University of Southern Queensland

Use of Marker Assisted Selection for the

Introgression of Quality Traits from Australian into
Chinese Wheats

A Dissertation Submitted by

Benedette Watson

Bachelor of Science

For the Award of

Masters of Science

2008
Abstract

Quantitative Trait Loci (QTLs) for polyphenol oxidase and xanthophyll

have a significant impact on variation in wheat flour for noodle colour and

colour stability. QTLs from two Australian wheat cultivars, Sunco and

Tasman, have been backcrossed into two Chinese wheat varieties,

Chuanmai 22 and Mianyang 11, to assess marker predictability for these

important traits in significantly different genetic backgrounds. The concept

of Marker-Assisted-Selection (MAS) is being trialled in this study as a

proposed method for wheat improvement. In this approach molecular

markers are used in conjunction with backcross breeding methods to

introgress specific characters into elite breeding materials, with the goal of

improving the quality attributes of wheat for the Asian noodle market. After

three single seed descent generations, the backcross populations generated

allow four QTLs to be investigated. These include two for polyphenol

oxidase (chromosome 2A and 2D) and two for xanthophyll (chromosome

3B and 7A).

This research was successful in identifying microsatellite markers that are

capable of predicting PPO activity levels and Xanthophyll content within

the backcross populations. These microsatellites were validated as useful

ii
markers for these quality traits, as they have also found to be important in

the Sunco x Tasman doubled haploid population. The combination of

marker assisted selection and backcrossing has generated three lines that

contain different combinations of the PPO activity and Xanthophyll content

QTLs. These lines have been found to produce low levels of PPO activity

and have a low Xanthophyll content. This improvement in flour colour and

colour stability highlights the potential of marker assisted selection as a

useful tool in wheat breeding.

iii
Declaration

I certify that the ideas, experimental work, results, analyses and conclusions

reported in this dissertation are entirely my own effort, except where

otherwise acknowledged. I also certify that the work is original and has not

been previously submitted for any other award, except where otherwise

acknowledged.

…………………………………… Date:

Signature of Candidate

Benedette Watson

(nee Cavallaro)

ENDORSEMENT

………………………………….. Date:

Signature of Supervisor

Professor Mark Sutherland

iv
Previously Published Material

Some of this work has been previously presented at scientific conferences.

Cavallaro, B., Storlie, E., Sutherland, M.W., Mares, D., Sheppard, J., &

Banks, P., (2004) Backcrossing QTLs from Australian to Chinese wheats,

Wheat Breeding Society of Australia, 11th Assembly, Canberra ACT.

v
Acknowledgements

While working as a Research Assistant I commenced this master’s project

part-time in 2002. Over the following six years a few major events occurred

that significantly slowed the completion of this master’s thesis. I was

married in the end of 2004, so that warranted a little holiday. But it was in

March of 2006 that my study life was successfully slowed to a halt by the

arrival of our son, Noah. At this time I took a leave of absence from study

and concentrated on the important job of being a new Mum. The final

distraction was during the first six months of 2008 when we all travelled to

Oxford, UK where my husband, Dr Mike Watson, spent a sabbatical leave.

It has been an eventful seven years after deciding to commence this master’s

project. And it wouldn’t have been possible without the support and

guidance from many people along the way.

This project came about through a large Australian Centre for International

Agricultural Research (ACIAR) funded project entitled “Wheat

Improvement in Sichuan Province” (1999-2006) on which I was employed

as a research assistant. This large project coordinated by Mark Sutherland

at USQ, was based on a close collaboration with the Crop Research Institute

within the Sichuan Academy of Agricultural Sciences (SAAS), located in

Chengdu, China, the Leslie Research Centre (QDPI & F), the University of

vi
Sydney, Cobbitty and the University of Adelaide, Waite Campus. I would

like to thank all involved in this overall project as they have supported me

during the time I was employed, but also while I completed my study.

During this project there were a number of trips to China to meet with our

colleagues from Sichuan. I would especially like to thank Dr Yuchun Zou

for his hospitality, but most of all for detailed knowledge of the wheat

varieties used in this master’s project, and all the information that I gained

from him throughout these last years.

The initial concept for this project was developed by Dr. Eric Storlie while

he was the Project Officer for the ACIAR project employed at USQ. It was

he who made the initial crosses to establish the four wheat populations

which are central to this Masters project. A field trial planted at the Leslie

Research Centre in 2005 was also an integral part of this research. This trial

was maintained by Lloyd Mason (Leslie Research Centre) and without his

help and hard work the trial would not have been such a success. I would

like to thank a number of the staff at the Leslie Research Centre, these being

David Martin and all the Quality Laboratory staff, and Mandy Christopher

and Raeleen Jennings of the Molecular Marker Laboratory.

I would like to thank Dr. Daryl Mares (University of Adelaide) for his

support and advice. It was through his guidance that I learnt the techniques

vii
required to complete the biochemical section of this project. Dr. Anke

Lehmensiek’s (USQ) support, knowledge and friendship have been

immeasurable throughout the years. Her successful curation of the genetic

maps and knowledge on all aspects molecular biology enabled my research

to progress and run smoothly. I would also like to thank Dr Ashley Plank

for his time and direction with the data analysis carried out in this project.

My support system at USQ has grown enormously over the many years that

I have been employed in the Dept of Biological and Physical Sciences. This

has enabled me to work successfully as a part of the Crop Improvement

Group (within the Centre for Systems Biology, USQ), but also as part of the

department. I would like to thank everyone that has been part of the Crop

Improvement Group and the technical staff within Biological Sciences.

Debbie White, Ros Gill, Adele Jones are friends and colleagues that have

given me the drive to continue with my studies when times got a little hard.

I would like to thank Vic Schultz who was an integral part of the department

for many years, he was always there for the less glamorous jobs and

offering unending emotional support when needed.

To my supervisor Professor Mark Sutherland, thank you for having faith in

me when I didn’t. Without your understanding and support this project

would not have been completed.

viii
My family and friends have been my constant and eternal support system

and without them I know the road would have been much harder. My

husband Mike and son Noah are my driving force and I thank them for their

love and understanding. To my wonderful babysitters, Helen and Peter

Cavallaro, Joan and Gordon Watson and Renee Nairn, thank you for your

help, long days and your unlimited support.

ix
Table of Contents

Chapter 1. Introduction and Literature Review……………1.

1.0 Introduction………………………………………………………..1.

1.1 Wheat Production…………………………………………………2.

1.1.1 Wheat Production in Australia……………………………..3.

1.1.2 Wheat Production in China………………………………...8.

1.2 Breeding Objectives in wheat improvement…………………...11.

1.2.1 Yield Potential………………………………………………11.

1.2.2 Heat and Drought Tolerance………………………………12.

1.2.3 Disease Resistance…………………………………………13.

1.2.4 Quality…………………………………………………….14.

1.2.4.1 Noodle Quality……………………………………….15.

1.2.4.2 Noodle Types………………………………………...16.

1.2.4.3 Noodle Colour……………………………………….18.

1.2.4.3.1 Impacting Factors on noodle colour

i) Polyphenol oxidase Activity Levels…………18.

1.2.4.3.2 Impacting Factors on noodle colour

ii) Xanthophyll Content………………………..21.

1.3 Breeding methods used in wheat improvement………………..23.

1.3.1 Pedigree Breeding Method………………………………..23.

1.3.2 Single Seed Descent……………………………………….24.

1.3.3 Recurrent Selection Method………………………………25.

1.3.4 Backcross Breeding……………………………………….25.

x
 1.3.5 Doubled Haploid Populations…………………………….27.

1.4 Genetic markers for wheat improvement………………………….28.

1.4.1 Types of Markers…………………………………………..28.

1.4.2 Applications of markers in wheat breeding………………..30.

1.4.3 Construction of genetic linkage maps……………………..32.

1.5 Quantitative Trait Loci (QTL) analysis…………………………..34.

1.5.1 Methods of detecting QTLs……………………………….34.

1.5.2 Factors affecting QTL analysis……………………………35.

1.6 Marker Assisted Selection (MAS)…………………………………36.

1.6.1 Gene/QTL Introgression…………………………………...37.

1.6.2 Marker Assisted Backcrossing…………………………….38.

1.6.3 Incorporating MAS into breeding programs………………40.

1.7 Rationale for the Current Study…………………………………..42.

Chapter 2: Establishment and Genotyping of Backcross

Populations …………………………………………………...51.

2.1 Introduction………………………………………………………….51.

2.2 Materials and Methods……………………………………………....53.

2.3 Results………………………………………………………………...62.

2.4 Discussion…………………………………………………………….72.

xi
Chapter 3: Phenotyping through Biochemical Testing……76.

3.1 Introduction…………………………………………………………76.

3.2 Material and Methods……………………………………………...78.

3.3 Results…………………………………………………………….....84.

3.4 Discussion………………………………………………………….102.

Chapter 4: Analysis of the Marker Assisted Selection in the

backcross populations………………………………………107.

4.1 Introduction…………………………………………………………107.

4.2 Materials and Methods……………………………………………..110.

4.3 Results…...…………………………………………………………112.

4.4 Discussion……...…………………………………………………..122.

Chapter 5: General Discussion……………………………128.

5.1 Research Outcomes and the Impact of this Project on Wheat

Improvement ………………………………………………………128.

5.2 Future Directions…………………………………………………..130.

5.3 Conclusion………………………………………………………….132.

References…………………………………………………...133.

Appendix A: Protocols used in the Genotypic and

Phenotypic assessment of the backcross populations…….148.

xii
List of Tables

Table 1-1: Total world production of Wheat, and wheat production by

major wheat producers: China, EU27, India, Russia and
USA………………………………………………………………………...3.

Table 1-2: Australian wheat production: comparing the past five years and
the forecast from September 2008 on cropping area, yield and total
production…………………………………………………………………..6.

Table 1-3: Australian production, consumption and export of wheat over

the last three cropping seasons…………………………………………......6.

Table 1-4: List of Priority traits/breeding objectives endorsed by the Pre-

breeding alliance and the GRDC………...………………………………...7.

Table 1-5: World Wheat Production and Consumption; and for Australia
and China (units: thousand metric tons)…………………………………....8.

Table 1-6: The National standard of wheat flour specifications for various
end-products consumed in China………………………………………….15.

Table 1-7: Wheat products consumed in China………………………….17.

Table 1-8: Markers that are currently used in wheat breeding programs
across Australia……………………………………………………………31.

Table 2-1: Molecular Markers used in the genotyping of plant material in

all four backcross populations – comparing years 2002 with 2003-20006..61

Table 2-2: Summary of the genotyping information of the revised and

original versions of the Sunco x Tasman linkage map……… …………...69.

Table 2-3: Summary of the lines included in the 2005 field trial. A wide
range of genotypes from each population were included to evaluate the use
of MAS.........................................................................................……..….71.

Table 3-1: A summary of the outliers evaluated from each PPO assay
conducted on the 2005 field trial material; genotypes of all four backcrossed
populations, parental varieties and standards………………………..…....86.

Table 3-2: Summary of the genotypes reanalysed because of a high initial

Standard Deviation…………………………………………..…………....89.

xiii
Table 3-3: Descriptive information for each backcross population for PPO
activity………………………………………………………………….....92.

Table 3-3: Legend………………………………………….………….....93.

Table 3-4: Summary of the results from the repeat xanthophyll content
Tests……………………………………………………….……………....95.

Table 3-5: Descriptive information for each backcross population for

Xanthophyll content……………………………………………………..100.

Table 3-5: Legend………………………………………………………….…101.

Table 4-1: Proportions of heterozygote genotypic data points removed from

the genotypic data collected from each of the four backcross populations for
the SPSS data analysis…….......................................................................112.

Table 4-2: Results from the linear Regression analysis for PPO
Activity…………………………………………………………………..114.

Table 4-3: Results from the linear Regression analysis for Xanthophyll
Content …………………………………………………………...……...116.

