food by using a model food 

system. This is a valuable tool to test ingredients on a small scale in a complex 

food system which can be extrapolated to real food products . Selected physical 

properties of starches have been used to study starch functionality using a food 

model system. Highly reproducible food models were used to evaluate functional 

and rheological properties of modified starches by Wischmann et al. (2002) . 

Following this line, the objective of this study was to investigate the 

functional properties of commercially available resistant starches in a fluid model 

food system. Moreover, it was the aim to determine the level of resistant starch 

that could replace the regular thickener without affecting the sensory properties 

of the system. Response surface methodology was used to collect rheological 

data on the food system which enabled us to describe the functionality of 

resistant starches and correlate it with their concentration . In this context , the 

claims made by manufacturers of resistant starches were also verified . 



Chapter 2: Literature review 3 

CHAPTER 2 

LITERATURE REVIEW 

2.1 Starch chemistry, properties, various sources and uses in 

foods 

Starch , a mixture of glucans, is the principal food reserve polysaccharide 

found mainly in the plant kingdom, where it may be utilized for growth . Out of the 

total calories consumed by humans worldwide , starch accounts for 70-80%. After 

cellulose, the most widely commercially utilized of all polysaccharides is starch 

(Greenwood , 1970). 

All starches occur in nature as minute granules , each with its inherent 

characteristic size and shape irrespective of the source . The commercial 

starches are divided into three groups. The first group comprises of tuber (potato) 

and root (tapioca , arrowroot and sweet potato) and pith (sago) starches. Cereal 

starches (corn , rice , wheat and sorghum) make up the second group. Both these 

groups differ from each other in chemical composition and physical properties . 

Group three contains waxy starches (waxy maize, waxy sorghum and waxy rice) , 

which are certain mutants of maize, rice and sorghum, cultivated for their special 

characteristics . Even though obtained from cereals , their physical properties are 

similar to those of root starches (Swinkels , 1985) . 

Starch is composed of a mixture of two polymers : an essentially linear 

polysaccharide called amylose and a highly branched polysaccharide called 

amylopectin . Amylogenin , a self-glucosylating initiator protein molecule , is 

attached to a single ~-D- glucopyranosyl unit, from which the starch molecule 

grows. Adenosine diphosphate molecules donate a -D- glucopyranosyl producing 

a chain joined by 1- 4 linkages. In addition to this , a branching enzyme is also 

active. In order to transfer a linear chain , the enzyme requires a linear chain of 

40-50 units . The transferred portion becomes an a - 1- 6 branch , whereupon 



Chapter 2. Literature review 4 

both non reducing ends continue elongation. This constitutes the amylopectin 

molecule. Some branches may occur as double helices of parallel chains 

(Whistler & BeMiller, 1997). 

Amylopectin constitutes about 75% of most common starches. Some 

starches be entirely amylopectin and are called waxy starches. These are called 

so because, when the kernel is cut, the new surface appears vitreous or waxy but 

there is no wax present Structures of amylopectin and amylase molecules are 

shown in Figure 2.1 (Whistler & BeMiller, 1997). 

Amylose is essentially a linear chain of (1-> 4) linked a -D- glucopyranosyl 

units with an average molecular weight of about 106 Amylose takes a helical 

shape due to the axial equatorial position coupling of ( 1- 4) linked a -D­

glucopyranosyl units. This results in formation of more elastic films and fibers 

than formed by cellulose. Amylase molecules occur among the amylopectin 

molecules and can diffuse out of partially water-swollen granules (Whistler & 

BeMiller, 1997). 

· .... ~ .. 
. \_;;..._ ' . ' I 

··~··· 
. ... . , .• \.:...,,,_ . .. ' 

a b 
... ~.· 

Figure 2.1: Representative partial structures of a) Amylose; b) Amylopectin 

(Pomeranz, 1991). 



Chapter 2: Literature review 5 

The quasi-crystalline nature of starch molecules in a granule is due to the 

radial ordered arrangement, as evident from the polarization cross (bifringence) 

seen using a polarizing microscope (Figure 2.2). The center of the cross is at the 

growth point, the hilum. Type-A X-ray pattern indicates parallel double helices 

separated by interstitial water, which is found in cereal starches. In tuber and root 

starches, a column of water molecules replaces one of the double helices, which 

produce Type-8 X-ray pattern. The structure of the granule is formed by 

amylopectin molecules, arranged with their reducing ends towards the center of 

the granule (Whistler & BeMiller, 1997). 

The unique particulate nature of starch and its properties such as high 

viscosity on gelatinization and its gelling or non-gelling characteristics, lead 

themselves to an array of uses. Their major applications in the food industry are 

as thickening agents (sauces, cream soups, pie fillings), as colloidal stabilizers 

(salad dressings), for moisture retention (cake toppings), as gel-forming agents 

(gum confections), binders (wafers, ice cream cones) and as coating and glazing 

agents (nut meats, candies) (Swinkels, 1985). 

Figure 2.2: An image of wheat granule (A) using the first microscope by Van 

Leewenhoek, and an image of potato starch (B) viewed under polarised light 

(Source: Wang et al., 1998). 



Chapter 2: Literature review 6 

In some instances, these properties are inherent in native starch , but 

occasionally, they can be augmented or introduced into the starch by (a) physical 

modifications , (b) non-degradative chemical modifications or (c) degradative 

modifications . Physical modifications are limited to variation in drying conditions . 

Starches react with a wide variety of chemical reagents to form ethers or esters 

which lead to an alteration of swelling properties along with associated properties 

of the starch molecule . Such starches comprise the non-degradative chemical 

modifications. The degradative mod ifications of starch lead to thin boiling 

starches, oxidized starches, dextrins and sugars and syrups (Pomeranz, 1991 ). 

2.2 Gelatinization 

Starch granules are insoluble in cold water due their semi-crystall ine 

structure, stabilized by the hydrogen bonds, formed either directly via neighboring 

OH groups or indirectly by water bridges. Even though the forces of the hydrogen 

bonds are weak , they are present in large numbers and can confer considerable 

stability . However, the granules can imbibe water reversibly and swell slightly 

(10-15% in diameter) in cold water. On drying , they shrink back (Swinkels , 1985). 

On progressively heating in water at high temperatures , the granules start 

to swell irreversibly and the characteristic polarization-cross starts to fade at the 

hilum. This loss of bifringence and the concurrent initiation of swelling is termed 

as gelatinization . Initiation of swelling takes place at the amorphous regions and 

starch molecules are hydrated by the disruption of weak bonds. The orderly 

radial arrangement is disturbed due to swelling leading to loss of bifringence . 

This is followed by more swelling and the disruption of hydrogen bonds in the 

crystalline region . Some amylase molecules may leach out increasing the 

viscosity of the surrounding phase. Because amylose is a relatively small , linear 

molecule, it can easily diffuse out of the granule (Swinkels , 1985) . 

Gelatinization occurs over a range of temperature with larger molecules 

gelatinizing first and the smaller ones later. However, this may depend on the 

granule type , starch-water ratio and the heterogeneities within the granule 



Chapter 2: Literature review 7 

population (Pomeranz, 1991) . Amylase content also affects the temperature of 

gelatinization as high amylase varieties of corn have a higher temperature range 

compared to the low amylase ones. The gelatinization temperature range of 

starches derived from different sources is shown in Table 2.1. The granule finally 

ruptures and collapses on persistent heating and agitation at high temperatures 

yielding a viscous colloidal dispersion of swollen granule fragments , hydrated 

starch aggregates and dissolved molecules (Swinkels , 1985). 

Gelatinization of granules occurs because the high temperature 

encourages vibration of the molecule , breaking the hydrogen bonds . More 

extensive hydration is produced when water molecules replace the broken 

hydrogen bonds. The starch molecules move freely and get sheathed in layers of 

water making it impossible for them to return to their original positions upon 

dehydration (Whistler & BeMiller, 1997). 

However, the same end can be achieved at room temperature by the use 

of solvents such as liquid ammonia and dimethyl sulfoxide or mechanically , by 

milling (Blanshard , 1987) . Gelatinization of starch can be examined by differential 

scanning calorimetry (DSC) , which measures both temperature and enthalpy of 

gelatinization . The Brabender viscoamylograph is used to examine the cooking 

behavior of different starches. This instrument records the viscosity change when 

temperature is raised to 95°C , where it is held for a short time and then lowered 

(Whistler & BeMiller, 1997) . The gelatinization of starch is affected by various 

factors . Concentration and degree of shear are the main ones. Proteins present 

in flour of cereal starches such as wheat interact due to the attraction of opposite 

charges. These form complexes during gelatinization . The interaction is low at 

alkaline pH levels , at which both starch and protein bear a negative charge , and 

high at acidic pH , at which both bear a positive charge. 



Table 2.1: Physical and chemical properties of starch (From: Pomeranz, 1991 ). 

Starch Granule size (µm) Amylose Swelling Solubility Gelatinization 
power Source Taste 

Range Ave rage (%) 
(at 95°C) 

at 95°C range (0 C) 

Barley 2-35a 20 22 56-62 Cereal Low 
Corn 

Regular 5-25b 15 26 24 25 62-80 Cereal Low 
Waxy 5-25 15 - 1 64 23 63-74 Cereal Low 
High 15 Up to 80 6 12 85-87 Cereal Low 

amylase 
Potato 15-100 33 22 1000 82 56-69 Tuber Slight 
Rice 3-8c 5 17 19 61-80 Cereal Low 
Rye 2-35d 23 57-70 Cereal Low 

Sago 20-60 25 27 97 60-74 Pith Low 
Sorghum 5-25e 15 26 22 48 52-64 Cereal Low 
Tapioca 5-35 20 17 71 48 52-64 Root Fruity 
Wheat 2-351 15 25 21 41 53-72 Cereals Low 
Oats 2-10 27 56-62 Cereal Low 

a Large starch granules above 5; small starch granules below 5. Small starch granules gelatinize at 75-80°C 
b Mainly 10-15. 
c Some clusters 9-30; some as large as 40. 
d 2-8; up to 35. 
e Mainly 10-12. 
1 2-5, 6-15, and above 15 

8 



Chapter 2: Literature review 9 

Wheat proteins denature during the heat treatment and form complexes 

with starch molecules, preventing the escape of exudates and thus interfering 

with the decrease in viscosity. Gelatinization is also affected by the interactions of 

lipids with starch complexes . The availability of water for starch is affected by the 

presence of pentosan gums, which imbibe a lot of water. Wheat flour contains 2-

3% of pentosan gums, one-fourth of which are water-soluble . They can absorb 

9.2 times their weight of water in a starch-protein mixture (Pomeranz, 1991 ). 