Table 4-5: Three lines that produce a low result for PPO activity and
Xanthophyll content. These lines also have marker alleles contributed by
Sunco, some of which are found to be significant and able to explain a
proportion of the variation (in its population).
The lines below are all performing better than either Chinese variety
(Chuanmai 22 and Mianyang 11)…………………..……………………121.

xiv
List of Figures

Figure 1-1: Areas of wheat production in Australia. Data from the

Agricultural Census of 2005-06 by the Australian Bureau of Statistics…....4.

Figure 1-2: Australian wheat production (kt) of six states over the last five
years (2002-2007), and the predicted season for 2008-09…………..……..5.

Figure 1-3: Area of wheat planted (‘000ha) in the six states over the last
five years (2002-2007), and the predicted season for 2008-09…………….5.

Figure 1-4: Wheat production Zones in China. Separated by location and

time of planting……………………...……………………………………...9.

Figure 1-5: The reaction catalysed by Polyphenol oxidase……...………19.

Figure 1-6: Diagram of the wheat grain………………………………....20.

Figure 1-7: Chemical structure of Xanthophyll………………..………...22.

Figure 1-8: Schematic Diagram of the Pedigree and Single-seed decent

breeding methods……………………………...…………………………..24.

Figure 1-9: Backcrossing Technique used in self pollinating plants.

A cross between donor parent A and recurrent parent B…………..……...26.

Figure 1-10: Schematic Diagram of production of Doubled haploids…..28.

Figure 1-11: Pedigree of parents; Sunco and Tasman selected for mapping
population: Sunco x Tasman doubled haploid population…………..……46.

Figure 1-12: Pedigree information of Varieties Fan 6…………………...48.

Figure 1-13: Pedigree information of Varieties Mianyang 11 and

Chuanmai 22………………………………………...…………………….49.

Figure 2-1: Flow chart of laboratory and field work for the phenotyping
and genotyping of the lines in the four backcrossed populations…………63.

Figure 2-2: QTL analysis carried out on chromosome 2A of Sunco x

Tasman DH population. All analysis used phenotypic data collected in the
1998 field trial; a. used the original genetic linkage map, b. used the revised
linkage map……………………………..………………………………...65.

xv
Figure 2-3: QTL analysis carried out on chromosome 2D of Sunco x
Tasman DH population. All analysis used phenotypic data collected in the
1998 field trial; a. used the original genetic linkage map, b. used the revised
linkage map……………………………………………………………….66.
.
Figure 2-4: QTL analysis carried out on chromosome 3B of Sunco x
Tasman DH population. All analysis used phenotypic data collected in the
1998 field trial; a. used the original genetic linkage map, b. used the revised
linkage map……………………………………………………………….67.

Figure 2-5: QTL analysis carried out on chromosome 7A of Sunco x

Tasman
DH population. All analysis used phenotypic data collected in the 1998 field
trial; a. used the original genetic linkage map, b. used the revised
linkage map……………………………………………………………….68.

Figure 3-1: Design of the microplate used for each PPO assay carried out
on the field trial material………………………………………………….80.

Figure 3-2: An example of a microplate of a PPO assay after the three hour
incubation period. Refer to Figure 3.1 for sample placement pattern…...85.

Figure 3-3: Frequency distribution graphs of PPO activity (OD 415nm)

results for each of the Sunco backcross populations……………………..90.

Figure 3-4: Frequency distribution graphs of PPO activity (OD 415nm)

results for each of the Tasman backcross populations…………………....91.

Figure 3-5: Frequency distribution graphs of Xanthophyll content (OD

436nm) results for each of the Sunco backcross populations…………….97.

Figure 3-6: Frequency distribution graphs of Xanthophyll content (OD

436nm) results for each of the Tasman backcross populations…………..98.

Figure 4-1: Chromosomes 2A, 2D, 3B and 7A of the Sunco x Tasman linkage
map, displaying the map distances (cM) between the microsatellites used in the
genotyping of the four backcrossed populations (’03-’05)
Only SSR markers were used in this figure, the complete QTL peaks can be
found in Chapter 2.....................................................................................118.

xvi
Chapter 1. Introduction and Literature Review

1.0 Introduction

“Plant breeding is essentially an election made by man of the best plants within a

variable population as a potential cultivar” (Sanchez-Monge 1993). The selection and

cultivation of grains for food has been a constant effort by man for centuries. It is

believed that wheat originated in south-western Asia, and wild cereals can still be

located within their natural and original habitats in the foot hills of Iraq-Kurdistan and

southeast Turkey (Feldman 2001). The cultivation of wheat occurred 10,300 – 6,200

years ago. Over this large period of time, basic but very important characteristics

would have influenced the selection of wheat plants. These included: plants with large

grains that didn’t shatter or lodge; and having the ability to survive through changes in

environmental conditions. These selections paved the way for modern wheat

breeding.

During the 19th century researchers focused on the origins of wheat and its genetic

background with the discovery of the work of Mendel. The onset of what is termed

“Modern Plant Breeding” has increased the understanding of all aspects of

plant/wheat breeding with respect to its genetics, physiology, pathology and

chemistry. Knowledge of these processes and the mechanisms involved has allowed

breeders to regulate and increase the variability needed to produce new varieties

(Feldman 2001).

1
Demand on wheat breeding has become even more important with the increasing

requirement for global food production. However, conventional breeding methods are

time consuming taking up to twelve years for the release of new varieties (Korzun

2002). It has been reported that using a combination of traditional breeding methods

with current diverse molecular techniques could help to reduce the time for release

and offer a reliable improvement of elite breeding material (Liu et al. 2004, Sorrells

and Wilson 1997).

1.1 Wheat Production

Wheat is a major stable food and is grown widely in many countries of the world. It

has been reported that by 2020 the majority of the world’s wheat consumption will be

in developing countries and this will have a direct affect on the amount of wheat

needing to be imported (CommodityOnline 2008).

The world wheat production is approximately 585 million tons per annum, and the

major producers are the European Union (EU-27), which is comprised of 27

countries, China, India, Russia and USA (CommodityOnline 2008). The Australian

Bureau of Agricultural and Resource Economics (ABARE) have produced economic

data on the production and consumption for Australia and the major world producers

for a number of years and seasons (Table 1-1) (ABARE 2008).

2
Table 1-1: Total world production of wheat, and wheat production by major wheat
producers: China, EU27, India, Russia and USA.
2006-07 2007-08 (Mt) 2008-09 forecast
(Mt) (Mt)
World Production 598 609 672
EU27 125 120 146
China 109 110 113
India 69 76 78
United States 49 56 67
Russian Federation 45 49 55

Consumption 612 612 643

- human 442 446 451
- feed 98 87 111
Price US$ 212 /t US$ 362 /t US$ 325 /t
(Data source: ABARE 2008)

For the purpose of this chapter the emphasis on wheat production will be on the

countries Australia and China as they have direct impact on this study.

1.1.1 Wheat Production in Australia

In Australia wheat is a winter grown crop that is sown between May and July and is

harvested in late spring/early summer. The amount of wheat produced throughout

Australia differs for each state with respect to yield and cropping area. What is termed

the Australian Wheat Belt spans most states with the exception of the Northern

Territory (Figure 1-1). A range of white-grained wheat varieties are grown in

Australia including bread wheats and durum wheats (for pasta) (Christopher and

Banks 2002). Soft grained wheat varieties grown in the Australian wheat belt are

concentrated in Western Australia and are generally used for biscuits and noodle

production. The varieties planted vary throughout Australia and are determined by

3
their adaptation for particular environments, resistance to diseases and their tolerance

to drought and frost.

Figure 1-1: Areas of wheat production in Australia. Data from the Agricultural
Census of 2005-06 by the Australian Bureau of Statistics (Data source:
http://www.abs.gov.au)

There is a noticeable difference in the wheat production and cropping areas between

the states of Australia, and it is Western Australia that is the highest producer of the

entire Australian wheat belt (Figure 1-2 and 1-3).

4
Figure 1-2: Australian wheat production (kt) of six states over the last five years
(2002-2007), and the predicted season for 2008-09 (Data source: ABARE 2008)

Wheat Production Area

50.0
45.0
40.0
35.0 2008-09f
Percentage

30.0
2007-08
25.0
20.0 Five year avg to
15.0 2006-07
10.0
5.0
0.0
NSW VIC QLD WA SA TAS
Weight (kt)

Figure 1-3: Area of wheat planted (‘000ha) in the six states over the last five years
(2002-2007), and the predicted season for 2008-09 (Data source: ABARE 2008)

Australian Wheat Cropping Area

40.0

35.0

30.0

25.0
Percentage

2008-09f

20.0
2007-08
15.0
Five year avg to
10.0 2006-07
5.0

0.0
NSW VIC QLD WA SA TAS
Area ('000 ha)

5
For the cropping season of 2008-09 an increase in wheat production in Australia of

approximately 72% to 22.5 million tons has been forecast, relative to the previous

season (ABARE 2008). An increase in production also leads to increased exports.

The expected tonnage to be exported will be nearly double that of 2007-08 (Table 1-2

and Table 1-3). Australia is a large exporter of wheat, attractive to many markets,

especially those in the Middle East and South East Asia, where Australia holds a 16%

share of the world’s wheat trade (AWB 2008).

Table 1-2: Australian wheat production: comparing the past five years and the
forecast from September 2008 on cropping area, yield and total production

Five year 2007-08 2008-09

average forecast
Wheat
Area Planted (‘000ha) 12375 12345 13552
Yield (t/ha) 1.50 1.06 1.66
Production (kt) 18828 13039 22460
(Data source: ABARE 2008)

Table 1-3: Australian production, consumption and export of wheat over the last
three cropping seasons
2006-07 2007-08 2008-09
(kt) (kt) forecast (kt)
Wheat
Production 25150 10822 22460
Domestic Use 6540 7381 6699
- Human and industrial 2287 2264 2287
- Feed 3672 4500 3740
- Seed 581 617 673
Exports 15969 8685 15689
(Data source: ABARE 2008)

Since the European settlement of Australia, wheat has been grown predominantly as a

food source. The initial varieties used over 200 years ago were not adapted for

Australia’s hard environment and these first crops were low yielding. With the

expansion of wheat production from coastal New South Wales into other states of

6
Australia, wheat production increased. This increase was also due to improved

farming techniques, application of fertilizers and pesticides, the introduction of

rotation crops and breeding of varieties specific to the Australian environment (AWB

2008).

In recent years breeding of wheat varieties has become essential in wheat production

and is focused in particular areas of development, varying slightly for the different

wheat producing areas of Australia. Breeding objectives are usually separated into

sections, these being Yield, Biotic stresses, Abiotic stresses and Quality attributes.

The GRDC (Grains Research and Development Corporation) is one of the major

funding bodies for wheat breeding in Australia drawing on funding from Australian

wheat producers and the Federal Government. Pre-breeding research

objectives/priority traits are regularly determined by discussions between the GRDC,

the Australian Winter Cereals Pre-breeding Alliance and breeding companies (Table

1-4). A more detailed look at wheat breeding objectives will be described in the

following sections of this chapter.

Table 1-4: List of Priority traits/breeding objectives endorsed by the Pre-Breeding

Alliance and the GRDC. (Data source: GRDC 2008)
Biotic Stresses Abiotic Stresses Quality
• Crown Rot • Drought • Defect elimination
• Leaf Rust tolerance • Late maturity alpha-
• Root Lesion • Water use amylase
Nematode:(Pratylenchus efficiency • Pre-harvest sprouting
neglectus) • Frost tolerance • Dough mixing
• Septoria tritici blotch • Salinity characteristics:
• Septoria nodorum blotch tolerance (strength, extensibility,
• Stem Rust mixing time)
• Stripe Rust • Flour water absorption
• Milling yield
• Quality for Asian
noodles
• Noodle manufacture
• Bread making quality

7
 1.1.2 Wheat Production in China

China’s population is just over 1.3 billion people and represents 20% of the world’s

population of 6.7 billion. Therefore maintaining or increasing wheat production is

important to China so that it can meet the needs of a growing population. China is the

world’s largest wheat producer, with a yield in 2007-08 of 109.9 million tons and has

a predicted yield of 114 million tons for the 2008-09 season (Table 1-5). After a poor

harvest in 2003/2004, the Chinese government has encouraged farmers to expand

wheat production with incentive policies. USDA (2008a) states that these incentives

include: tax reductions, direct subsidies for inputs and high quality seeds and

minimum support prices.

Table 1-5: World Wheat Production and Consumption compared to Australia and
China (units: thousand metric tons)
2004/05 2005/06 2006/07 2007/08 2008/09
Oct
Production
Australia: 21, 905 25,173 10,822 13,039 21,500
China 91, 950 97, 450 108,470 109,860 114, 000
World 625,738 620,851 596,304 610,883 680,200
Consumption
China 102,000 101,500 102,000 104,000 107,000
World 606,681 624,182 616,927 618,102 655,585
(Data source: USDA 2008b)

The Chinese Academy of Agricultural Sciences has divided the country into ten zones

of wheat production (Figure 1-4). These zones are determined by geographical

location which then governs sowing and harvesting times. Zone V: is the South-

western Autumn-sown spring wheat zone and includes most parts of the Sichuan

Province. The Chinese varieties used in this study were developed and grown in

Zone V.

8
Figure 1-4: Wheat production Zones in China. Separated by location and time of
planting

Legend:
I Northern Winter Wheat Zone
II Yellow and Huai River Valleys Facultative Wheat Zone
III Middle and Low Yangtze Valleys Autumn-Sown Spring Wheat zone
IV Southern Autumn-Sown Spring Wheat Zone
V Southwestern Autumn-Sown Spring Wheat Zone
VI Northeastern Spring Wheat Zone
VII Nothern Spring Wheat Zone
VIII Northwestern Spring Wheat Zone
IX Qinghai-Tibetan Plateau Spring-Winter Wheat Zone
X Xinjiang Winter-Spring Wheat Zone
(source: Figure 1.3 Wheat zones of China. Chapter 1: “A history of Wheat Breeding in China” He et
al. 2001)

9
Since 1949 wheat breeding in China has become more important and has proven very

successful in producing varieties with higher yield potentials, a reduction in plant

height which has aided lodging resistance and resistance to disease. For Zone V, the

Sichuan Province, the major breeding objectives as stated by He et al. (2001) are to

produce lines with a) high yield, short stature and with good lodging resistance; b)

resistance to stripe rust, powdery mildew and head scab; c) early maturity to fit the

multiple cropping system; c) drought tolerance for wheat sown in hilly areas without

irrigation, and tolerance to poor soil fertility; and d) preferred white kernel colour.

Producing varieties with a high yield potential is essential to maintain the necessary

increase in production. Establishing disease resistance in the breeding material is

very important especially with rust and powdery mildew being such a large problem

in China. Landraces have been used with lines from Italy, Mexico (CIMMYT), USA,

Australia, Canada and other countries to improve the varieties grown by the local

farmers (He et al. 2001).

The Ministry of Science and Technology of China (MOST 2006) has stated that

between the years of 2006 – 2020 China aims to produce 50 new varieties of what

they have termed “super wheat”. The reports explain that these varieties will increase

total wheat production by 30%. This equates to 33.3 million hectares with a

forecasted output of 10,500 to 12,000 kg/ha and yield of 10.45 tons/ha (china.org.cn

2006).

10
1.2 Breeding Objectives in wheat improvement

It is the goal of a wheat breeder to create new genotypes with improved features

compared to previous varieties (Poehlman and Sleper 1995). Parental material is

selected to fulfil breeding objectives relative to the environment and economic

market. A wheat breeder needs to consider the following when developing new

genotypes: increasing yield potential and stability; improving tolerance to biotic and

abiotic stresses; and maximising the end product quality.

1.2.1 Yield Potential

Yield of grain is of the primary importance to any cereal breeder as a high yield

potential ensures appropriate economic return to the grower and the ultimate success

of the variety. Yield potential and stability requires evaluation over several years and

in a range of locations/environments before a new variety can be released.

Traditionally, breeding for high-yielding potential is accomplished by the crossing of

selected genotypes to supply genes for yield potential through the generation of

transgressive segregates expressing a superior yield phenotype (Poehlman and Sleper

1995). Yield potential, a complex quantitative trait, takes into account a number of

morphological and physiological features. It is a trait that may benefit with the use of

modern molecular techniques when used with traditional breeding methods. It has

been reported that there is a need to develop varieties that have high yielding

potentials to ensure the rate of production equals or betters the rate of consumption.

To maintain this, improved breeding methods based on physiological criteria have

11
been implemented (Reynolds 1998). This physiological selection was based on the

genetic evaluation and mechanisms of stomatal aperture-related traits.

Fischer (2007) states that rate of yield progress (using conventional breeding

methods) is between 0.5-1% per year and is limited by the following agronomic

factors: available soil moisture, biological soil additives and seasonal climatic

conditions. In the past yield has increased through the establishment of new varieties,

improved farming techniques and the application of soil additives.

1.2.2 Heat and Drought Tolerance

Damage to wheat crops due to heat or drought stress is a common occurrence in the

Australian environment, therefore breeding for heat and drought tolerance remains an

important area for improvement. The effect of heat stress varies with the stage of

plant growth and the duration of the stress, but is most critical during the flowering

period, due to reduced pollen viability, stigma receptivity and seed formation

(Poehlman and Sleper 1995). Genetic control of heat and drought tolerance is a

combination of complex physiological and morphological traits and is therefore

difficult to evaluate. Improved water use efficiency is a mechanism by which wheat

varieties can support drought tolerance. An experiment of this type was carried out

by Najafian et al. (2004) where varieties were planted under water stress at three

locations to produce lines with high yield. McIntrye (2006) has been studying the

genetic basis of obtaining drought tolerance in wheat, and is concentrating the search

in genomic areas that code for flowering period, height, grain size and number and

stem carbohydrates. As there are many traits that can contribute to a wheat plant

12
tolerating a high level of water stress, it continues to be a very complex area of

research.

1.2.3 Disease Resistance

The development of wheat varieties with resistance to a range of disease pathogens is

a major objective for wheat breeding worldwide. Each crop disease is treated as a

separate breeding objective, as many have a high degree of pathogen race

specialisation (Poehlman and Sleper 1995). Australian wheat breeding programs

focus on a number of diseases, these include: Crown Rot, Root Lesion Nematode

(Pratylenchus neglectus), Septoria diseases (Septoria tritici blotch, Septoria nodorum

blotch) and rust diseases (stem rust, leaf rust and stripe rust) (GRDC 2008). In the

Northern Region of Australia, breeding programs are developing resistant lines to:

Yellow Spot, Common Root Rot, Barley yellow dwarf virus and Karnal Bunt (Smut)

(QDPI&F 2008).

Breeding for disease resistant varieties can be a complicated process as the resistance

may be controlled by a single or multiple genes and the disease pathogen can be

active with a number of races. To take leaf rust as an example, a good source of

partial resistance is contributed by the gene Lr34. To reduce the likelihood of this

resistance gene being overcome by a change in the pathogen; durable adult plant

resistance needs to be achieved. To undertake this, Singh and Rajaram (1992) suggest

adding the Lr34 gene with two or three additional slow rusting genes.

There are other methods to prevent or control diseases during the cropping season,

and these include: the use of partial resistant varieties, planting of disease free seed

stock, eliminating contamination from weeds and crop rotation (QDPI&F 2008).

13
 1.2.4 Quality

The market demands high quality products in all aspects of wheat production, from a

sound wheat kernel, to high quality milled flour. Wheat quality is a complicated area

of wheat breeding and Dewan (1988) states that quality can be broken down into four

important factors: protein content, grain hardness, flour dough strength and milling

quality (flour yield and flour colour). Milling procedures and end product

requirements are affected by varietal type and the environmental growing conditions.

Environmental growing conditions include the amount of rainfall, soil fertility and

other stresses. In Australia there are three groups of wheat cultivars grown; hard or

bread wheat; soft wheat and durum wheat. Flour milled from each of these groups

differ significantly for a number of quality components, some of which are: flour

yield, water absorption, particle size index and dough mixing time (Poehlman and

Sleper 1995). Martin (1996) states that there have been measures put in place in

breeding programs in the Northern regions of Australia to ensure that new varieties

comply with the quality requirements of all Australian wheat. The quality

requirements included bread wheats produced having a hard grain, high milling

quality, medium dough strength and a high extensibility. This high level of grain and

flour quality tailors Australian wheat for the domestic and export markets.

Within Australia a number of the large research institutes have concentrated research

projects on wheat and flour quality, some of which include: the Value Added Wheat

CRC, Molecular Plant Breeding CRC (WA), South Australian Research and

Development Institute (SARDI) and Queensland Department of Primary Industries

and Fisheries, along with collaborations with many Australian Universities. These

programs all concentrate on the enhancement of quality, and are generating and

14
optimising analytical techniques that will help wheat breeders, researchers, growers,

buyers and all processors of wheat in Australia.

1.2.4.1 Noodle Quality

There are a number of research programs established in Australia that concentrate on

perfecting the quality of wheat grain for many types of Asian noodles. Given that

China is a large importer of wheat from a number of countries, it is important to

research the areas that influence noodle production, colour and taste. This will ensure

Australia with a reliable source of high quality wheat for the growing export market

into China (Brettell 2006).

There are a number of quality requirements to be considered with the production of

Chinese noodles. He (1999) expresses the importance of protein type and quality in

Chinese noodle production compared to other wheat end-products consumed in

China. China has generated a set of quality standards for wheat flour used in these

end-products (Table 1-6).

Table 1-6: The National standard of wheat flour specifications for various end-
products consumed in China
Ash (%) Wet gluten Farinograph Falling
(%) Stability (m) Number (s)
Bread <0.60 >33.0 >10.0 250-350
Noodles <0.55 >28.0 >4.0 >200
Dumplings <0.55 28.0 - 32.0 >3.5 >200
Steamed <0.55 25.0 – 30.0 >3.0 250
Bread
(Data source: National standard for Wheat Flour, 1993 and He 1999)

15
Another important aspect of noodle quality is the texture and sensory perception of

cooked noodles. A number of research groups have concentrated their studies in this

area, and have used taste panels as a means of assessing the noodle quality (Hatcher et

al. 2002, Storlie et al. 2006, Zhang et al. 2005). Storlie et al. (2006) established a

taste panel to assess the influence the puroindoline genes and environment have on

texture of fresh Chinese noodles. They reported that both the environmental growing

conditions, and the puroindoline genes affecting grain hardness, had a significant

effect on noodle softness and texture. Researchers Hatcher et al. (2002) established

different requirements for suitable noodle palatability, illustrating that a fine flour

particle size and minimal starch damage during milling combines to give a superior

noodle after cooking. Other studies have continued using tests such as flour swelling

indexes as a predictive tool of noodle quality during the production and cooking of the

noodles (Hu et al. 2004, McCormick et al. 1991).

As early as 1970, Moss (1971) was researching the use of Australian wheats for

noodle production, with the emphasis of the study directed at white Japanese noodles

and yellow Chinese noodles. He found that noodle quality was affected in three

ways: 1. noodle whiteness/brightness decreased with increased protein levels, 2.

noodles had differing responses to boiling and absorption of water, 3. the intensity of

the noodle yellowness was due to the alkaline medium used. The basic parameters for

noodle quality have remained unchanged.

1.2.4.2 Noodle Types

A large proportion of Australian wheat exports enter eastern Asia, where its high

quality makes it ideal for a number of end-products (Brettell 2006). These include

16
white salted noodles, yellow alkaline noodles, instant noodles, steamed bread and

dumplings (Table 1-7). All of which are a staple part of the daily meals in many parts

of Asia.