The extent of swelling , the swelling power, of granules is peculiar to 

particular starches. It can be calculated at a pasting temperature as the weight of 

sedimented swollen granules per gram dry starch . This indicates the nature of 

internal bonding . A relative ease in swelling indicates weak internal bonding and 

vice-versa (Swinkels, 1985) . 

The critical concentration value of starch is the weight of starch per 100 

ml , on dry basis , required to produce a paste in which the swollen granules 

occupy the entire volume at 95°C. When the starch concentration is above the 

critical value , the swollen granules will form a continuous paste . Below this value 

there will be some free water left (Swinkels , 1985) . 

2.3 RETROGRADATION 

The phenomenon in which some of the crystalline structure of the starch 

molecule, lost during gelatinization , is restored during storage is known as 

retrogradation . If a dilute starch solution is allowed to stand for a prolonged time , 

it gradually becomes cloudy and eventually deposits an insoluble white 

precipitate. Cooling leads to a rapid elastic gel formation in more concentrated 

systems. This may be accompanied by syneresis , which is the leakage of water 

from the gel along with hardening of the gel , increase in viscosity and turbidity 

development (Swinkels , 1985) . The quality of food texture and other physical 

properties deteriorate making retrogradation an important factor in the inclusion 

of starch as an ingredient (lshiguro et al., 2000) . 



Chapter 2: Literature review JO 

Amylase is predominantly responsible for the retrogradation processes. 

Dissolved amylase molecules can orient themselves in a parallel alignment, so 

that a large number of hydroxyl groups of one chain are in close proximity to 

those in the adjacent chains. This leads to formation of aggregates bound 

together by interchain hydrogen bonding of hydroxyl groups, which are insoluble 

in water (Swinkels, 1985). 

A double helical association of 40-70 glucose units similar to structure to 

native granular starch may form (Ting-Jang et al., 1997). In dilute solution, the 

aggregates of amylose chains precipitate whereas in concentrated dispersions, 

aggregated amylase molecules entrap fluid in the network forming a gel (Figure 

2 .3). To re-dissolve the retrograded amylase a temperature of 100-160°C is 

required (Swinkels, 1985). 

Solution 

Slow I \ Rapid 

Precipitate 
Gel 

Figure 2.3: Schematic representation of retrogradation mechanism. 



Chapter 2: Literature review 11 

Studies of amylose dispersions indicate that amylose gels are composed 

of a network of double helical , semi-crystalline junction zones separated by 

amorphous regions. Ting-Jang et al. (1997) reported that retrograded amylose is 

a mixture of crystalline and amorphous regions and exhibits a 8-type X- ray 

pattern. The crystalline regions are resistant to amylo-lytic and acidic hydrolysis. 

At elevated temperatures the double helix formation is slower and longer chains 

are required to maintain a stable double helical arrangement. Amylose with a 

degree of polymerization (OP) of 80-100 has the highest retrogradation tendency. 

Amylopectin is much less prone to retrogradation than amylose due its 

highly bran ched structure. Therefore , amylopectin tends to be soluble and does 

not form gels under normal conditions and the double helices formed are shorter 

due to the restriction imposed by branching structures (Klucinec & Thompson , 

1999). However, under freezing conditions and high starch concentration the 

molecules may undergo retrogradation. The amylopectin fraction in common 

starch gels has a moderating effect on the retrogradation of linear amylase 

fraction by slowing down the precipitation and diminishing the gel tendencies 

(Swinkels , 1985). 

Normal retrogradation occurs generally on cooling and storage of starch 

pastes at a temperature of 70°C or below. High temperature retrogradation (75-

950C) is observed when corn starch is gelatinized at temperatures of 120-160°C. 

The precipitated particles are formed from inclusion complexes of cornstarch 

amylose with fatty acid , which occurs naturally in corn starch . Normal cereal 

starches contain 0.5 - 1.5% internal lipids (i.e. , lipids extractable only by polar 

solvents such as water containing methanol, butanol , etc.), which remarkably 

influence the retrogradation of starch (Swinkels , 1985). 

However, high temperature retrogradation does not occur when the corn 

starch is defatted . The dissociation of such complexes occurs at temperatures 

above 95°C, as the complex of amylase with fatty acids is not formed above 95°C 

(Swinkels, 1985). Lipids can indirectly affect the behavior of amylopectin towards 

water through complex formation with amylase, which may be associated with 



Chapter 2: Literature review 12 

amylopectin within a starch granule (Hibi et al., 1990). 

Retrogradation is a complex process and its rate depends on various 

factors , such as source of starch , starch concentration , cooking procedure , 

temperature , storage time, pH , cooling procedure, and the presence of other 

compounds (Jacobson et al., 1997) . Starches from different botanical sources 

retrograde differently. This is due to factors like the amylase and amylopectin 

ratio , the molecular weight of amylase and the chain length of amylopectin 

(Swinkels , 1985). lshiguro et al. (2000) noticed rapid retrogradation in gels having 

small amylase molecules and long unit chain amylopectin . Sweet potato starch 

retrograded more slowly than tapioca starch due to its lower amylase content. 

Retrogradation is also affected by the chain length distribution of amylopectin as 

pea amylopectin retrogrades faster than cereal amylopectins , which have shorter 

average chain lengths. 

Retrogradation , determined quantitatively by turbidometric analysis , in 

stored (4°C for 56 days) 2 % pastes prepared by atmospheric cooking under mild 

shear conditions of starches from various botanical sources revealed that 

retrogradation rates followed the order of wheat, common corn > rice , tapioca , 

potato>> waxy maize. Amylase in stored starch systems underwent very obvious 

changes whereas there was little change in amylopectin microstructure over the 

storage period . Solubilized amylase generally co-crystallized or precipitated with 

amylopectin and/or crystallized or precipitated onto the amylopectin-rich granule 

remnants . The latter interactions were weak indicating an interaction between 

amylase and amylopectin to form a network (Jacobson et al., 1997) . . 

Lin et al. (2001) measured the degree of retrogradation of milled rice as 

the melting enthalpy (llH) of recrystallized starch for 30 days. The llH for rice 

varieties with high amylase content increased rapidly in the early stage and then 

reached a plateau . For varieties with intermediate amylase content, the llH did 

not change significantly for the first two days but rose as the storage time 

increased. For waxy varieties, the llH did not change for 9-12 days , and then 

started to increase. The llH for the latter two did not reach a plateau during the 



Chapter 2: Literature review 13 

30 days. 

Fastest rates of retrogradation are observed at a pH of 5-7. The rate slows 

down below pH 2 and ceases above pH 10 (Swinkels , 1985). In dilute amylase 

solution , the rate of retrogradation , measured by decrease in amylase 

concentration , decreased with an increase in incubation temperature (Table 2.2) . 

Different molecular sized amylose fractions have different retrogradation 

tendencies. Small sized amylase molecules such as potato amylose (DP 110) 

have a higher retrogradation tendency. Retrogradation temperature also affects 

the chain length of crystalline regions of retrograded amylase. The crystalline 

chain length increased as the incubation temperature of the retrograded amylase 

increased (Ting-Jang et al., 1997). 

Table 2.2: Total retrogradation of original potato amylose and retrograded 

amylose during incubation at different temperatures1 (From Ting-Jang et al., 1997). 

Temperature (0 C) Total retrogradation (%) 

Original amylose 0 
Retrograded amylose 

5 88.3 ± 4 .5 
15 70 .8 ± 0.9 
25 44 .8 ± 3.2 
35 14.2 ± 2.6 
45 8.6 ± 0.4 

1 Means and tandard dev iation o f duplicate sample fo r tota l retrogradation 
2 Aller 85 days incubation 

The retrogradation of waxy starches is directly proportional to the mole 

fraction of branches with DP 14-24, and inversely proportional to the mole 

fraction of branches with DP 6-9 (Lin et al., 2001) . The polymeric form obtained 

during amylopectin recrystallization depends upon the storage conditions. Low 

water content and/or high temperature conditions favor the formation of the A 

polymorph while high water contents and /or low temperatures lead to polymorph 



Chapter 2: Literature review 14 

B form with intermediate conditions leading to mixed crystals (C type) (Farhat et 

al., 2000) . 

A clear evidence of amylopectin retrogradation was observed in extruded 

potato starch stored at constant moisture conditions for 2 and 14 days at three 

different temperatures (22 , 40 , 60°C) . The material stored at 22°C retrograded to 

B form while the one stored at 60°C gave a polymorph of type A. An intermediate 

of pure A and B polymorph was observed at a storage temperature of 40°C 

(Farhat et al., 2000) . Type-A polymorphs are known to have denser unit cells and 

contain very few bound water molecules as compared with polymorph B (Shamai 

et al., 2003) . The digestibility of the extruded material decreased on storage as a 

result of retrogradation depending on storage time and temperature (Table 2.3) . 

The results suggest a higher resistance to digestion in polymorph A than 

polymorph B (Farhat et al., 2000) . 

Table 2.3: PA1 digestibility(% starch, 6 h incubation) of potato starch extruded at 

35% water (w/w, wb) and stored at different temperatures. The digestibility of the 

freshly extruded material was 77.0% ± 0.92 (From Farhat et al., 2001). 

20 °C 

2 days 45.8 ± 1.6 
14days 42 .1 ±0.3 

1 Porcine Pancreatic a amylase 
~Results are means± standard dev iation. 

40 °c 

37 .8 ± 0.3 
35.4 ± 1.1 

60 °C 

32 .9 ± 1.5 
28 .0±1 .7 

The phosphorous in starch is mainly present in the form of phospholipids. 

Generally root starches contain low amounts of phosphorous compounds. Potato 

starch is the only commercial starch that has appreciable amounts of chemically 

bound phosphate ester groups (Swinkels, 1985). The effect of phosphate 

covalently linked to the starch is less known . Starch phosphorylation increases 

gel hydration and peak viscosity. However, potato starch with low bound 

phosphate shows a reduced peak viscosity, stronger gel , increased stickiness 



Chapter 2: Literature review 15 

and increased turbidity indicating increased retrogradation behavior. However, 

high amylose content, short chain amylopectin and high phosphate content 

suppress retrogradation (Thygesen et al., 2003) . 