Table 1-7: Wheat products consumed in China (Data source: He 1999)

Food Type % Noodle Variety %
Steamed bread including 46 Fresh noodle (home made) 33
flat bread
Noodles and dumplings 39 Dry noodles 24
(machine made)
Cookies and biscuits 6 Instant noodle 21
(machine made)
Western bread 3 Fresh noodle (machine made 13
available at store)
Others 6 Extended alkaline noodle 5
(La Mian)
Extruded noodle (Helu) 2
Frozen noodle and others 2

The homemade fresh noodle is the most popular noodle in China, as they can be

produced easily with basic equipment (He 1999).

White salted noodles are popular in Japan, South Korea and China. These noodles

can be found in a number of forms, including fresh, dried, boiled and steamed. These

noodles are required to have a smooth appearance and should be soft and elastic in the

mouth therefore starch and protein quality are important aspects (Bushuk 1998,

Crosbie et al. 1998). Yellow alkaline noodles (YAN) have the same mouth feel

requirements as white salted noodles, but require high yellow flour colour that is

stable. This yellow colour may then reduce the amounts of additives (alkaline salts)

needed in the production of the YAN (Crosbie et al. 1998).

17
 1.2.4.3 Noodle Colour

Colour is an important sensory aspect of Asian noodle production and maybe the most

important as it is the first assessment made of noodles by a consumer (Bhattacharya et

al. 1999, Crosbie et al. 1998, Mares and Panozzo 1999). Colour and discolouration of

flour and dough products are affected by a number of quality factors. These being

genotypic and environmental effects, protein content, colour of wheat pericarp,

inherent pigmentations, flour processing conditions and storage conditions

(Bhattacharya et al. 1999). Flour colour can also be affected by the milling of the

wheat, whereby bran flakes/ash can enter the flour during the grain milling thereby

hindering the whiteness of the flour (Brettell 2006). Flour colour can be affected by

its inherent accumulation of carotenoids in the wheat grain endosperm. These yellow

pigment components, usually of the Xanthophyll type, can produce a colour range

from bright white to creamy yellow (Brettell 2006). Colour is not always constant;

therefore colour stability of flour and end-products needs to be considered. A large

amount of research has concentrated on the causal effects of colour and colour

stability, and it is effect on the appearance of noodles, but it is the polyphenol oxidase

(PPO) activity and Xanthophyll content that will be assessed in this chapter.

1.2.4.3.1 Impacting factors on noodle colour i) Polyphenol Oxidase

Activity Levels

Discolouration of fresh noodles has been highlighted as a significant problem in the

production of noodles in China. It has been shown that Polyphenol oxidase (PPO)

activity is a cause of the time-dependent discolouration of fresh noodles

18
(Bhattacharya et al. 1999, Demeke et al. 2001, Kruger et al. 1992, Mares and Panozzo

1999). It has been reported that PPO can be found on the seed coat of wheat grains

and it is an enzyme which catalyses the oxidation of phenolic acids to quinones.

These react with amines and thiol groups or undergo self-polymerisation to produce

dark brown or black pigments (Demeke et al. 2001, Steffens et al. 1994) (Figure 1-5).

Figure 1-5: The reaction catalysed by Polyphenol oxidase (a: Data source:
http://plantphys.info/plant_physiology/enzymelab.html) (b: Data source: Rajesh et al. 2005)

a)

(di-hydroxyphenol)

b)

Many studies have been undertaken to both understand the mechanism by which this

enzyme works, and to measure the PPO activity of harvested grain products.

Kruger (1976) has found evidence for the presence of 12 isozymes of PPO in different

parts of the wheat kernel. This study also found that as the kernel approached

maturity, the level of PPO in the endosperm dropped and the PPO in the germ and

scutellum increased (Figure 1-6). These findings are consistent with the research of

19
Taneja et al. (1974) who also reported that the number and type of isozymes decline

once the seed is fully developed.

Figure 1-6: Diagram of the wheat grain (Data source: http://www.nutrition.org.uk)

Demeke and Morris (2002) has studied the PPO gene and characterised the gene in

wheat. This research has produced a significant polymorphism between DNA

fragments amplified from different cultivars. This polymorphism can clearly separate

cultivars with a high PPO from those with a low PPO activity. The Polymerase Chain

Reaction (PCR) was performed by these researchers using genomic DNA and

oligonucleotides designed from the conserved binding regions in all PPO genes. It is

the conserved copper binding regions that are responsible for the interaction with

oxygen and phenolic compounds (Steffens et al. 1994). Demeke and Morris (2002)

were successful in amplifying a DNA fragment which showed a high level of

sequence homology to other plant PPO genes in a BLAST search.

Research has also taken place to determine the location of the PPO genes on the

wheat genome. Mares and Campbell (2001) identified Quantitative Trait Loci (QTL)

20
associated with PPO activity on chromosome 2A and 2D using a doubled haploid

population. Demeke et al. (2001) identified QTLs located on chromosomes 2A, 2B,

3D and 6B in a recombinant inbred line population. This research group found

different QTLs were identified using different PPO substrates, L-DOPA and L-

tyrosine, whereas Mares and Campbell (2001) only employed L-tyrosine to generate

their phenotypic data.

Raman et al (2005) have also identified a significant QTL on 2A for PPO activity in a

doubled haploid population, again using both PPO substrates. In order to develop

functional gene markers Raman et al. (2007) have developed a SNP and a CAPS

marker tightly linked to the QTL identified on chromosome 2A.

In the past, selection on wheat germplasm for varying PPO activities has been

conducted through the development of whole seed assays using particular PPO

substrates to generate high quality phenotypic data for line selection and QTL

identification research (Bernier and Howes 1994, Demeke et al. 2001, Demeke and

Morris 2002, Mares and Campbell 2001, Raman et al. 2005, Raman et al. 2007).

1.2.4.3.2 Impacting factors on noodle colour ii) Xanthophyll content

Plant tissue contains pigments that contribute to the colour of that tissue. In wheat

flour, the pigment that constitutes part of its yellow colour is the carotenoid pigment,

xanthophyll (Moros et al. 2002). Xanthophyll is a complex structure of alternating

double and single bonds and it is this structure that contributes to the compounds

shape, chemical reactivity and light absorbing properties (Figure 1-7).

21
Figure 1-7: Chemical structure of Xanthophyll (Data source:
http://chemistry.about.com/library/graphics/blxanth.htm)

The light absorbing properties of xanthophyll produces various shades of red to

yellow colours in foods (Moros et al. 2002). Lepage and Sims (1968) developed a

method to characterise the carotenoid pigments in wheat flour, demonstrating that

they are composed of xanthophylls in the form of free lutein and its esters. Total

xanthophylls can be extracted from wheat flour using water-saturated butanol or

methanol (Burdick 1956). This method can be used as a predictive tool to assist in the

selection of wheat flours for the Asian noodle market since it is now recognised that

the variation in noodle ‘yellowness’ is due to the presence of xanthophylls (Mares et

al. 1997). It has been shown that xanthophyll content has an impact on the stability of

colour in fresh noodle sheets. Mares et al (2000) reported a decrease in brightness

and whiteness in fresh noodles over time. This was associated with a 20 - 37%

decrease in xanthophyll content in the first 6 hours of noodle production. After this

time the noodle colour and the xanthophyll content remained more constant. This

study also indicated that the genotypes that contained the most xanthophyll had

relatively low rates of darkening.

Miskelly (1984) first detected a correlation between flour colour and Asian noodle

sheet colour. Subsequent research by Elouafi et al. (2001); Mares and Campbell

22
(2001) and Parker et al. (1998) have identified chromosomal locations of linkages

between yellow pigment in wheat to chromosome 7. Parker et al. (1998) identified

QTL’s for flour colour on 3B and 7A in a ‘Schomburgk x ‘Yaralinka’ single seed

descent population accounting for 13% and 60% of the genetic variation respectively.

1.3 Breeding methods used in wheat improvement

There are a number of breeding methods that can be used to select agronomically

superior lines from populations containing desired traits. The lines become varieties

only after they are selected from segregating populations and are carried through until

homozygosity is reached following at least one hybridisation (Stoskopf et al. 1993).

1.3.1 Pedigree Breeding Method

The pedigree system is a widely used method which produces new superior varieties

and pure lines for breeding programs (Moreno-González and Cubero 1993). This

method is used to introduce disease resistances and simply inherited traits into new

varieties (O'Brien and De Pauw 2004). The F2 generation, produced by crossing two

lines both contributing beneficial alleles, is the first opportunity for selection (Figure

1-8). Plants are then re-selected in each subsequent generation until a level of

homozygosity is reached and the plants express the desired phenotype (Poehlman and

Sleper 1995, Stoskopf et al. 1993). Pedigree breeding is a simple and effective

method, but is also labour intensive as extensive field trials are required (Moreno-

González and Cubero 1993, Stoskopf et al. 1993).

23
 1.3.2 Single Seed Descent

The Single Seed Descent method has been used in breeding programs when large

numbers of lines need to be advanced in a short period of time. Only one seed is

required from each plant each generation and this permits progression to

homozygosity in limited spaces (Figure 1-8). Controlled environmental conditions in

growth chambers and glasshouses allow more than one generation to progress a year

thereby shortening line development time (Moreno-González and Cubero 1993,

Poehlman and Sleper 1995). This method reduces the loss of genotypes with

desirable alleles during the segregating generations, by maintaining a large number of

F2 plants and advancing them through another four to six generations (O'Brien and

De Pauw 2004). Other than ensuring speed and simplicity, another advantage of the

Single Seed Descent method is that it is suitable for traits that have low heritabilities

and where visual phenotyping is difficult (Moreno-González and Cubero 1993).

Figure 1-8: Schematic Diagram of the Pedigree and Single-seed decent breeding
methods (Source: O'Brien and De Pauw 2004)

Pedigree with two Single-seed descent

generations per year

Choose parents that combine all the desired Time Line

Winter Cycle Summer Cycle

Parent A x Parent B Parent A x Parent B

F1 Self to F2 F1 Self to F2 Plant in End year 1

Glass House

Plant F2, select F3 seed Select for

for agronomic increase & F3 harvest F4 harvest & agronomics & End year 2
& disease selection & plant increase disease

F4, select for F5 seed

disease & yield increase Regional yield & quality End year 3
evaluations

Regional yield & Year 4

quality evaluations

24
 1.3.3 Recurrent Selection Method

Recurrent selection is a method that combines the techniques of both single seed

descent and pedigree breeding to improve adapted genotypes in breeding programs

(Stoskopf et al. 1993). “Recurrent selection is a population improvement method that

increases the frequency of favourable alleles while maintaining genetic variation in

breeding programs” (Doerksen et al. 2003). To accomplish this increased frequency

of desirable alleles, superior genotypes are selected and intercrossed to produce the

next generation (Poehlman and Sleper 1995). The technique requires artificial

emasculation in self-fertilising plants, therefore being more easily used in cross-

fertilising plants (Moreno-González and Cubero 1993).

1.3.4 Backcross Breeding

Backcrossing is a technique that allows important genes to be transferred from one

genotype into another with relative ease. Backcrossing is a form of recurrent

hybridisation which substitutes a desirable allele for a particular trait with the current

allele in an adapted cultivar. The method is carried out with an initial cross between a

donor and a recurrent parent. The progeny produced in the F1 generation is then

crossed with the recurrent parent. The number of backcrosses required is dependent

on the performance of the donor alleles within the recurrent parent genotype and the

proportion of homozygosity needed (Moreno-González and Cubero 1993). After the

first backcross to the recurrent parent, the progeny have 50% of the alleles from the

recurrent parent. After only three more backcrosses to this recurrent parent the

progeny will have 93.75% of the alleles from the recurrent parent (Figure 1-9). The

types of genes that can be considered for this technique are: single dominant genes,

25
single recessive genes and polygenes. The method can become more complicated

when the trait is more complex (Poehlman and Sleper 1995). When considering

several genes, they can be transferred simultaneously or in a stepwise progression. To

introgress two or more genes stepwise, each gene is backcrossed into the recurrent

parent separately and then assessed at the end. Simultaneous selection of several

genes is a more complex method and requires large plant populations.

Stoskopf et al. (1993) explains for a backcross population to be successful, the

following is required; an ability to identify the trait being transferred; a simply

inherited trait; and a sufficient number of backcrosses carried out to recover the

recurrent parent genotype.

Figure 1-9: Backcrossing Technique used in self pollinating plants. A cross between
donor parent A and recurrent parent B. (Data source: Poehlman 1994)

Parent A Parent B
Donor Parent X Recurrent Parent

F1
‘AB’
50% genes from Parent B

1st Backcross

BC1
75% genes from Parent B

2nd Backcross

BC2
87.5% genes from Parent B

26
 1.3.5 Doubled Haploid Populations

The development of doubled haploid populations has become important in wheat

breeding in recent years because of their usefulness in molecular marker development

and identification for desirable traits. The major benefit of doubled haploids is that

homozygosity is achieved in a very short time compared to other breeding methods

(Stoskopf et al. 1993). To produce these populations, a number of steps are required.

Firstly, haploid plants are generated through crossing of wheat spike florets with

collected maize pollen. The maize chromosomes are lost and haploid wheat embryos

form which will abort if left in the developing grain. Consequently the forming

haploid embryos are “rescued” to plant growth media and incubated under a sequence

of conditions which stimulate embryo germination and plantlet formation. These

small sterile plants are then transplanted and grown in soil. At the tillering stage they

are treated with colchicine which artificially doubles the chromosomes in some cells

within the crown, generating diploid tillers that produce fertile seed (Figure 1-10)

(Kammholz et al. 1996, Kammholz (2001) PhD). These fertile seed are homozygous

at all alleles, and therefore beneficial for research purposes and for the release of

varieties in a shorter time period (O'Brien and De Pauw 2004).

27
Figure 1-10: Schematic Diagram of production of Doubled haploids, (Source: O'Brien
and De Pauw 2004)
Doubled Haploids
Parent A x Parent B

Time Line

F1 Pollinate with maize

Embryo rescue

Double chromosome
number with colchicine
End year 1
Homozygous plants, select
for agronomic & disease

End year 2

Regional yield &

quality evaluations

1.4 Genetic markers for wheat improvement

1.4.1 Types of Markers

A genetic marker is a score-able identifier linked to a trait of interest at particular

chromosomal location.

There are three broad classes of markers, these being morphological, biochemical and

molecular. Morphological markers have been associated with traits that can be

visually assessed, but these types of markers have limited success in wheat.

Secondly biochemical markers have been used within wheat breeding, but their use is

limited by the lack of functional biochemical markers available. Biochemical markers

can be identified by assays for the production of particular proteins. For example

28
glutenin proteins can be extracted and visualised by electrophoresis (Eagles et al.

2001). Because of the limitations of the above marker types in recent years there has

been a move towards using molecular markers. There are a number of different types

of molecular markers frequently used in wheat research. These being; random

amplified polymorphic DNAs (RAPDs), restriction fragment length polymorphisms

(RFLPs), amplified fragment length polymorphisms (AFLPs), simple sequence

repeats (SSRs or microsatellites) and single nucleotide polymorphism (SNPs).

RAPDs and RFLPs were frequently used in the early years of molecular research into

wheat but were replaced by SSRs and SNPs because of their low levels of

polymorphisms and difficultly in use. AFLPs have been used less in recent years as

they restrict researchers to only dominant markers (Landjeva et al. 2007). Overall

SSR markers have been a very popular marker type in wheat genetic studies as they

are highly polymorphic, reproducible and are co-dominant in nature (Röder et al.

1995). As breeding programs are demanding large number of lines genotyped by a

number of different markers, SNPs are becoming more important (Landjeva et al.

2007). Varshney et al. (2005) states that expressed sequence tag (EST)-derived SNPs

and SSRs are highly functional molecular markers as they have been developed from

gene sequence data. These markers have an advantage over other marker types

because they are linked to the desired trait allele. This is an important step in

generating true associations between markers and QTLs (Quantitative Trait Loci),

which is linked to a particular gene of interest. The closer the marker is to being

perfect (the gene), the more useful they are to breeders.

The usefulness of any molecular marker relies upon its ability to reveal

polymorphisms within important breeding material and to make breeding programs

more efficient than those currently in use.

29
 1.4.2 Applications of markers in wheat breeding

Both Babu et al. (2004) and Francia et al. (2005) have reviewed the number of

factors/requirements that influence the success of marker application in breeding

programs, and these are: a) a genetic map with an adequate number of uniformly

spaced markers for the location of desired QTL/s of major genes; b) a tight

association of desired QTL/s or major gene of interest and flanking markers ; c) an

adequate recombination between the markers and the rest of the genome; d) an ability

to analyse a number of plants in a time and cost effective manner. Young (1999)

states that scientists can maximise the efficiency of DNA markers by using them to

introduce novel traits of interest into breeding programs, rather than using them to just

perform tasks of conventional breeding.

Before molecular markers can become a mainstream asset to breeding programs, their

individual usefulness has to be demonstrated. It is very important to establish a link

between the markers identified and the QTLs/traits they are associated with. To do

this marker validation has to be carried out for all traits, over a number of

environments and in multiple genetic backgrounds to test the robustness of these

markers.

Christopher and Banks (2002) explains the necessary steps for marker validation as:

“1) markers must be assessed for technical robustness and transferability; 2) markers

must be assessed for polymorphism; 3) markers must be assessed in a cross involving

local breeding material to establish that a similarly close linkage exists as has been

identified in the original mapping population.”

30
There has been a large input of funding for the development and identification of

molecular markers useful to plant breeders. The next step is for these markers to be

used as a tool to predict phenotype and enhance the development of breeding lines

towards new varieties. William et al. (2007) describes the ways that breeding

programs can benefit from the introduction of molecular markers: a) to pyramid genes

for specific traits; b) to select for recessive genes; c) to use when phenotypic

screening is difficult; d) to use when the environment can affect the phenotype; e) to

be used when the heritability of a particular trait is low and f) can be used at early

growth stages; with seed or young plant tissue.

There are a number of molecular markers that have been used on a regular basis in

wheat breeding programs across Australia to assist the introduction of specific traits

into elite breeding material (Table 1-8).

Table 1-8: Markers that are currently used in wheat breeding programs across
Australia

Traits assessed with markers used Australia Marker

wide

Cereal Cyst Nematode resistance Cre1; Cre3

Pratylenchus neglectus resistance Rlnn1
Noodle Quality (Granule bound starch Wx-A;Wx-B1;Wx-D1
synthase)
Boron Tolerance Bo1; Bo2
Stem Rust resistance Sr2; Sr39; Sr36; Sr24/Lr24
Grain Protein Nor1
Coleoptile length Rht8
Tiller inhibition Tin
Barley yellow dwarf virus Bdv2
VPM rust resistance Sr38/Lr37/Yr17
Flour colour FC3
(Data source: Eagles et al. 2001)

31
Christopher and Banks (2002) has reported that markers linked to traits important to

the Northern wheat region of Australia are also being integrated into the breeding

programs. These traits being: rust resistances, russian wheat aphid resistance, late

maturity alpha-amylase activity (absence of), polyphenol oxidase activity, grain

hardness, semi-dwarfing status, crown rot resistance, root lesion nematode resistance

(P.thornei), yellow spot resistance, pre-harvest sprouting resistance, black point

resistance, max extensibility, flour milling yield and flour water absorption.

1.4.3 Construction of genetic linkage maps

As a very large genome of 16x109 base pairs, undertaking genome wide studies on

wheat is a difficult and time consuming effort. A number of studies have been

successful in developing large numbers of microsatellite (SSR) markers with the

purpose of generating wheat linkage maps. The Wheat Microsatellite Consortium

(WMC) is an international collaboration with the main objective of developing

microsatellite markers to be mapped on the International Triticeae Mapping Initiative

(ITMI) map (W7984 x Opata85). This mapping population has been used to map 396

SSRs which were developed through WMC, 279 SSRs developed by Röder et al.

(1998), 55 SSRs by Pestsova et al. (2000), 66 SSRs by Gupta et al. (2002), 53 by

Stephenson et al. (1998), 337 SSRs by Sourdille et al. (2001), 144 SSRs by Harker et

al. (2001) and 347 SSRs by Song et al. (2005).

The process of generating a genetic linkage map is made possible by the above

numbers of microsatellite markers available. Three major steps are required to

construct a linkage map; 1) Development of a mapping population; 2) Screening of

32
the markers against the parents of the population; 3) Generating genotypic data from

the population for each polymorphic marker.

There are a number of population types that can be used for mapping, these being,

doubled haploids, Recombinant inbred lines, backcross populations and F2

populations (Langridge and Chalmers 2005). The mapping population is dependent

on the phenotype or specific traits that are being investigated, and it is the selection of

parents that holds the most importance in this step (Grandillo and Fulton 1998). The

size of the population has a large affect on the linkage map, and its ability to generate

the correct marker order. Mapping populations can vary in size, as in the doubled-

haploid populations generated by Kammholz et al. (2001), where the five populations

produced contained numbers up to 321 lines. Wu et al. (2003) states that the ordering

power of markers in mapping populations depends on the population size and the

marker distances. If the marker distance is to be 5cM apart, then the mapping

population needs to consist of 200 lines. If the marker distance is increased to 10cM

then the required population size is a more manageable 150 lines.

Mapping populations generated by wide crosses will generate the high level of

polymorphism required in the marker screening step. The polymorphic markers are

then assayed across the entire population to generate the genotypic data required to

produce a linkage map. Computer programs, most of which are freely available over

the internet are used to perform the linkage analysis. Some of these programs

include: MapMaker/EXP (Lander et al. 1987); JoinMap (Stam 1993); MapManager

QTX (Manly et al. 2001); RECORD (Van Os et al. 2005) and Q-gene (Nelson 1997).

33
The analysis required to link markers together is based on the maximum likelihood

ratio (LOD) and it is a LOD score value of >3 that is generally considered to indicate

a significant linkage between markers (Collard et al. 2005). It is suggested by

Hackett and Broadfoot (2003) that the effects of scoring errors reduce the accuracy of

maps impacting the length of the linkage map as well as the marker order. Therefore

managing linkage maps as indicated by Lehmensiek et al. (2005) will ensure that

genotypic data collected will not hinder any further analysis.

1.5 Quantitative Trait Loci (QTL) analysis

1.5.1 Methods of detecting QTLs

QTL analysis is carried out to identify a linkage or association between a particular

phenotype and the genotype of markers (Collard et al. 2005). The three methods for

detecting QTLs used on a regular basis are single-marker analysis, simple interval

analysis and composite interval analysis.

Mapping software like Map Manager QTX allows all three of these methods to be

carried out. Single-marker analysis is a linear regression that tests for an association

between the phenotypic values with the genotypic data of a single marker locus

(Manly et al. 2001). The linear regression produces a R2 value which explains the

amount of phenotypic variation being explained by that particular marker. Another

method, Interval Mapping can be carried out to identify a QTL and to estimate its

position on the genome. Interval mapping analyses two markers at a time that are

located next to each other on the same chromosome (Lander and Botstein 1989). The

34
next step on is Composite Interval Analysis which Manly et al. (2001) explains as a

method of using the most significant QTL to control the effects of subsequent

significant QTLs. This method was developed by Zeng (1994) to undertake QTL

analysis and identification by taking into account the effects of possible multiple or

linked genes.

1.5.2 Factors affecting QTL analysis

Reliable identification of QTLs is very important if these QTLs are to be used in

Marker assisted selection. The accuracy of QTL analysis can be hindered by a

number of factors, one of which is the reliability of the related phenotypic data. To

detect QTLs with small effects, having conclusive phenotypic data is essential

(Collard et al. 2005). Asins (2002) has reviewed the use of QTLs in plant breeding

and has highlighted other factors that can affect QTL analysis. These are: the marker

order of the linkage map, environmental variation, and epistatic effects from other

QTLs. Accurate mapping and linkage map generation can be supported by an

appropriate experimental design, population size and robust molecular markers.