2.4 Rheological properties of starch dispersions 

In most food products, starch is used as a texturing ingredient, and the 

texture depends on the concentration of the starch . Low levels are usually 

employed to impart thickness to the product whereas if it is the main component 

responsible for texture , the levels are relatively higher. Because of settling of the 

ungelatinized granules in rheological studies on starch dispersions of low 

concentrations , considerable heed should be exercised to ensure that the 

samples taken from the dispersion contain the same concentration of starch 

granules. Similar precautions are necessary in vertically oriented viscometer 

geometries like concentric cylinder and vertical capillary/ tube (Rao , 1999) . 

Starch dispersions, after gelatinization , have a continuous biopolymer 

matrix conta ini ng dispersed granule fragments along with swollen starch 

granules. Accordingly , rheology of gelatinized starch suspensions reflects the 

properties of the dispersed phase , the continuous phase, and the interactions 

between the components . The viscosity versus time profile is important for most 

food products containing starches (Rao, 1999) . 

Starch dispersions undergo crucial changes while being heat treated in 

heat exchangers. Significant changes in velocity profile occur resulting in 

pressure drops inside the process equipment. Shear rates ranging from 10 to 103 

s-1 are encountered during most thermal processes carried out using plate heat 

exchangers or scraped surface heat exchangers. The flow regimes in heat 

exchangers depend heavily on the preparation procedure and the difference in 

viscosity of the product. Thus, for product development and process control 

strategies, modeling and rheological data for the starch dispersion are critical 

because of the effect on the process and the end product (Lagerrigue and 

Alvarez , 2000). 



Chapter 2: Literature review 16 

Extreme structural modifications take place during gelatinization, thermo­

mechanical treatment and retrogradation of starches. These mainly result in the 

dramatic changes in apparent viscosity. During gelatinization, the starch granules 

swell irreversibly producing an increase in viscosity and an eventual leaching of 

amylase (linear component of starch granule). As soon as the granules start to 

rupture, due to shear, the viscosity starts to decease. The solubilized starch 

polymers and remaining starch granules re-associate in an ordered structure 

upon cooling during retrogradation (Lagerrigue and Alvarez, 2000). 

Gelatinized starch dispersions exhibit a non-Newtonian, time-dependent 

and viscoelastic behavior. Their rheological properties revolve around the 

procedure adopted for gelatinization. Table 2.4 summarizes the rheological 

behavior of gelatinized starch dispersions found by different researchers for 

measurement conditions above and below 100°C. Minor variations in the 

procedure adopted for gelatinization were present along with differences in 

composition cause differences in viscosity (Lagerrigue and Alvarez, 2000). 



Table 2.4: Modeling of rheological behavior of gelatinized starch suspensions: (a) T < 100 °C; (b) T > 100°C (From: 

Lagerrigue and Alvarez, 2001 ). 

Authors Starch type and Viscometer Gelatinization procedure Measuring condition Modelling 
concentration (w/w) 

a) T < 100 °C 
Corn, potato , tapioca Haake rotovisco RVI, Heating at a fixed T in rotating T = 60°C; flow Power law model , Herschel Bulkley 

Evans and and modified corn (0.5· Weissenberg flasks curves :0.0007·56 s·' and model 
Haisman (1979) 10%) Rheogonimeter R16 (cone 7-1142 s 1 

and plate) 
Bagley and Wheat (7·25%) Haake rotovisco coaxial Corn industries research T stud ied : 60°C and 23 Dilatancy if cooking 15· 45 min at 60 
Christianson cylinders (MV system cup, viscometer cooking T: 60· 75°C °C; fl ow curves 1 · 1000 s °C, master curve 11/cO function of cO 
(1982) MV·II bob) cooking t : 15-75 min 1 at a60 °C 
Christianson Corn (8·24%) Haake rotovisco coaxial Corn industries research T studied: 60°C and 23 Herschel Bulkley model yield stress 
and Bagley cylinders (MV system cup, viscometer cooking T: 60·75°C °C; fl ow curves 3·500 s ' at 23°C is a function of cO 
(1984) MV-11 bob) cooking t : 15-75 min 
Doublier et al. Wheat and maize (3· Rheomat-30 coaxial cyl inders Double-walled round bottom T = 70°C up and down Herschel Bulkley model K, n and a0 

(1987) 10.5) (A-system) vesse l. stirring : 200 or750 rev. curves: \1) 0·6 .6 s ', (2) depends on the gelatinization 
min 1 0·660 s , (3) from 660 s procedure, thi xotropy 
Cooking T: up to 96 °C, cook ing ' to O constant shear 
t: up to 30 min: heating rate 1 ° rate experiments, 
C min ' or exp. rate 

Harrod Cross-linked and Rheomat-30 coaxial cylinders Bra bender viscograph scraped T = 10·90°C up and Arrhenius law, power law, thixotropy 
(1989a,b,c) eslerified potato (3·10%) surface heat exchanger sown step wise curves: 

6-450 s 1 

Lagarrigue et Potato (3-6%) Tube viscometer D=23mm; Steam-jacketed vessel cooking T = 90°C; flow curves: Power law model . Bingham plastic 
al. (1990) L=10.5m; q=0 .05·0.5 kg s ' T: 90°C 40·400 1 model 
Dolan and Corn (5.5·7.3%) and Brookfield RVTD mixer Brookfield RVTD mixer T = 50·95 °C; impeller Shear ·temperature - time model 
Steffe (1990) bean (6%) viscometer viscometer cook ing T: 85·90°C, speed: 20·100 rpm 

cooking t: 2·25 min 
Ramaswamy et Cross-linked waxy Hake RV20 Pregelatinized at 85·95 °C (2h) T = 25°C up and down Power law model, Herschel Bulkley 
al. (1995) maize (3-4 %) in a steam jacketed kettle curves :0·200 s ' model Casson model 

thermal processing in an 
agitated retort (110·130°C; 10· 
20 rpm) 

Okechukwu and Corn (2.6%) Carri-med CLS 100 cone· Isothermal healing in jacketed T = 20 °C; v= 0.05· 1200 Power law model (20- 1200 s'): n 
Rao (1995) plate (6 cm, 2°C) stainless steel vessel cooking T s ' function of the standard deviation in 

: 70·90 °C; cooking t:0·540 min mean granule size, K and Casson 
yield stress increase with mean 
granule diameter, dilatant behaviour 
for cookinq at 67 °C 

17 



Okechukwu and Rao (1996 Cowpea (2.6 %) Carri-med CLS 100 cone- Isothermal hating in T = 20°c. v= 0.05-1300 s ' Power law model: k increase 
a,c) plate (6 cm, 2°C) iacketed stainless exponentially with mean granule 

steel vessel cooking diameter, n decrease linearly with 
T 67-86°C: cooking t: increasing extend of gelatinization 
0-2880 min dilatancy at early stage of 

gelatinization 
Bhattacharya and Debranned maize Rheotest -2 coaxial cylinders Brabender viscograph T = 60°C: flow curves: 3-1326 Herschel Bulkley mode, Mizrahi-
Bhattacharya (1996) flour (2-10%) s' berk model, K, a0 increases with c, n 

decreases with c 
Breton- Dollet (1996) Maize and waxy Carri-med CLS 100 cone- Autoclave cooking T: T = 60°C: up an down curves: Power law model, K function of 

maize (5%) plate (6 cm, 4°C) 100-136°C: cooking t: 0.1- 100 s cooking T, n increases with cooking 
30 min T thixotropy depends on the kind of 

starch, c, cooking T 
Nguyen et al. (1998) Normal maize and Haake rotovisco RV 100 Cooking T 50-75°C T = 25- 50°C, transient Power law model: K (c,T) n is 

waxy maize (6-7%) coaxial cylinders (waxy) or 50-85°C experiments at constant y(90- constant 0.48, structural kinetic 
(regular): Mating rate: 225 s '), flow properties at model for thixotropy (which depends 
2°Cmim 1 

steady state on y, c, and T) 
Xu and Raphaelides (1998) Maize (7-30%) TR-1 tube rheometer Gelatinization at 95 T=75-95 °C; 10-1000 s 1 Power law model, n increase with 

diameter: 2.05 mm; length: °C under continuous cooking time 
30mm stirnng 

Nurul et al. (1999) Sago (3-5.5%) Contraves rheomat 115 Autoclave cooking T: T= 40-80 °C; v= 13.64-704 s 1 Power law model: exponential 
25°C cooking t: 30 dependence with c. Arrhenius law 

m1n 
(b)T>100°C 
Dial and Steffe Waxy maize (1.82 Tube viscometer D=1.27 cm: Gelatinization in a T studied: 121. 1 132.2- Power law model k depends on c 

2.72%) L= 4.59 m tubular heat 143.3°C: flow curves 10-150 and T, n increases with c and 
exchanger s' decreases with T dilatant behaviour 

Abdelrahim et al. (1995) Cross-linked waxy Haake RV 20 equipped with a Gelatinization at 140 T studied: 60-140°C flow Power law model, modelling of K 
maize (3-6%) high Tihigh P attacl1ment and 0 c in an aseptic curves: 0-500 s 1

; constant y and n as a function of C and T 
a magnetic coupling coaxial processing system experiment 
cylinders 

Heydon et al. (1996) Colflo 67 and C Flo Tube viscometer Gelatinization in a T studied 100-135°C: 10-70 s Arrhenius law, power law, 
(4-7%) D= 3.46-2.17 am: scraped surface heat 

2.85-32 m exchanger 
0= 0.5-3.1 min 1 

18 



Chapter 2: Literature review 19 

Predominantly, power law or Herschel- Berkley models are used to 

represent gelatinized starch dispersions in the shear rate range of 1-1500 s-1 . 

However, other models such as Bingham plastic, Casson and Mirzrahi-Berk have 

been tested as well. These are illustrated below. 

O=Kt 

o- Oo = K t 

( Power law Model) 

(Herschel-Bulkley Model) 

(Casson Model) 

Here, o , oo, o0 05 are shear stress, yield stress and its square root , K is 

consistency index, y is shear rate, n is flow behavior index, 

Pre-gelatinization starch suspensions can show dilatant behavior . After 

gelatinization shear-thinning behavior is most often observed. The consistency 

index (K) and flow behavior index (n) obtained from the power law model depend 

upon the starch concentration and the processing temperature . Over the 

gelatinization temperature range , the consistency index (K) is also dependant on 

temperature and follows the Arrhenius law (Lagerrigue and Alvarez, 2000). 