When the environment and other QTLs/genes are influencing the phenotypic and

genotypic data advanced methods of QTL analysis including Multiple QTL Mapping

(Jansen 1993) and Composite Interval Mapping (Zeng 1994) can assist in improving

QTL identification.

Young (1999) emphasises that for high quality QTL analysis and identification of

robust DNA markers, the following needs to be maintained: a) high quality scoring

35
methods; b) large population sizes; c) multiple replications and environments and d)

validation of markers.

1.6 Marker Assisted Selection (MAS)

Through the aid of molecular markers, Marker Assisted Selection (MAS) is a

technique that selects plants with particular genes/QTL involved in the expression of

traits of interest (Babu et al. 2004). With the development of molecular technologies,

it is now possible to combine molecular markers with phenotypic data collected from

the field, to increase the efficiency of selecting superior lines. A modified method of

using markers in the early generations and combining the phenotypic selection in the

later generations has been recommended as an efficient system to integrate MAS into

breeding programs (Liu et al. 2004). MAS has been indicated to be a useful

technique at the times when conventional phenotyping procedures are too expensive,

time consuming or unreliable (Francia et al. 2005, Koebner and Summers 2003).

MAS in plant breeding has the ability to save time by selection of genotypes at early

stages of plant development. This method also assists the minimisation of linkage

drag, selection for traits with low heritability, the introgression of specific traits of

interest and assists gene pyramiding into elite varieties (Collard et al. 2005, Francia et

al. 2005, Hospital and Charcosset 1997). There have been a number of simulation

studies performed to assess the effectiveness of MAS as a breeding tool. The main

conclusions from these studies are that in large populations, when heritability values

are low and the markers are relatively close to the QTLs, MAS could be more

efficient than pure phenotypic selection (Edwards and Page 1994, Hospital et al.

1997, Lande and Thompson 1990, Moreau et al. 2004, Moreau et al. 1998). It has

36
been shown that using molecular markers at early stage generations can be a very

efficient use of MAS, as the rate of fixation of QTLs with larger effects is faster with

markers compared to phenotypic selection (Koebner and Summers 2003, Kuchel et al.

2007). Also applying MAS at early generations reduces the complication of

recombination between molecular markers and the trait of interest/QTL (Edwards and

Page 1994, Hospital et al. 1997, Ribaut and Betrán 1999).

1.6.1 Gene/QTL Introgression

In order to use MAS successfully great care is needed to choose the most appropriate

marker. Having a molecular marker located within the gene being transferred is the

most favourable strategy for MAS (Babu et al. 2004, Dubcovsky 2004, Francia et al.

2005). It is generally difficult to find these ideally located molecular markers

especially when considering polygenic traits. In these instances a genomic region

carrying the QTL or genes is transferred using two flanking markers. It is important

that the distance between the flanking markers is no larger than 20cM. The chance of

recombination events occurring lessens when the markers are close to the gene (Babu

et al. 2004, Edwards and Page 1994, Francia et al. 2005, Moreau et al. 1998). It is

agreed that tightly linked markers is a key element of MAS having an advantage over

pure phenotypic selection. This advantage cannot be maintained as the distance

between marker and QTL increases (Chao and Ukai 2000, Edwards and Page 1994).

Francia et al. (2005) states that there are limitations to using QTL mapping

information in breeding through the use of MAS. These limitations include: accurate

identification of the QTLs controlling the trait; uncertainty of the QTL position;

incorrect measurement of the QTLs effect; limitations of QTL-marker associations

37
over a range of breeding material; the possibility of loosing the target QTL through

double crossover events during MAS; and the difficulty in assessing other limiting

effects. Manipulating complex polygenic traits through MAS is a more complicated

effort. When several genes are involved the plant tends to exhibit smaller phenotypic

effects, therefore making it more difficult to capture using conventional methods as

well as using MAS (Ribaut and Hoisington 1998). There have been successful studies

completed by a number of researchers worldwide. The first studies into this area were

those that simulated MAS through computer programs in different plant species to

assess the possibility of introgressing one or more genes/QTLs into elite breeding

material (Frisch and Melchinger 2001, Knapp 1998, Ribaut et al. 2002, Reyes-Valdés

2000). In recent years there have been a number of studies that have introgressed

genomic regions from wild relative germplasm (Fulton et al. 1997, Robert et al.

2001); and QTLs (Toojinda et al. 1998, Tomar and Menon 1998) into useful and

agronomically successful varieties in tomato, barley, wheat and other plant species.

1.6.2 Marker Assisted Backcrossing

Marker-assisted Backcrossing has been a means of transferring genomic regions from

a donor parent (variety or wild relative germplasm) into elite breeding material.

Marker assisted selection can be very useful backcross breeding when it is required to

select the lines with the necessary amount of target genomic material but an otherwise

high proportion of recurrent parent germplasm. This reduces the amount of ‘linkage

drag’ occurring. Markers can also assist in selecting the lines that have undergone

recombination between the recipient DNA and the target gene/QTL introgressed

(Babu et al. 2004, Francia et al. 2005).

38
Under conventional backcross breeding methods, recovery of all the recurrent

parental genome would typically take 6-8 backcross generations (Allard 1960, Collard

et al. 2005). With the aid of tightly-linked flanking markers conferring a trait of

interest and background (recipient DNA) marker selection, the efficiency of recurrent

parent recovery is greater and can be reduced down to three generations (Arús and

Moreno-González 1993, Frisch et al. 1999, Frisch and Melchinger 2001). Therefore

with a combination of backcrossing techniques and molecular markers, a large step

ahead can be made within wheat breeding programs.

Researchers have successfully put the theory of MAS into practice in a number of

plant species. Through the use of marker-assisted backcrossing, Steele et al. (2006)

was able to successfully introgress five chromosomal segments into a rice variety. To

complete such a feat, this research undertook many steps to complete their task; initial

mapping, three backcrosses, two further crosses, thousands of marker assays,

phenotyping and validation. All of which took eight generations and six years to

produce one ideotype that contained all the five target regions in an elite rice variety.

Kuchel et al. (2007) also had success in transferring disease resistant traits into an

important wheat line. They achieved this with a shorter period of time than other

researchers because they introduced a doubled haploid step into their methods. This

technique may not be an ideal step to use in all breeding programs as it does increase

the cost. In a similar manner, another successful example of MAS was reported by

Toojinda et al. (1998) whereby a disease resistance QTL introgressed into a barley

line.

39
Advanced backcross QTL analysis (AB-QTL) is a technique that was proposed by

Tanksley and Nelson (Tanksley and Nelson 1996, Tanksley et al. 1996) as a

mechanism to transfer QTLs of important traits into a variety using wild species

germplasm. Other studies have been carried out whereby genomic regions and QTLs

have been transferred through marker assisted backcrossing to enhance breeding lines

or elite material (Huang et al. 2003, Huang et al. 2004, Tomar and Menon 1998,

Yousef and Juvik 2002). Varshney et al. (2005) states that large populations sizes are

required for selection of the useful alleles and maybe a limitation of AB-QTL.

1.6.3 Incorporating MAS into breeding programs

Markers have been used to characterise parental material used within breeding

programs, allowing this information to improve the efficiency and effectiveness of

selection in crosses and in segregating lines throughout the selection process (William

et al. 2007). The integration of molecular marker technologies into conventional

breeding strategies could produce genetic gains with greater speed and efficiency

(Babu et al. 2004). Babu et al. (2004) also suggests using molecular markers in

breeding strategies at one key selection step can maximise their impact as a prediction

or diagnostic tool. It provides a reliable method of selection without the effect of

environmental conditions. It is a method that is becoming more attractive with wheat

breeders and is being used for traits that are difficult to screen for through traditional

phenotyping methods (O'Brien and De Pauw 2004).

Francia et al. (2005) states that one of the most important aspects to ensure MAS

efficiency in large breeding programs is the implementation of high-throughput

molecular marker techniques that are efficient in both time and cost. Also, obtaining

40
the most effective marker for the trait of interest is imperative. New technologies into

comparative genetics may identify positions of genes associated with the QTLs

previously discovered. Having more closely linked markers will then improve MAS

for complex traits.

A successful integration of MAS into wheat breeding programs has occurred in

Australia with the introduction of resistance to cereal cyst nematode into wheat lines

(Eagles et al. 2001). The two Cre genes have assisted the protection of wheat

varieties from a disease that can significantly reduce wheat production in southern

Australia. Testing for cereal cyst nematode is a high cost, unreliable bioassay,

therefore this pest was an ideal test case for MAS.

Dubcovsky (2004) reports that MAS programs are a positive way to link wheat

genomics with wheat breeding. It is important that the transfer of valuable genes into

selected breeding programs continue therefore improving wheat varieties available for

commercial growers. A combined effort between wheat researchers and breeders has

shown to be successful in Australia through the National Wheat Molecular Marker

Program implemented in 1996. Likewise in the United States of America, with the

program MASwheat, which has allowed well documented genomic techniques to be

easily accessed so that they can be used to implement MAS within wheat breeding

programs worldwide (Dubcovsky 2004).

Computer simulations have been used to investigate the effectiveness of using

markers to increase genetic gain and have found combining MAS with phenotypic

evaluation a useful combination (Lande and Thompson 1990, Hospital et al. 1997).

41
Hospital et al. (1997) has found alternating MAS with and without phenotypic

selection over several successive generations to be more efficient that using purely

phenotypic selection, as well as being more cost effective. Integrating phenotypic

assessment with MAS allows the markers to be re-evaluated at each generation.

Without re-evaluating the markers, the additional gain provided by MAS rapidly

decreases with several cycles of selection due to segregation of marker and trait

(Hospital et al. 1997).

Ribaut and Hoisington (1998) state that using MAS to incorporate single-gene traits

of interest has proved successful, however when considering polygenic traits, more

innovative strategies need to be developed. To date MAS has been used in breeding

programs to transfer genes or QTLs of polygenic traits when the phenotypic testing

for these traits is difficult. Holland (2004) states that implementation of MAS for

complex traits is dependent on accurately identifying, locating and estimating the

effects of the QTLs. These effects also need to be seen over a number of important

breeding lines to be a useful tool in plant breeding.

1.7 Rationale for the Current Study

A comprehensive study was carried out by Mares and Campbell (2001) whereby

characteristics of flour and noodle colour were investigated in crosses of Australian

wheat varieties. Major parameters linked to flour and noodle colour were found to be:

Flour protein levels, Flour colour L* (brightness) and b* (yellowness), PPO activity

and Xanthophyll content. The discolouration of fresh White Salted noodle and

Yellow alkaline noodle sheets was investigated over defined periods of time,

42
measuring brightness and yellowness of both noodle types. QTLs were identified in

three doubled haploid populations, one population being the Sunco x Tasman

population. In this population, a highly significant QTL was identified on

chromosome 2D for PPO activity, located in the same area contributing to the time

dependent darkening and brightness of noodles, and to a lesser degree on a QTL

located on chromosome 2A. QTLs significantly affecting Xanthophyll content were

also identified in the Sunco x Tasman population, located on chromosomes 3B and

7A. These QTLs are located within regions corresponding to b* (yellowness) levels

within flour and noodle sheets.

This evidence for links between PPO activity and noodle colour stability and between

Xanthophyll content and noodle colour provided an excellent starting place for the

current project.

When the current project commenced research into the practical aspects of Marker-

assisted selection was limited. The majority of the research into this area were

computer simulation studies concentrating on the comparison between the traditional

phenotypic methods and the marker-assisted methods, using the backcrossing

breeding method (Kruger et al. 2002, Ribaut et al. 2002). These studies evaluated the

efficiency of MAS as well as the ability to transfer genes/QTLs from one genetic

background to another (Frisch et al. 1999, Frisch and Melchinger 2001). These

studies evaluated the population sizes required to carry out MAS successfully

(Bonnett et al. 2002, Chao and Ukai 2000, Moreau et al. 1998).

43
This master’s research was part of a larger collaborative, ACIAR funded project

(Wheat Improvement in Sichuan Province: Application of Modern Breeding

Technologies) in which I was employed as a research assistant. The ACIAR project

combined the expertise of a number of scientists from the University of Southern

Queensland, the Queensland Department of Primary Industries & Fisheries – Leslie

Research Centre, the University of Adelaide, Waite Campus and the Crop Research

Institute within the Sichuan Academy of Agricultural Sciences (SAAS) located in

Chengdu, China. This project ran from July 1998 to April 2006 and was initiated to

investigate a number of key areas in wheat breeding important to Australia and the

Sichuan Province, China. There are a number of factors that influence the wheat

production and quality in Sichuan and have been identified as necessary areas of

research. The project concentrated on the following areas that were limiting yield and

production of wheat in Sichuan: increasing genetic gains, disease resistance,

intermittent reproductive sterility, pre-harvest sprouting and wheat quality.

Quality of wheat grains and its end products is important to both countries and this

project was successful in utilising the technologies present in Australia to assess and

improved the quality of Chinese fresh noodles in China. This research focussed on

identifying effective methods to develop wheat varieties exhibiting high levels of

noodle quality. To enhance this quality improvement, molecular markers were used

to assess quality attributes in the germplasm grown in Sichuan. Information and

germplasm gained through this ACIAR project enabled this current study to be carried

out, also highlighting the potential to further improve Chinese varieties in the areas of

colour and colour stability through the use of Marker-assisted selection.

44
This current research was carried out on populations that were generated through the

original ACIAR project combining important germplasm from Australian and China.

The initial crosses and first backcross was conducted by the Project Officer Dr. Eric

Storlie in 2001. The varieties selected for these populations were the Australia

varieties: Sunco and Tasman, and the Chinese varieties: Chuanmai 22 and Mianyang

11.

The Australian varieties were selected for a number of reasons, firstly because they

are used often in wheat breeding and therefore are known to perform well. They also

contribute to the traits being studied: Polyphenol oxidase (PPO) activity and

Xanthophyll content. Pedigree’s of the two Australian varieties; Sunco and Tasman

(Figure 1-11), contain lines very important in Australia’s wheat breeding history.

Sunco includes the wheat variety Cook in its parentage, which was a wheat grown in

Queensland in the late 1800’s displaying a good milling quality and rust resistance

(Spennemann 2001). Sunco has a quality classification of Australian Prime Hard,

released in 1986 with resistance against stem rust, and performs with a moderate

susceptibility to lead and stripe rust (NSW-DPI 2008). Sunco also has shown good

flour and noodle colour quality (Mares and Campbell 2001) and therefore was an

important inclusion to this study.

Tasman has a more complex lineage and includes lines from Mexico (CIMMYT –

International Maize and Wheat Improvement Centre), Italy and South Africa.

Tasman is classified as an Australian Hard wheat with a semi-dwarf stature, very

strong straw and resistance against stem, leaf and stripe rust in Australia. Tasman

contributes high levels of Xanthophyll content; however, it also contains an

45
unfavourably high level of Polyphenol oxidase. Tasman was released as a variety in

1993 (Brennan et al. 1993).

Figure 1-11: Pedigree of parents; Sunco and Tasman selected for mapping
population: Sunco x Tasman doubled haploid population (Source: Kammholz et al.
2001)

SUNCO
3Ag14 X Sun9E-27

4 backcrosses

BC4 X WW15

3 backcrosses

BC3 X Cook

Sunco
TASMAN
Chino X Barleta Klein Universal II X Barleta 7D

Sinvalocho X H44

La Estanzuela
Bage

Bage

Gabo 55 X Penjamo 62

Ciano 67 X Bluebird
Condor X 3Ag3

Torres

Tasman

46
The Chinese varieties used in this study were developed in China’s Zone V. Both

varieties, Mianyang 11 and Chuanmai 22, have the variety Fan 6 as part of their

pedigree. Fan 6 was the first landmark variety for Zone V in the 1970s and

contributed greatly to wheat production and wheat breeding (He et al. 2001). A

derivative of Fan 6, Mianyang 11 was developed by the Miangyang Prefectural

Agricultural Research Institute (PARI) in 1979 and possesses the broad adaptation,

high yielding potential, short stature and stripe rust resistance from Fan 6. In

addition, Mianyang 11 has powdery mildew resistance, escape from head scab and

early maturity. He et al. (2001) states “that it was the leading variety in the 1980s and

contributed 1.5 million ha in 1984, out yielding Fan 6 by 11.3%.”

Chuanmai 22 is also a derivative of Fan 6 and was developed from Mianyang

11/Chuanmai 20 in 1989 by the Sichuan Academy of Agricultural Sciences (SAAS).

This improved variety maintains the resistance to stripe rust, powdery mildew and

head scab, but has also showed tolerance to drought and poor soil fertility which are

major concerns for the hilly cropping areas of the Sichuan basin of Zone V (He et al.

2001). The detailed pedigrees of Fan 6, Mianyang 11 and Chuanmai 22 have been

displayed in Figure 1-12 and 1-13.

47
Figure 1-12: Pedigree information of Varieties Fan 6, (Data source: He et al. 2001)

i) Pedigree of Fan 6
Chengdu Guangtou Zhongnong 483

reselection reselection

Chengdu Guangtou Branched Wheat X Zhongnong 483 Branched Wheat

IBO 1828 X NP 824 Wuyimai X F1 Zhongnong 28 X IBO 1828 NP 824 X Funo

Branched Wheat
F1 F1 F1 F1

F2 X F1

Fan 6 Fan 7 69-1776

48
Figure 1-13: Pedigree information of Varieties Mianyang 11 and Chuanmai
22, (Data source: He et al. 2001)

ii) Pedigree of Mianyang 11

IBO 1828 X NP 824

Airongsui X Yaanzao

70-5858 X Fan 6

Mianyang 11

iii) Pedigree of Chuanmai 22

Hechuan
Hongpaideng X Ardito

X Shannong 205 406 X Yananza

Yananzao

Abbondanza X Zhuyeqing F3 X 69-1776

603-15443 X 987-1-2

Mianyang 11 X Chuanmai 20

Chuanmai 22

49
Specifically, the objectives of the study are:

1. To use the QTLs previously identified by Mares and Campbell (2001) to

improve the flour colour characteristics in backcross populations

generated between Australian varieties (Sunco and Tasman) and

Chinese varieties (Chuanmai 22 and Mianyang 11).

2. To evaluate Marker-assisted selection as an efficient method of

transferring QTLs located in Australian germplasm to a Chinese genetic

background.

3. To test the usefulness of microsatellite markers as a tool to predict

phenotype (Polyphenol oxidase activity and Xanthophyll content) in a

novel genetic background.

4. To increase awareness and usefulness of molecular markers in wheat

breeding in China.

50
Chapter 2: Establishment and Genotyping of Backcross

Populations

2.1 Introduction

Backcrossing is a form of recurrent hybridisation, which integrates into an elite

recurrent parent background specific alleles from donor lines while keeping the

overall integrity of the recurrent parent intact (Poehlman and Sleper 1995).

Using the Backcrossing method with in-breeding crops such as wheat,

generates lines homozygous for particular traits in a relatively short period of

time. If selection for the desired allele from the donor parent is conducted at

each generation, a breeding program can introduce important genes from

different sources and integrate them into highly adapted elite wheat varieties.

Traditionally this selection of important alleles has been conducted

phenotypically. In the past decade molecular markers have been successfully

used to track the genetic sections from the donor parent, as well as select lines

with a genome close to the recurrent parent (Langridge and Chalmers 2005).

Combining traditional backcross methods with molecular markers has

successfully introduced specific alleles for disease resistance, quality attributes

51
and many other traits (Frisch and Melchinger 2001, Frisch 2005, Fulton et al.

1997, Huang et al. 2004, Servin and Hospital 2002, Tanksley et al. 1996,

Willcox et al. 2002, Yousef and Juvik 2002).

Including marker assisted selection into the breeding process will only quicken

the process and decrease the number of lines requiring phenotyping (Bonnett et

al. 2005). Marker Assisted Selection (MAS) can be particularly useful when the

markers employed are co-dominant and reveal heterozygous individuals.

PCR and DNA sequencing have been integrated into traditional breeding

programs to assist the identification of important traits in the wheat genome

(Jones et al. 1997). A number of different molecular markers have been used in

plant research, of which have been described in Chapter 1. A very important

aspect of molecular marker use is their reproducibility in different genetic

backgrounds and by different laboratories and researchers. The study by Jones

et al. (1997) considered this aspect of the marker technology, comparing

RAPD, AFLP and SSR markers. He concluded that RAPD markers were

initially easy to perform, but the reproducibility of these markers was poor.

Alternatively, the AFLP and SSR markers produced reliable and robust

genotyping between different laboratories in Europe. Technical aspects of

52
markers also have to be considered. This includes the optimisation of

chemicals, enzymes, DNA template and thermocycling machines used in the

process.

In this current study SSR markers were employed due to their robust, co-

dominant nature, wide availability and ease of use with backcrossing strategies

(Korzun 2003).

This chapter will describe the process undertaken to generate the backcross

populations used in this project. A genotypic evaluation of the four populations

has been carried out to provide all the necessary information to complete the

aims of this project.

2.2 Materials and Methods

Establishment of Backcross Populations

The parental material used in the production of the backcross populations were

two Australian varieties: Sunco and Tasman, and two Chinese varieties:

53
Chuanmai 22 and Mianyang 11. The initial F1 crosses and the first backcross

were made by Dr. Eric Storlie in 2001, then as a postdoctoral fellow at the

University of Southern Queensland, before the commencement of this project.

These four populations were Sunco x Chuanmai 22, Sunco x Mianyang 11,

Tasman x Chuanmai 22 and Tasman x Mianyang 11.

Both the first and second backcross, of the four populations, were carried out

using the Chinese variety as the recurrent parent. After the backcrossing,

individuals within each of the four populations went through three further

generations of self-fertilisation (Single Seed Descent) before lines from each

population were planted in a field trial. After each generation each line

produced a different yield of grain, therefore the numbers of seed available for

further plantings varied. The amount of seeds (of individual lines) planted in

the three single seed descent generations, ranged from a minimum of 6 seeds to

a maximum of 34 seeds. Between the first and second backcrosses, molecular

markers were used to select the lines that would undergo the second backcross.

During the second backcross the important lines, those carrying the donor

marker alleles, were given priority. The backcrossing continued until there was

no more viable pollen available. These lines were genotyped again at the

54
second and third generation of self-fertilisation. This process has been

displayed in Figure 2-1.

All stages of population development took place in the Glasshouse facilities at

the University of Southern Queensland (USQ).

QTL analysis and Molecular Marker Selection

The initial QTL analysis was carried out using the Sunco x Tasman linkage

map (genotypic data) and the available phenotypic data (Mares and Campbell

2001). This linkage map was developed through GRDC funding and as an

integral part of the Australian Winter Cereal Molecular Marker Program

(AWCMMP). The phenotypic data used for this QTL analysis included

Polyphenol oxidase (PPO) and Xanthophyll evaluation of Sunco x Tasman

population material planted at Narrabri, NSW in 1998. This phenotypic data

was collected and analysed by Daryl Mares (then at the University of Sydney).

A second round of QTL analysis was carried out after a refinement of the

Sunco x Tasman linkage map. This was done through interval mapping

analysis using the Q-gene software, version 3.04 (Nelson 1997). The revised

55
Sunco x Tasman linkage map can be viewed at the following website:

http://cbbc.murdoch.edu.au/cmap.