K a exp ( Ea/RT) 

Here, Ea is the activation energy in kJ mol- 1 

Flow behavior index (n) has also been found to depend on concentration 

and temperature. A dependence of K and n on granule size has also been 

suggested (Lagerrigue and Alvarez, 2000) . 



Chapter 2: Literature review 20 

2.5 Resistant Starch 

2.5.1 History 

Resistant starches were originally associated with raw starch with an X-ray 

diffraction pattern of Type B (raw banana and potato starch) as starch granules 

had been found in faeces of animals and human subjects. They were seen to 

decrease the energy value of starchy products . The medical and physiological 

interest in RS was boosted only after the identification of a specific fermentation 

profile (rich in butyrate) and the first studies on the effect of butyrate on cell 

prol iferation and differentiation (Champ et al. , 2003). 

The quantitative importance of the undigested fraction of starch was 

recognized , at first , by the laboratories of Cummings and Stephen (1983). 

Sandberg et al. (1981 ) studied the intestinal digestibility of ileostomy patients 

(patients without a colon ) and made in viva measurements. By doing this one 

could measure the RS content without administering invasive techniques . 

Besides th is, intubation and perfusion techniques were developed for both small 

as well as the large intestine. The European Flair-Concerted Action Research 

Programme (EURESTA) , which started in 1990, has since then instigated further 

research on resistant starch in Europe as well as in other parts of the world 

(Champ et al., 2003) . 

2.5.2 Relationship with Glycaemic Index 

An important number of studies in the past decade suggest a beneficial 

effect of low Glycaemic index (GI) diets in relation to insulin resistance syndrome . 

A low GI diet not only improves metabolic aftermaths of insulin resistance but 

also causes a reduction in insulin resistance per se (Bjorck et al., 2003) . Rapidly 

absorbed starch produces high blood glucose and insulin levels after meals . Free 

fatty acids thus released from the liver cause insulin resistance at the next meal , 

which is considered as a potential risk factor for the development of insulin 

resistance syndrome e.g . diabetes, atherosclerosis and obesity (Akerberg at al. , 

1998). 



Chapter 2: Literature review 2 1 

However, food products rich in RS consistently give low GI and 

insulinaemic responses (Akenberg et al., 1998) . This is due to the indigestibility 

of RS in the small intestine , which prevents any contribution to postprandial 

glucose response . With addition of RS to meal there is no alteration in 

postprandial glucose. Propionic acid , a fermentation product of RS, has been 

shown to be involved in the metabolism as a moderator of hepatic glucose 

( Bjorck et al., 2003). 

Bjorck et al. , (2003) reported an improved insulin economy with a modified 

low GI diet compared to the usual high-GI one in women at risk to Type-2 

diabetes. However, no correlation could be established between a low - GI diet 

and the postponement of Type-2 diabetes . 

2.5.3 Resistant Starch as a Functional Food Ingredient 

Uses of starch in the food industry are manifold , such as for bulking , 

functionality and texture and for nutritional quality as well. Therefore, its 

modification can easily be used in the field of disease prevention . In the UK, the 

daily intake of RS by an individual is estimated to be about 3g , which is quite low 

(Niba et al., 2002) . RS can be used in functional-food product development. By 

definition , functional foods are " foods similar in appearance to conventional foods 

that are consumed as a part of the normal diet and have demonstrated 

physiological benefits and/or reduce the risk of chronic disease beyond basic 

nutritional functions" (German et al., 1999) . 

According to Niba et al. (2002) , resistant starch can be exploited for the 

development of functional foods in following ways : 

• Convert foods, with high RS, into more palatable and appealing forms ; 

• Increasing the level of RS in foods by various processing techniques ; 

• Developing processing techniques that ensure stabilization of resistant 

starch levels in foods. 



Chapter 2: Literature review 22 

Whole grains are considered as a rich source of RS and other fermentable 

carbohydrates as these escape digestion. Making these rich repositories of RS 

into palatable food material by appropriate processing techniques improves their 

usefulness for functional foods. Several methods have been suggested that can 

potentially enhance the RS content in foods such as high-pressure autoclaving 

(Escarpa et al., 1996), enzymatic debranching (Shi et al., 2003), partial 

degradation (Schmied! et al., 2000) etc. 

Hitherto, most macronutrients including RS are difficult to quantify in viva. 

In spite of that, there are a number of in vitro procedures, which provide an 

estimate of the physiologically available levels of RS (Englyst et al., 1999). 

Therefore, physiologically relevant levels of resistant starch can be determined 

and applied to the production of functional food products. 

2.5.4 Classification of Resistant Starch 

Until recently, only three classes of RS were described in the literature but 

a fourth class has also been identified (Englyst et al. 1992, 1996). The four 

classes of RS are: physically inaccessible starch (RS1 ); RS granules (RS2); 

retrograded starch (RS3); chemically modified starch (RS4). 

Physically inaccessible starch 

RS1 is found in partly milled grains and seeds. Other main sources of RS1 

are legumes such as beans or lentils where the starch is entrapped in a cellular 

matrix. These escape digestion due to the thickness of the cell walls. The 

preparation and cooking process is vital as these can disrupt the cell wall, 

lowering the resistance to digestion (Champ et al., 2003). 

RS granules 

RS2 includes Type-B starches, such as raw potato and banana starch, 

which are known to be acutely resistant to enzymic hydrolysis when uncooked. 

Type-B refers to the X-ray diffraction pattern of the starch. Most raw starch in 

food gets gelatinized eventually leading to the disappearance of RS2. Banana is 

thought of as the main source of RS2 in the human diet, but the content in the 



Chapter 2: Literature review 23 

fruit depends on the degree of ripening (Champ et al., 2003). 

Retrograded starch 

RS3 is present in most starchy foods after they have been hydrothermally 

treated . After gelatinisation , if the starch gels are allowed to cool , recrystallization 

occurs often forming double helices. Amylopectin , the branched fraction takes a 

longer time to retrograde than amylose , which is linear. Cooked and cooled 

potatoes have high RS3 mainly due to recrystalization . Retrogradation is partly 

reversible as reheating of starch reduces the RS3 content of the potato . Several 

cycles of heating and cool ing , however, allow an increase in the RS3 levels. RS1 , 

RS2 and/or RS3 can co-exist in the same food e.g. a meal of beans (Champ et 

al., 2003) . 

Chemically modified starch 

RS4 , wh ich has been recently established , includes starch ethers and 

esters as well as cross-bonded starches (Champ et al., 2003). 

2.5.5 Digestion and Fermentation of RS 

By definition , RS is the fraction of starch that escapes digestion (Langkilde 

et al., 1987). Therefore , it has been defined exclusively in terms of large bowel , 

as the rate of digestion in the small intestine is not relevant. Raw starch is 

digested poorly but cooking in presence of water enhances digestibility . While 

gelatinization increases digestibility , subsequent retrogradation increases 

resistance to digestion (Botham et al., 1995) . Another important factor affecting 

starch digestibility is the amylose : amylopectin ratio. Higher amylase content 

increases resistance to gelatinization and also makes the starch more 

susceptible to retrogradation (Topping et al., 2003a). 

Some physiological factors also influence the digestibility of starch that 

varies between individuals. Factors relating to gender, e.g. the female menstrual 

cycle , influence the RS uptake greatly with the uptake increasing in the mid­

cycle . Also , particle size plays a crucial role. Large particle sizes have faster 

transit time and allow less access to digestive enzymes. Therefore , it is more 



Chapter 2: Literature review 24 

likely that people, who chew their food enough , would get less RS from a high RS 

diet as compared to the people who masticate sparingly. This is thought to be the 

biggest difference between resistant starch and non-starch polysaccharide . RS is 

a physiological outcome whereas the later is characterized by its chemical 

structure (Topping et al., 2003b). 

The digestion process does not end when food enters the large bowel. 

However, the main difference between the large and the small intestine is that 

the breakdown in the former is affected by bacterial enzymes rather than human 

ones. It is this bacterial system in the large bowel, which is responsible for 

metabolizing the undigested components in diet. RS is largely fermented , 

producing Short Chain Fatty Acids (SCFA) and bacterial cells . The molar quantity 

of SCFA produced depends on the accessibility of RS to the colonic microflora . 

Hitherto , there exists poor understanding of this complex bacterial system mainly 

due to limitations in technology as the present methods are labor intensive and 

consume enormous time. The net reaction of bacterial fermentation in adults is 

given below (Topping et al., 2003a) . 

Carbohydrates + H20 ----+ Acetate + Propionate + Butyrate +C02 + CH4 + H+ + 
heat + more bacteria 

In humans, the principal SCFA are acetate, propionate and butyrate. They 

are useful metabolically and are readily absorbed from the large bowel lumen 

and used by the viscera for salt and water uptake (Henningsson et al., 2002). 

The fermentability of RS is generally very high which makes them relatively poor 

laxatives. Thus, the effects of RS are mediated through their metabolic products 

rather than their physical presence so that SCFA become one of the key 

biomarkers for RS action . Fermentation is high in the proximal large bowel and 

so is the absorption of SCFA (Topping et al., 2003a) . However, as the faecal 

stream passes the fermentation slows down due to substrate depletion (Figure 

2.4) . 



Chapter 2: Literature review 

Small 
intestine 

Digestion Other 

:::====~> !at l'rntc111 Cll(l nutncnb 

hll 
ac1J\ xH:ch nutrll'l1h 

-aridcs 

Absorption 

Large 
intestine 

Starch l nahsorhed Sccrct1on, 
!l1gh 
ferrnentat 10n 

lo\\ 
fermentatH_)n 

:::===========:::::=-
()II gosacc har1des 

SCF.\ 
\b,orrtl()Il 
grad1cnh 

Faeces 

\' 01dcd undigested 
carboh~drme. !1gnin 

b1orrn.1'.-.s and un<.1h\orbed 

nutrient 

:::=======> 

25 

Figure 2.4: An overview of the transit of food through the small intestine and the 

large bowel (From Topping et al., 2003a). 

Of the principal SCFA, acetate does not have any specific actions in the 

large bowel and is largely transported to the liver for oxidative metabolism. RS 

fermentation by faecal bacteria leads preferentially to butyrate and propionate 

production (Brous et al., 2002). Le Blay et al. (1999) reported that long-term 

intake of resistant starch for 2 to 6 months, derived from potato, increased 

butyrate production throughout the colon and also decreased the pH, effects 

which are considered beneficial. However, no key butyrate producing bacteria 

have been identified so far in human pilot studies (Schwiertz et al., 2002). 

Butyrate is thought to be pivotal for human colonic function as it 

contributes to the maturation of colonic epithelium. It has an anti proliferative 

effect on tumor cells and also promotes apoptosis (programmed cell death) in 

vitro in cancer cell lines (Brouns et al., 1998). A study showed complete tumor 

regression by apoptosis probably induced by butyrate while testing the efficacy of 

sodium butyrate (NaB) and immune-factor interleukin 2 (IL- 2) against 



Chapter 2: Literature review 26 

experimental carcinomatosis induced in rats . Evidence based on these 

experimental observations suggests that butyrate plays a controlling role in cola­

rectal cancer but such a link remains to be established in humans (Brouns et al., 

2002 ; Topping et al., 2003b) . However, Sakamoto et al. (1996) reported an 

increase in the butyrate production in rats with consumption of RS (raw potato 

starch) rich diet but this had no inhibiting effect on the dimethylhydrazine induced 

colonic carcinogenesis . Therefore , the literature on the effect of RS on colon 

cancer is contradictory (Champ et al., 2003) . 

Many physiological attributes of butyrate are shared by propionate but at 

higher concentrations (Topping et al., 2003a) . Propionate also affects metabolism 

in peripheral tissue and its inhibitory action on hepatic cholesterol has also been 

proposed but not yet consistently shown in vitro (Henningsson et al., 2002) . 

RS reduced body fat in rats and also increased fermentation in the lower 

GI tract , which is beneficial for colonic health (Hegsted et al., 2003) . However, 

this effect has not been confirmed in normolipidemic human subjects (Haralampu 

et al., 2000) . 

The use of RS in probiotic compositions has been suggested as it 

promotes the growth of beneficial microorganisms such as Bifidobacterium 

(Haralampu et al., 2000) . Brown et al. (1997) reported an increase in the faecal 

concentration and excretion of Bifidobacterium longum after oral ingestion of RS 

as high amylase starch compared to those consuming conventional starch . In 

comparison to other Prebiotics, RS from high amylase starch yielded similar 

results in humans . However, pigs fed with rice baby food , with higher RS than 

conventional foods , did not show significant increase of ingested probiotics in 

faeces. 

Ramakrishna et al. (2000) reported a major reduction in fluid loss and the 

halving of recovery time in children suffering from cholera induced diarrhoea after 

consuming RS (high amylase starch) along with the usual hydration therapy . This 

is due to greater fluid uptake as a result of increased SCFA production in the 

proximal colon. 



Chapter 2: Literature review 27 

2.5.6 Analysis of RS in Foods 

The inclusion of RS in functional foods has led to the need for a valid 

analytical method for its appropriate quantification. Asp (1992) defined RS as 'the 

sum of starch and products of starch degradation that are not absorbed in the 

small intestine of healthy individuals. Thus , the analytical method should be 

applied to all the starch and a-dextrin present to quantify the RS content of foods . 

Besides, the ratification of the analytical method by a direct comparison of the 

data obtained in vitro with the true in vivo data from healthy subjects is important 

(Champ et al., 2003). 

A formalization of the physiological definition of RS seems to be 

particularly difficult for several reasons even though it would be conceptually 

most gratifying . Firstly , differences like the structural organization of the starch 

and/or the functional and physiological environment during the process of 

digestion exist between starch and RS . These factors are known to affect the 

digestive enzyme accessibility to the RS substrate. Others factors like the 

efficacy of chewing , the gastrointestinal transit time and the quantitative enzyme 

secretion also effect the starch digestion . All these factors may vary from one 

subject to another. These differences between digestible starch and RS are 

difficult to incorporate in an analytical method (Champ et al., 2003) 

Therefore , the in vitro method of choice has to provide results in contrast 

to the average response of a population of healthy individuals. Ideally, a large 

range of RS sources should be used to ratify the in vitro method (Champ et al., 

2003) . 

2.5.6.1 Current in vitro methods 

The main step for the quantification of RS is to remove digestible starch 

from the sample by using a- amylase. To avoid a possible inhibition of the a­

amylase by the products of the digestion (mainly maltose and maltotriose) , some 

methods use amyloglucosidase. Proteolysis may precede amylolysis in order to 

reflect the action of pepsin and trypsin in the stomach secreted in the pancreatic 

juice along with a- amylase . RS is quantified after the digestion either from the 



Chapter 2: Literature review 28 

residue (isolated by ethanolic precipitation) or by calculating the difference 

between total starch and digestible starch (Champ et al., 2003) . 

Bjorck et al. ( 1986) used the official AOAC method for estimation of dietary 

fibre to quantify total RS, as there was no regulatory definition of RS at that time. 

The residual dietary fibre was acquired after enzymatic solubilization (bacterial a­

amylase {Termamyl} treatment at 95-100°C, according to Asp et al. (1983) and 

Prosky et al. (1988)). The amylase digestion was conducted at 95-100°C, which 

melted down some of the RS structures leading to an underestimation of the total 

RS . Therefore, to correct the method porcine pancreatin at 37°C was included 

(Haralampu, 2000). 

The estimation of Rapidly Digestible Starch (RDS), Slowly Digestible 

Starch (SOS) and RS was first proposed by Englyst et al. (1992) . The enzymatic 

digestion was accomplished at 40°C , which was close to that in the in vivo 

process, using protease , amylase and amyloglucosidase. A schematic of the 

method is given in Figure 2.5. 

Berry's (1986) method was based on the Total dietary fibre (TDF) analysis 

methodology with remodelled scheme to improve the RS quantification. Goni et 

al. (1995) modified this method by including a proteolysis step before digestion , 

which was missing in Berry's (1986) method. In this method ethanol precipitation 

and acetone washing drying were also omitted from the Berry's ( 1986) method . 

Ethanol precipitation was thought of as time consuming and non-physiological 

and drying affected the RS values. 



Chapter 2: Literature review 29 

Sample preparation 
Mill/homogenize 

Extract lipid 
Analyse free glucose 

! 
r 

Add 
Guar gum and 

Glucose standard 

! 
Enzymatic digestion 

Sodium acetate buffer 
Amyloglucosidase Rapidly digested 

Pancreatin starch 
lnvertase (RDS) 

37 °C, 20 min 

i 
Enzymatic digestion 

Sodium acetate buffer Slowly digested 
Amyloglucosidase 

starch 
Pancreatin (SOS) 
lnvertase 

37 °C, 100 min 

! 
Dissolve resistant starch 

Concentrated KOH 
0°C, 15 mins 

! 
r 

Starch Digestion Resistant starch 
Amyloglucosidase (RS) 

70°C, 30 min 

'" ~ 

Figure 2.5: Englyst et al. (1992) method for resistant starch analysis (Haralampu, 

2000). 



Chapter 2: Literature review 30 

Another method similar to the Berry (1986) method was proposed by 

Champ et al. (1999). The main modification to the Berry's (1986) method was the 

addition of amyloglucosidase. This method was derived from the one published 

earlier by the same author in 1992. A summary of the most recent methods is 

presented in Table 2.5. Mccleary & Monaghan (2002) studied the strengths and 

weaknesses of the methods present in earlier literature (Englyst et al. 1992; Muir 

& O'Dea, 1992, 1993; Gani et al. 1996; Akerberg et al. 1998; Champ et al. 1999) 

and used the RS data derived from ileostomy model. 

The following parameters were studied systematically: 

• concentration of pancreatic a-amylase; 

• need for pepsin pre-treatment; 

• pH of incubation; 

• importance of maltose inhibition of a-amylase; 

• need for amyloglucosidase inclusion; 

• effect of shaking and stirring on obtained RS values; 

• problems in recovering and analysing the RS-containing pellets. 



Table 2.5: Main in vitro methods to quantify resistant starch (RS) (Champ et al. 2003) 

Bjbrck et al. (1986) Englyst et al. (1992) Muir& O'Dea (1992, Gani et al. (1996) Akerberg et al. (1998) Champ et al. (1999a) 
1993 

Sample size 100 mg fibre residue 0·8-4·0 g depending About 0.1 g 100 mg dry sample 1 g total starch basis 50 mg total starch 
(Asp et al. 1983 or on the water and carbohydrate basis basis 
Prosky et al. starch content of the 
1988(AOAC method) sample 

Sample pre- Minced (9 mm O holes) Chewing Dry samples, Chewing (15 X, 15 s) Minced (9 mm 0 holes) 
treatment milled(o 1 mm); fresh 

samples. homogenized 
Protein hydrolysis Pepsin pH 1 ·5, 1 h then Pancreatin (see ·starch Pepsin treatment (pH Pepsin treatment (pH Pepsin treatment (pH No protein hydrolysis 

pancreatin or bacterial hydrolysis') 2, 37 C. 30 min) 1 ·5,40 C, 60 min) 1.5, 37 C, 30 min) 
protease pH 7.5 at 
60'C, both within OF 
analysis 

Digestible starch Gelatinization step Pancreatin + a-Amylase (Speedase Pancreatic a- Pancreatin + Pancreatic a-amylase+ 
hydrolysis at100°c + heat- amyloglucosidase + PNA-8) + amylase(pH 6·9, 37 C. amyloglucosidase (pH amyloglucosidase 

resistant amylase then invertase + glass amyloglucosidase (pH 16 h) 50,40C,16h) (37C, 16 h) 
pancreatin or balls+ guar gum (pH 50.37C.15h) 
amyloglucosidase both 5.2, 37'C). Time: 20 
within OF analysis min =>RDS Samples 

collected 
ett1anol (64-4 ~o 1n 
final concentration) 

Removal of starch Ethanolic precipitation Ethanolic precipitation. No ethanolic No ethanolic Precipitation, 76 Precipitation, 80 
hydrolysis products filtration using celite as centrifugation (1 O min. precipitation. precipitation. %EtOH, filtration %EtOH 

filter aid, within OF 3000 g) centrifugation (10 min. centrifugation (15 min, centrifugation 
analysis 1200g) 3000g) 

Dispersion of RS Boiling (20 min) + 2 M- No Boiling (20 min) 2-M-KOH (room 2-M-KOH (30 min Boiling (20 min) + 2 M-
KOH(room temperature. 30 min) KOH (O"C. 30 min) 
temperature, 30 min) 

RS hydrolysis Amyloglucosidase (pH No Amyloglucosidase Amyloglucosidase (pH Termamyl + amyloglucosidase (14 
4·75, 60'C, 30 min) enzymic GOD- Amyloglucosidase units ml; 70"C, 30 min 

PAP4 75, 60 C. 45 + 100'C, 10 min) 
min) 

Glucose Enzymic, GOD-POD Enzymic, GOD-PAP Enzymic, GOD-PAP Enzymic, GOD-PAP Enzymic, GOD-POD Enzymic, GOD-PAP 
detemiination 
Validation In vivo data obtained In vivo ileostomy data In vivo ileostomy data No In vivo ileostomy data In vivo ileostomy and 

with antibiotic-treated (mostly from literature) intubation data 
rats 

Specificity RS = TS fRDS+SDS) 

31 



Chapter 2: Literature review 32 

Consequentially , the predigestion step with pepsin was omitted . The 

refined procedure contained pancreatic a-amylase and amyloglucosidase acting 

together at pH 6·0 under defined shaking conditions followed by alcohol 

precipitation. RS was hydrolysed by amyloglucosidase after dissolution in 2 M­

KOH and glucose was measured by using the GOD-POD reagent (glucose 

oxidase-peroxidase reagent; Megazyme International Ireland Ltd , Wicklow, 

Republic of Ireland) . This method most resembles the Champ et al. (1999) 

method (Champ et al. , 2003) . Various in vitro methods are compared with in viva 

data using ileostomy method in Table 2.6 (Champ et al., 2003) . 

Table 2.6: Comparison of RS (as percentage total starch) determined in vivo and in 

vitro (From Champ et al., 2003). 

Source of In vitro RS 
starch Englyst Faisant Champ et Goni et McCleary In vivo RS 

et al. et al. al. al. (1996) & 
(1992) (1995a) (1999a) Monaghan 

(2002 
Potato 66 .5 83 .0 77 .7 77 .0tt 78·8t 

starch , raw 
HACS, raw 71.4 72 .2 52 .8 51 .7tt 50 ·3t 

HACS, 30.5 36.4 29.6 42 .0tt 30·1t 
retrograded 
Bean flakes 10.6 12.4 11 .2 14.3tt 9-10·9+ 
Cornflakes 3.9 4 .9 4.3 4.0tt 3·1-5·0§ 

Beans 17.1 17.1 16.5tt 16·511 
C* Actistar, 59 ·3t 

Retrograded 63.0 57 .0 57 .0tt 58.0tt 78 ·8t 
HAGS, high-amylose maize starch . 
tlleostomy model ; AM Langki lde, H Andersson and F Bouns (personal commun ication). 
:j:lleostomy model ; Schweizer et al. (1988). 
§lleostomy model ; Muir & O'Dea (1993) and Englyst et al. (1992). 
II Intubation technique ; Noah et al. (1998). 
,Jlntubation technique; analysis by Dr Kettlitz , Cerestar Research and Development Centre , 
Vilvoorde , Belgium . 
ttlntubation technique; presented at AACC Meeting in Montreal 13-17 October 2002 . 



Chapter 2: Literature review 33 

2.5.6.2 Possibilities of validation in vivo 

At present, three prospective methods are available to obtain the in viva 

values on the RS content of foods that are required for a ratification of the in vitro 

methods. These methods are discussed below. 

Hydrogen breath test 

One of the end products of carbohydrate fermentation is H2. It is 

exclusively formed in the colon by bacterial fermentation , partly absorbed and 

cleared in a single passage of the lungs to be excreted in the expired air. The 

gas perfusion technique assumes that H2 production is proportional to the rate of 

breath H2 excretion . To quantify the mal-absorbed carbohydrate , a non­

absorbable but qu ickly fermented oligosaccharide (lactulose) is used to 'calibrate ' 

the subject. The area under the curve after the test meal is plotted together with 

the area under the curve after the intake of lactulose value , which gives the data . 

The amount of carbohydrate from the experimental meal can be calculated by 

knowing the amount of fermented lactulose . Rumessen (1992) proposed several 

ways of quantify ing the data . Although th is procedure was quanti ta ti ve fo r 

oligosaccharides, it was only qualitative for insoluble or slowly fermented 

substrates (Champ et al., 2003). 

lleostomy model 

The ileostomy model can be performed over those subjects who have had 

a conventional ileostomy after colectomy . By minimising the bacterial 

degradation , direct and quantitative determination of small -bowel excretion can 

be obtained . The effluent is usually collected in a bag , which is changed every 2 

hr during the day. The ileostomy bags are immediately deep-frozen on solid 

carbon dioxide. During the experimental period the subjects are given a plant 

polysaccharide-free diet with addition of the RS source under study (Champ et 

al., 2003) . 

Intubation technique 

Intubation technique uses a triple-lumen polyvinyl tube , which is passed 

through the gut with the help of a terminal inflatable bag conta ining Hg . The bag 



Chapter 2: Literature review 34 

is deflated when it reaches the caecum, which is confirmed fluoroscopically, and 

the subjects have to remain in a semi-supine position. The sample used in 

perfusion is drawn from 50mm above the ileocaecal junction in one lumen, and 

250mm proximal to the aspiration port in the other. NaCl and polyethylene glycol 

4000 are the main components of the perfusate used to mark the recovery so as 

to estimate water flow through the distal ileum. Throughout the experiment the 

tube is positioned in the same place and is confirmed fluoroscopically (Champ et 

al., 2003). The advantages and shortcomings of the three methods for the in viva 

validation of RS content are summarized in Table 2. 7. 

lleostomy model v. intubation technique 

Even though it is difficult to compare two methods, which use different 

type of meals as sample and have totally different data, one comparison has 

been made. The ileostomy model and the ileal intubation method have been 

compared with the same meal containing 16·3 g RS (30 g) from green banana. 

The ileostomates excreted 15·8 (SEM 0-4) of starch (i.e. RS) whereas healthy 

subjects showed 19·3 (SEM 0·7) g/d of ileal excretion. The explanation of the 

difference may be the underestimation of RS in ileostomates and/or the 

overestimation of RS when intubation techniques are used (Champ et al., 2003). 



Table 2.7: Advantages and shortcomings of the studies performed with human subjects (Champ et al. 2003). 

H2 breath test* 

lleostomy modelt 

Advantaqes 

Simple and non-invasive 
Healthy subjects 

Direct collection of the ileal 
effluent (i.e . quantitative) 

Easy to perform 

Shortcominqs 

Semi-quantitative 
Strict standardization necessary 

Large intra- and inter-individual variation in H2 excretion 

Cannot be considered as healthy 
Physiological adaptation 

Water and electrolytes absorption 
Bacterial overgrowth 

Transit time (different from normal) 

lntubations of healthy 
subjects Healthy subjects Disturbance of the normal physiology by the long triple lumen 

tube 
Direct collection of the ileal effluent Quantification of the flow rate using a liquid phase marker 

Risk of selectivity of the tube in case of heterogeneous food 
Ex_i:>_ensive and long_ 

* Determination of the increase in H2 in the breath after the consumption of malabsorbed carbohydrates. 
t Patients who have had a colectomy for ul cerative colitis. 
:j: Collection of the ileal content in healthy subjects after intubation using a constant perfusion technique of solution containing an unabsorbable 
marker. 

35 



Chapter 2: Literature review 36 

The microbial population is different in a normal distal ileum with 105-106 

bacteria/g as compared to that of the terminal ileum in ileostomy subjects which 

is 107-108 bacteria/g . However, these differences might sound small when 

compared with a population of 1012/g found in the caecum. Changing the bag 

every 2 h and deep-freezing it prevents the substantial bacterial degradation of 

NSP and RS. 

Thus, the model gives a slight underestimated amount of carbohydrate 

recovered , especially of easily fermented carbohydrates such as 

oligosaccharides. In contrast , when intubation is used, the tube is believed to 

decrease intestinal efficiency because of the decrease in the oro-ileal transit 

time. This is believed to be the cause of overestimation of RS . 

Of the three techniques , the intubation technique might be the only one , 

which can be exploited directly. The main disadvantage is the presence of the 

tube along the small intestine and its possible control on the passage of food. 

Using ileostomy, the direct quantification of excretion can be achieved with 

minimal bacterial degradation but there is a slight underestimation of the RS 

content. The hydrogen breath test will be more acceptable if the calibration could 

be improved (Champ et al., 2003) . 

2.5.7 Production of Resistant Starch 

Present day eating habits tend to result in diets which are very high in 

digestible carbohydrate leading to obesity. This could be ameliorated by 

increasing the level of RS in the diet , with the added benefit of increasing the 

production of butyrate in the large bowel. The most common RS in diet , out of the 

classified four categories , is retrograded starch RS3 as it is formed during food 

processing (Escarpa et al., 1996) . This is also the most extensively studied type. 

The class and extent of derivatization that may be legally permitted restricts the 

use of RS4 in the food industry on a large scale . 

Inside the native granule , starch is tightly packed in a radial pattern , 

relatively dehydrated . Due to the compact nature of the native starch granule the 



Chapter 2: Literature review 37 

accessibility of digestive enzymes is limited (Haralampu , 2000). Banana , which is 

predominantly consumed raw, is the chief source of RS2 (Resistant starch 

granules) in the human diet (Champ et al. , 2003) . Heating in excess of water 

disrupts the native semi-crystalline granule rendering it accessible to digestive 

enzymes. 

However, if the gelatinized starch is allowed to cool , recrystallization of 

polymers occurs forming microcrystalline filaments , often as double helices. 

These are created from amylase chains and upon further retrogradation these 

double helices form hexagonal units. When this happens, a partial crystalline 

structure is formed , which is responsible for their resistance to the degradation by 

digestive enzymes (Haralampu , 2000) . A scheme of the process is shown in 

Figure 2.6. 

An elementary method for production of RS3 is by controlled hydrothermal 

treatment. Pomeranz and Sievert (1990) reported an increase of up to 20 -35 % 

in the RS3 content of high amylase cornstarch (Hylan VII) by repeated cycles of 

autoclaving and subsequent cooling. It was quite evident from the literature that 

the factors that affected the yield of RS were amylose/amylopectin ratio , along 

with the temperature and sample/water volume ratio (Escarpa et al. , 1996) . Table 

2.8 shows how the yield of RS3 is related to the amylase content in the source . 

Random Junction Zones- Crystallites 
Coil Double Helices 

Figure 2.6: Schematic of amylase retrogradation (From Haralampu , 2000) . 



Chapter 2. Literature review 38 

Table 2.8: RS yields in autoclaved and retrograded amylase I amylopectin 

mixtures (From Escarpa et al., 1996). 

Amylose/Amylopectin (%. dm) 
100/0 
40/60 
75/25 
25175b 
50/50 
15/85 
0/100 

RSa (%. dm) 
36.45 ± 2.31 
1907 ± 040 
28 06 ± 146 
18.16 ± 0.23 
2148 ± 041 
8.97 ± 0.29 
7.61 + 0.38 

a Values are average of three gelatinization treatments in HCHPA 
b Potato starch. 

1" con;rr,Uer 

. -- - . 

Figure 2.7: Scheme of high pressure autoclave (Escarpa et al., 1996). 



Chapter 2: Literature review 39 

Table 2.9: comparison of RS yields by different authors (From Escarpa et al., 

1996). 

Standard 
0% Amylase 

1 00% Amylase 
Potato starch 

RS (%) RS (%) 

a Values of Berry (1986) , b Va lues of Escarpa et a/.(1994 ), c values of Sievert & Pomeranz (1989) 

The Hydrothermal Treatment used by Escarpa et al. (1996) improved on 

previous gelatinization studies by using a heat controller within the high-pressure 

autoclave , shown in Figure 2.7, so as to standardize the internal conditions . 

Higher yields of RS3 were reported as compared to the studies conducted earlier 

(Table 2.9). 

Skrabanja & Kreft ( 1998) and Huth et al. (2000) showed that debranching 

and/or lintnerization were much more effective procedures for the molecular 

we ight reduction of starch than other nonenzymatic hydrothermal treatments , 

e.g., autoclaving or extrusion cooking (Lehmann et al., 2003) . The fine structures 

of branched molecules markedly influence the rate and extent retrogradation of 

amylopectin (Botham et al., 1995). 

Besides being slow, the retrogradation of amylopectin is reversed by 

heating at 70°C. Debranching and /or lintnerization aims at releasing linear, 

recrystallizable polymers chains by debranching the amylopectin , which results in 

an effective retrogradation in fairly quick time (Lehmann et al., 2003; Kettlitz et 

al., 2000) . Schmied I et al. (2000) illustrated that the formation of RS from gels 

containing 30% poly-1,4 -a-0-glucans (linear chains with low molecular weight) 

was very rapid as 50-65% resistant structures were achieved after two hours of 

retrogradation . With the same gel concentration and a retrogradation temperature 

of 4 - 25°C , up to 94% of high a -amylase resistant starch can be obtained . 



Chapter 2: Literature review 40 

The retrogradation of debranched maltodextrin generates a much higher 

RS3 content. Maltodextrins are starch hydrolysis products with a wide distribution 

of oligomeric and polymeric a-D-glucans with a DP ~ 20. To obtain linear, low 

molecular weight, recrystallizable polymer chains, maltodextrin was partially 

debranched. An RS3 content of up to 65 % could be achieved if the 

recrystallization was carried out at 25°C on a 30% (w/w) maltodextrin gel 

previously deb ranched by isoamylase (Schmied I et al., 2000). 

High concentrations of oligosaccharides retarded RS3 formation and their 

reduction in the gel markedly accelerated the formation of RS3 structures up to 

56% within 24 h. Large amounts of optimally long linear polymer chains (poly-1,4 

-a-D-glucans) were produced using amylosucrase. Figure 2.8 shows a 

chromatogram of poly a-D-glucans preparation obtained by in vitro synthesis with 

amylosucrase according to the method described by Buttcher et al. (1997). A 

relatively high concentration of polymers with DP 10 - 20 can be seen from a 

range distribution between DP 10 and 35 (Schmied! et al., 2000) 

Figure 2.8: Chromatogram showing poly a-D-glucans preparation containing 

polymers ranging between DP10and 35 (From Schmeidl et al., 2000). 



Chapter 2: Literature review 41 

Interestingly, there was no separate step for retrogradation in the 

technique outlined by Kettlitz et al. (2000) . The maltodextrin solution (45%) was 

cooled to 50 °C and incubated for 48 hrs at a pH of 4 after the addition of 

isoamylase (0 .1 % starch d.b.) . Enzymatic debranching and retrogradation 

occurred at the same time during the incubation , which made the process faster 

and cheaper, according to the author. 

Lehmann et al. (2003) used lintnerization in addition to debranching for 

reduction of molecular weight. It was accomplished by stirring of starch for 1 to 7 

days after adding a fivefold volume 7.5% HCL. Later, the samples were freeze­

dried after being washed and centrifuged . Retrogradation after lintnerization up to 

7 days led to a low yield of approximately 20% probably due to insufficient 

reduction in molecular weight. However, acid hydrolysis followed by enzymatic 

debranching led to an increase in RS3 content in pea starch. Table 2.10 shows a 

total of 51 % RS3 yield after one day which could probably be explained by the 

higher susceptibility of a-1 ,6 linkage to acid hydrolysis than a-1,4 linkages. 

According to Brumovsky & Thompson (2001) , the native starch granule 

consists of a metastable state , packing the semi crystalline material structure 

inside the granule quite efficiently. However, the resistance to digestive enzymes 

is reduced or lost completely, on heating the granules in water. The heat stability 

of granular starch can be effectively increased by hydrothermal treatment. 

Annealing (ANN) and Heat Moisture Treatment (HMT) are two types of 

hydrothermal treatments that have been used to modify the starch granule , 

physiochemically, without wrecking its structure. ANN relates to the treatment 

given to starch at a moisture level of more than 40% whereas the treatment given 

at lower than 40% moisture is known as HMT. Annealing is achieved by 

incubating the sample at a gel concentration of 30% (w/w) at specified 

temperatures (50-70°C) above the glass transition temperature but below the 

gelatinization temperature. Granular RS yield was higher with ANN followed by 

Partial Acid Hydrolysis than ANN alone (Brumovsky & Thompson , 2001 ). 



Chapter 2: Literature review 42 

Table 2.10: Resistant starch content of native enzymatically de branched and 

lintnerized pea starch products analyzed after different retrogradation conditions 

(From Lehmann et al. , 2003). 

Sample Retrogradation conditions Yield of RS (%) 
Storage Starch 

temperature (°C) concentration in gel 
(%) 

Native starch 21.4a±0.7 

Deb ranched 4 10 37.3 b ± 1.6 
starch 

4 20 42.4 C ± 0.1 
25 10 38 .0 b ± 0.9 
25 20 42 .1 c±1 .3 

Acid hydrolyzed 25 10 51 .3 d ± 0.9 
(1 day) and 

Deb ranched 25 20 37.5 b ± 0.9 
starch 

Ac id hydrolyzed 25 10 17.2 a± 0.4 
(7 days) and 
Deb ranched 25 20 19.7 a±1 .5 

starch 
Data are means ± SO; n = 3, Numbers in the column followed by a letters in common were not 
significantly different (p > 0.05). 

Table 2.11: Resistant starch content of a 1 day lintnerized , enzymatically 

debranched, and retrograded (24 h at 25 °C, 20% w/w starch concentration in gel) 

products after 10 min of tempering at 93 °C and a further tempering at 93 °C for 

different storage times (From Lehmann et al. , 2003). 

Sample 

Starting substance 

Storage time (hr) 

1 
8 

24 

Yield of resistant starch (%) 

57.3 a± 0.7 
46.2 b ± 0.7 
46.9 b ± 0.6 
74.4 C ± 2.9 

Data are means ± SO ; n = 3; numbers in the column followed by a letter in common were not 
sign ificantly different (p > 0.05). 



Chapter 2: Literature review 43 

Lehmann et al. (2003) tempered (annealed) the lintnerized , enzymatically 

debranched and retrograded product with the highest RS3 yield to investigate the 

increase, if any, in polymer interaction . The retrograded product was stored for 

24 hrs at 93 °C , which eventually generated 74% RS3, even higher than the 

commercially available products (Table 2.11 ). This was probably due to the 

perfection of crystalline regions and recrystallization of imperfect crystals . The 

product formed from a higher starch concentration in gel increased the 

gelatinization temperature indicating a higher thermal stability . Schmied! et al. 

(2000) reported a rise in thermal stability of RS3 product with an increase in 

recrystallization temperature . 

The technique described by Haralampu et al. (1998) to produce RS2 

related to the heating of starch granule under conditions sufficient to make them 

swell but preventing them from rupturing . The product obtained from this method 

had a Total Dietary Fibre (TDF) of 20% to about 50% by weight. Interestingly, the 

product displayed heat stability in the range of 90°C to 150 °C rendering it quite 

resistant to normal cooking temperatures in most food processes. 

2.5.8 Thermal analysis of resistant starches using differential scanning 

calorimeter (DSC) 

DSC is the most widely used instrument to study the thermal behavior of 

starches because of its high sensitivity to of weak transitions and its high 

resolution (Liu et al. , 1991 ). Generally, DSC is used to determine the specific 

heat, heat of fusion , and heat of reaction or heat of polymerization of materials 

and involves heating or cooling a sample and a reference over a temperature 

range , under such conditions that the two are always maintained at identical 

temperatures. The additional heat required to maintain the sample at the same 

temperature as the reference is measured , and is a function of chemical or 

physical changes , which are talking place. Basically, the difference between the 

independent supply of power to the sample and reference is recorded against the 

programmed temperature (Brown , 2001). A schematic is shown in the Figure 2.9. 



Chapter 2: Literature review 44 

Sample pan Starch sample Reference pan 

LJ 

I I eaters computer interface 

Figure 2.9: Diagram of Differential scanning calorimeter (DSC) with a computer 

interface to monitor and regulate heat flow. 

The first reported application of the DSC to starches was the 

measurement of heat of gelatinization in 1971 (Biliaderis, 1990) . Since then it has 

been widely used to study the thermal behavior of starches including 

gelatinization, glass transition temperature and crystallization. However, due to 

different measuring conditions and the complexity of starch structure, the results 

are often controversial. Many physiochemical changes can take place during the 

thermal analysis of starches including gelatinization, melting , glass transition , 

crystallization , molecular degradation , volume expansion and water migration 

which makes it a complex process (Yu and Christie, 2001 ) . 

Typical thermal profiles of starches containing more than 50% moisture 

reflect two kinds of thermal phenomena. The first involves the rearrangement of 

starch crystals and the second one connotes the disorganization of amylase-lipid 

complexes. The melting of the starch crystallites follows swelling of the 

amorphous region of the starch granule by the water in the system. If the water 

content is limited the melting of the crystallites is spread over a wider 



Chapter 2: Literature review 45 

temperature range and can sometimes give rise to a second peak. The melting 

temperature thus depends upon the moisture content (Lui et al., 2005) . Overall , 

the structural reorganization of starches and their thermal behavior is water 

content dependent. Also , there are reported variations in thermal profiles due to 

heating rates (Biliaderis, 1990). 

To avoid sloping and bending of DSC curves , baseline subtraction has 

been widely used particularly when the starch sample contains water. It has also 

been noticed that shaken samples or starch granules packed tightly using an 

ultrasonic bath gave better defined peaks than unshaken ones. Because the 

thermal enthalpies of starches are quite low, the mass of sample used for DSC is 

generally high . This leads to a compromise between resolution and sensitivity . It 

is generally recommended to have a sample mass of 5-20mg to obtain well ­

defined endotherms (Yu and Christie , 2001 ). 

2.5.9 Applications of RS 

These days most Americans seem to view bread as the "kiss of death". 

That's because millions of people in US, as well as in many other countries have 

adopted low-carbohydrate diets or the so called "Carb Craze" that push proteins 

and cut carb-laden foods such as pasta , potatoes and bread to a minimum. 

The survey by Morgan Stanley analysts , US estimated that 13 percent of 

the U.S. population was on the Atkins , South Beach , or other low-carb diet in 

January , 2003 . Participation has since tailed off to 11 percent. 

Caloric control is paramount to combat obesity. Depending on the RS 

product used , resistant starches contribute 1.6-2 .5 kcal/gm versus 4 kcal/gm for 

rapidly digested starches. In addition to this, RS has numerous potential health 

benefits and functional properties. The physiological benefits of RS, which are 

similar to those of fibre , include increased faecal bulk and the production of 

butyrate, both of which have been shown to promote good colonic health . 

Because it is not absorbed in the small intestine, the products , which contain RS, 

don't raise blood glucose levels as other carbohydrate sources do . 



Chapter 2: Literature review 46 

Products like, "NiteBite", "Gluc-0-Bar", and "ExtendBar" snack bars all 

contain RS2 . These foods are specifically formulated for people suffering from 

type- 2 diabetes. The "Choice" bar which contains RS3 called Crystalean , 

manufactured by Opta food ingredients based in Bedford , US is another example 

of a medical food. 

Quite a few leading food ingredients companies are now manufacturing 

RS. The leading supplier of resistant starch to Australia and New Zealand is the 

National Starch Company. HI-MAIZE and NOVELOSE, which are RS2 and RS3 

respectively, have made a considerable impact on the food industry of Australia 

and New Zealand . A number of products with HI MAIZE as an ingredient are 

being marketed in Australia and New Zealand (Appendix 1 ). 

Another product called Cerestar ACTISTAR, manufactured by the Cargill 

Company, USA, is another functional food ingredient rich in RS. It is claimed to 

have 53% of RS in the end product. It is promoted as being an appropriate 

ingredient for incorporation in breads biscuits/ cookies, muesli , milk drinks etc. 

The MGP Ingredients Inc. began marketing its wheat based RS in July , 

2003. FIBERSYM, as it is called now, is capable of delivering 70-80% of tota l 

dietary fibre. The company refers to it as an "invisible fibre" because of its ease in 

blending. The product can be incorporated in a variety of bakery products and 

snack foods as well. 

Using functional ingredients like resistant starch is a step forward to a 

balanced approach to eating and improving diets. Balancing nutritional needs 

and taste preferences is the chal lenge for food scientists and for food 

manufacturers as well. 



Chapter 2: Literature review 47 

2.6 Conclusions 

The versatility of starch as a food ingredient is incomparable to any other 

ingredient in terms of application in the food industry. Designed by nature as a 

plant energy reserve, this polymeric carbohydrate is only second to cellulose in 

abundance. Modification of starches has resulted in numerous highly functional 

derivatives. The choice of starch for a certain food product depends upon a 

range of features. These include sensory properties of final product, 

manufacturing process , shelf life requirement and other ingredients in the food 

product. 

Gelatinization of starch is critical to building the structure and texture of 

most food products. Starches upon cooking are high in viscosity and imparts 

desirable body to a variety of foods. The amount of thickness of the final product 

depends of the concentration of starch, amount on water and the shear appl ied 

during processing . The type of starch applied depends on the textura l 

requirements of the final product. In baked goods, where limited amount water is 

present, it is desirable to have starches, which bind with water early on whereas 

in liquid foods such as soups and gravies quick thickening starches are used. 

By definition , resistant starch is the fraction of starch that escapes 

digestion. RS is classified in to four classes i.e. , physically inaccessible starch 

(RS1 ); resistant starch granules (RS2) ; retrograded starch (RS3); chemically 

modified starch (RS4). 

Raw starch is digested poorly but cooking in presence of water enhances 

digestibility. While gelatinization increases digestibility, subsequent retrogradation 

increases resistance to digestion. Another important factor affecting starch 

digestibility is the amylase: amylopectin ratio . Higher amylose content increases 

resistance to gelatinization and also makes the starch more susceptible to 

retrogradation 

Earlier, RS was quantified using the official AOAC method for estimation 

of dietary fiber. Berry (1986) modified the AOAC method, which gave better 



Chapter 2: Literature review 48 

estimates of RS. Since then , many researchers over the past decade have tried 

to devise an in vitro method to precisely quantify RS in foods without total 

success . However, McCleary & Monaghan (2002) refined the procedure of 

quantifying RS by studying the strengths and weaknesses of the methods in 

earlier literature and is expected to become generally accepted method. 

At present, three prospective methods are available to obtain the in viva 

values on the RS content of foods that are required to ratify the in vitro methods. 

These are (1) hydrogen breath test, (2) intubation technique and (3) ileostomy 

model. Of the three techniques , the intubation technique might be the only one, 

which can be exploited directly. 

An elementary method for the production of RS3 is by control led 

hydrothermal treatment e.g . autoclaving. It has been showed that debranchi ng 

and/or lintnerization were much more effective procedures for the molecular 

weight reduction of starch than other nonenzymatic hydrothermal treatments , 

e .g., autoclaving or extrusion cooking . Debranching and /or lintnerization aims at 

releasing linear, recrystall izable polymers chains by debranching the 

amylopectin , which results in an effective retrogradation in fairly quick time . 

RS2 can be produced by heating of starch granule under conditions 

sufficient to make them swell but preventing them from rupturing. An nealing 

(ANN) and Heat Moisture Treatment (HMT) have been successfu lly used to 

modify the starch granule, physiochemically, without wrecking its structure in 

order to produce boiling stable RS2. 

With increasing awareness and health consciousness among consumers , 

the demand of RS has grown rapidly. Quite a few leading food ingredients 

companies are now manufacturing resistant starch as a standard product. Most 

of these products fall under the solid food category, like breads, pasta products , 

fruit bars etc. In most of these products RS has been included as an addition 

ingredient to the existing formulation rather than substituting a fi ller ingredient. 

However, replacement of starch with RS in food systems has not been 

studied . There is a need to test the substitution of resistant starch for the other 



Chapter 2: Literature review 49 

thickeners in a fluid food system. Such a study will help in understanding the 

functional behavior of RS in real foods with respect to rheological and sensory 

characteristics. The results for the study can be used in evaluating RS in small­

scale food model, which can be extrapolated later to real food products. 



Chapter 3: Materials and methods 50 

CHAPTER3 

MATERIALS AND METHODS 

3.1 Materials 

The RS samples were , a commercial resistant maize starch type II with 

52% TDF (Hi-Maize 1043, National starch and Chemical NZ Ltd ., Green Mount, 

Auckland , New Zealand .), another Type II commercial resistant maize starch with 

22% TDF (Hi-Maize 958 , National starch and Chemical NZ Ltd ., Green Mount, 

Auckland , New Zealand .), a commercial Tapioca RS Type Ill with 54% RS 

(Cerestar Actistar , Cargill food company , Cedar Rapids , Iowa, United states) , and 

a Type Ill commercial RS with 30% TDF (Novelose 330, National starch and 

Chemical NZ Ltd. , green mount, Auckland , New Zealand.) . For standard 

formulations commercial waxy maize starch (Novation 2700 , National starch and 

Chemical NZ Ltd ., green mount , Auckland , New Zealand .) was used . 

Other ingredients (salt , sugar, pepper, garlic powder) were purchased 

from a local supermarket and onion powder was supplied by Stonemill (G .S. Hall 

and company , Auckland , New Zealand) . Dairy Products Supply , Emmen 

(Holland) and Chr. Hanson Pty. Ltd ., Bayswater (Victoria , Australia) supplied 

chicken flavor and spray dried vegetable fat respectively. Archer Daniels Midland 

Company, Decatur, USA, provided xanthan gum NF/FCC , and skim milk powder 

was provided by NZMP, New Zealand . All chemicals used in the analysis were of 

analytical grade obtained from either Sigma Chemical Co . (St. Louis , MO) or 

Megazyme international Ltd . (Wicklow, Ireland) . 



Chapter 3: Materials and methods 51 

3.2 Approach to development of Model Food system for 

functionality testing of resistant starch 

In order to be able to evaluate RS with new functional properties under 

realistic conditions of use , a soup model was developed . The choice of food 

system was based on the ease of rheological and sensory measurement along 

with a straightforward method of preparation . The model also presented an 

opportunity to verify thermal stability of RS. 

The approach utilized has been 

1. Various formulations for the manufacture of particular food system of 

interest were reviewed. 

2. The simplest combination representative of the majority of formulations 

was selected . Multiple use of a single additive was avoided wherever 

possible . 

3. The laboratory process of making the food system on small scale was 

then investigated. 

4. For the evaluation of the system, quantifiable food characteristics were 

selected . 

5. A suitable experimental design was developed thereafter. Generally 

fractional factorial or central composite design is most applicable for a 

number of ingredients at multiple compositional levels . 

6 . The formulation that provided for differentiation of RS by standard 

functionality testing was chosen . 

Four representative cream of chicken soup formulations were the basis of 

initiating the study. These formulations were sourced either from text books or 

from the Internet (Appendix 2) . The formulations included both industrial starch 

and wheat flour as thickening agents. A modified formulation based on these four 

basic formulations was selected (Appendix 3) . 



Chapter 3: Materials and methods 52 

3.3 Method of Soup Preparation 

The soup was prepared in laboratory by slurring the dry mixture of 

ingredients with water at 50°C and rapidly heating in a rotary evaporator 

(Rotavapor-R, Nicholas Watson Victor Ltd) to a temperature of 95°C by placing it 

in a boiling water bath. The heating was done without the application of vacuum 

by using just the rotating mechanism of the