The recent genotyping and map refinement on the Sunco x Tasman population

was conducted by Dr. Anke Lehmensiek of USQ whereby a greater range of

SSR markers were mapped to the areas of interest. These areas were located on

chromosomes 2A, 2D, 3B and 7A. During this revision, all the markers were

re-ordered and apparent double recombinants were removed. The refinement

process is described fully in Lehmensiek et al. (2005).

Plant Material and DNA Extraction

DNA was extracted from plant material collected from four week old seedlings.

For the first three screenings, the plant material was collected from seedlings in

the USQ glasshouse.

The first genotyping of the four backcrossed populations was carried out in

2002 on 246 BC1F2 seedlings. At this stage the Phenol/Chloroform DNA

extraction method was used. The genotyping carried out in 2003 was

56
conducted on BC2F3 seedlings using a kit supplied by Qiagen Pty Ltd. This

method was used to extract genomic DNA from a total of 216 lines (over the

four backcrossed populations). The decrease in total of lines (from 246 – 216)

was due to the selection of lines (using molecular markers) between the first

and second backcross. The Wizard Genomic DNA Extraction Kit was the

method used in 2004 and 2005. This method allows tubes or microplates to be

used. Tubes were used in 2004 on BC2F4 seedlings and the plate system was

used in 2005 on BC2F5 seedlings. DNA was extracted from all four

backcrossed populations, with 216 lines extracted in 2004 and 384 lines in

2005.

Full descriptions on all DNA extraction techniques are present in the Protocol

Section in Appendix A.

Field trial of BC2F5 material

Lines selected after the 2004 genotyping were planted in a “Latinised row-

column” designed field trial conducted at the Leslie Research Centre (QDPI &

F). The field trial statistical design was developed by statistician Allison Kelly

of QDPI & F. Each line was replicated three times, planted in a 0.5m row with

57
6 seeds per row. In the instances that there was not enough seed for a particular

line, two replicates were planted and the third was substituted with the parental

line, Sunco. All parental lines (Sunco, Tasman, Chuanmai 22 and Mianyang

11) and biochemical standard lines (Krichauff and Batavia) were planted within

the trial site, replicated a total of nine times.

During the field trial, plant material was sourced from each genotype at various

stages of development. Leaf material was taken from the first and the third

(four week old) seedling of each row. Genotyping was carried out on the first

seedling material only.

Once the plants had reached maturity the first plant from each row was

harvested separately. The remaining five plants were bulked together. The

grain was removed from the bulked samples at the Leslie Research Centre

(LRC) using the large threshing machines. These machines remove and

separate the grain from the wheat heads. The grain from these bulked samples

was stored in humidity and temperature controlled units at the LRC. The single

(first) plants were bagged and transported back to the USQ. The seed was then

removed from the wheat heads by hand and stored for use in the biochemical

tests (see Chapter 3).

58
Genotyping of lines produced in the four backcrossed populations

Genotyping was carried out at four different times throughout the project. It

was very important to track the genotype at each selected marker locus from

under the four different QTLs (PPO: 2A and 2D; Xanthophyll: 3B and 7A)

highlighted in the rationale section of Chapter 1.

The SSR primer sequences were obtained from the Wheat Microsatellite

Consortium (WMC), Agrogene (Röder et al. 1998), Pestova et al. (2000) and

Graingenes (http://wheat.pw.usda.gov). The amplification of the SSRs used in

this project was carried out in a PTC-100 thermocycler (Biometra). The

samples for amplification contained approximately 20ng of wheat genomic

DNA, 5µM each of the forward and reverse primer, 100µM of each dNTP,

1.5mM MgCl2 and 0.5U Taq DNA polymerase in a total volume of 10µl. The

following PCR profile was used on all samples: 1 cycle of 94oC for 3 mins,

followed by 40 cycles of 94 oC for 30secs, 50-60 oC (depending on the primer

combination) for 30 secs, 72 oC for 1 min, and 1 cycle of 72 oC for 10mins. The

amplified products were electrophoresed on 6% polyacrylamide gel using the

Bio-rad Sequencing Gel Electrophoresis System, or on a 5.6% polyacrylamide

59
gel using a Corbett Robotics Gel Scan 2000. The single AFLP used in this

project was carried out using the methods described in Chalmers et al. (2001).

The molecular markers (SSRs and AFLP) used throughout the genotyping

process changed between the years 2002 and 2003. This occurred because

more closely linked SSR markers were mapped to the revised Sunco x Tasman

population (Lehmensiek et al. 2005). These differences are displayed in Table

2-1. The genotyping carried out in 2002 enabled lines containing the Sunco or

Tasman allele to be selected for the second round of backcrossing. Likewise in

2003, all of the lines for each population were genotyped again to track the

lines heterozygous or homozygous for the Australian parent alleles. The

genotyping in 2004 was an important step in selecting lines for the field trial.

A number of lines were selected, containing different combinations of

Australian and Chinese alleles. The complete list of these selected lines can be

found in Table 2-3.

60
Table 2-1: Molecular Markers used in the genotyping of plant material in all four backcross populations –
comparing years 2002 with 2003-2005.

Population PPO - 2A PPO – 2D Xanthophyll – 3B Xanthophyll – 7A

2002 2003- 2002 2003- 2002 2003- 2002 2003-
2005 2005 2005 2005
Sunco x - gwm372 wmc18 wmc18 gwm285 gwm285 wmc17 wmc17
Chuanmai gwm526 gwm157 P36/M35 gwm566 wmc346 wmc346
22 cfd233
gwm301
Sunco x - gwm372 wmc18 wmc18 gwm285 gwm285 wmc17 wmc17
Mianyang gwm526 gwm157 gwm566 wmc346 wmc346
11 cfd233
gwm301
Tasman x wmc170 gwm372 wmc18 wmc18 gwm285 gwm285 wmc17 wmc17
Chuanmai wmc170 gwm102 gwm566 wmc346 wmc346
22 gwm526 wmc181 gwm108
gwm349
gwm301
Tasman x wmc170 gwm372 wmc18 wmc18 gwm285 gwm285 wmc17 wmc17
Mianyang wmc170 gwm102 gwm566 wmc346 wmc346
11 gwm526 wmc181 gwm108
gwm349
gwm301

61
2.3 Results

The process by which this study was carried has been depicted

diagrammatically in Figure 2-1. It allows the process to be visualised as a

whole, displaying the total volume of work conducted in the laboratory,

glasshouse and the field.

The lines carried through the backcrossing stage were selected by markers

identified through QTL analysis. The genotypic and phenotypic data generated

by previous research (Mares and Campbell 2001) allowed the re-identification

of the QTLs for Polyphenol oxidase activity (PPO) and Xanthophyll Content on

chromosomes 2A, 2D (PPO) and 3B and 7A (Xanthophyll).

62
Figure 2-1: Flow chart of laboratory and field work for the phenotyping and genotyping of the lines in the four
backcrossed populations

Australian Variety x Chinese Variety Time Scale

F1’s for 4 Populations 2001

Sunco x Chuan Mai 22
1st Backcross to Tasman x Chuan Mai 22 Marker Analysis
the Chinese parent Sunco x Mianyang 11 On 246 lines with 5 SSRs 2002
Tasman x Mianyang 11
Marker selection to decrease line numbers
nd
2 Backcross to the Marker Analysis 2003
BC1F2
Chinese parent On 216 lines with 13 SSRs

BC2F2
3 cycles Single Seed Descent
All populations
1. BC2F3 Marker Analysis
2004
On 216 lines with 13 SSRs
2. BC2F4

Field trial lines selected on the basis of marker

Field Trial 3. BC2F5 analysis
131 selected lines only
Marker Analysis 2005
BC2F5 seedlings On 384 lines with 13 SSRs
Biochemical Tests
PPO Activity Test: 11,328 assays
Xanthophyll Content Test: 696 BC2F6 seed 2006

63
The identification of these QTLs was performed twice, firstly with the original

Sunco x Tasman linkage map and then again with the revised and curated

version. The QTL traces using the original map have been displayed in section

(a) of Figures 2-2, 2-3, 2-4 and 2-5. The section (b) of the Figures 2-2, 2-3, 2-4

and 2-5 contain the QTL traces generated with the revised Sunco x Tasman

linkage map.

64
Figure 2-2: QTL analysis carried out on chromosome 2A of Sunco x Tasman DH population. All analysis used phenotypic data collected in
the 1998 field trial; a. used the original genetic linkage map, b. used the revised linkage map.

a. b.

a. Analysis carried out in 2002 using the original linkage map b. Analysis carried out in 2003 using the revised linkage map

Legend: the circled markers on the QTL trace indicate the Microsatellite markers selected and used in the genotyping of the four backcross populations

65
Figure 2-3: QTL analysis carried out on chromosome 2D of Sunco x Tasman DH population. All analysis used phenotypic data collected in
the 1998 field trial; a. used the original genetic linkage map, b. used the revised linkage map.

a. b.

a. Analysis carried out in 2002 using the original linkage map b. Analysis carried out in 2003 using the revised linkage map
Legend: the circled markers on the QTL trace indicate the Microsatellite markers selected and used in the genotyping of the four backcross populations

66
Figure 2-4: QTL analysis carried out on chromosome 3B of Sunco x Tasman DH population. All analysis used phenotypic data collected in
the 1998 field trial; a. used the original genetic linkage map, b. used the revised linkage map.

a. b.

a. Analysis carried out in 2002 using the original linkage map b. Analysis carried out in 2003 using the revised linkage map
Legend: the circled markers on the QTL trace indicate the Microsatellite markers selected and used in the genotyping of the four backcross populations

67
 igure 2-5: QTL analysis carried out on chromosome 7A of Sunco x Tasman
Fig
DH population. All analysis used phenotypic data collected in the 1998 field
trial; a. used the original genetic linkage map, b. used the revised linkage map.

a. b.

a. Analysis carried out in 2002 using the original linkage map

b. Analysis carried out in 2003 using the revised linkage map

Legend: the circled markers on the QTL trace indicate the Microsatellite markers selected and
used in the genotyping of the four backcross populations

68
The refinement of the Sunco x Tasman linkage map allowed a greater number

of microsatellite markers to be used in the genotyping of the backcross

populations. Table 2-2 displays work carried out on this linkage map and the

benefit gained through the map curation.

Table 2-2: Summary of the genotyping information of the revised and original
versions of the Sunco x Tasman linkage map.

Revised Map Original Map

Markers Map Size Markers Map Size Double Recominants
(no.) (cM) (no.) (cM) Removed
(no.) (%)

Sunco x Tasman 345 2150 300 2920 635 1.03

(Source: Lehmensiek et al. 2005)

The volume of genotyping carried out changed substantially between the years

2002 and 2003. A total of five SSR markers were amplified on the 2002 lines.

This was increased to ten SSR markers for the Sunco populations and thirteen

SSR for the Tasman populations (2003 – 2005 lines). The number of SSRs

used in the genotyping increased for all QTLs except the 7A Xanthophyll QTL.

Extensive research was carried out to increase the number of SSR markers in

this location, but the SSRs wmc17 and wmc346 were still the most closely

linked microsatellites to this trait on 7A (Lehmensiek et al. 2005).

69
There were a wide range of lines included into this field trial: Ones that

contained only the recurrent parent alleles; those that contained the donor

alleles for one or more QTLs; those with alleles still remaining in the

heterozygote state and those with a mixture of alleles. The complete range of

lines, and total numbers, included into the field trial has been displayed in Table

2-3.

70
Table 2-3: Summary of the lines included in the 2005 field trial, A wide range of genotypes from each population were included to evaluate
the use MAS.

Chinese Hetero- 2A 2D 3B 7A 2A/ 2A/ 2A/ 2D/ 2D/ 3B/ 2D/3B 2A/7A 2A/2D 2A/2D Mixture
alleles zygotes 2D 3B 7A 3B 7A 7A /7A /3B /7A /3B of the
only only alleles
Sunco x 0 6 2 5 6 1 1 5 0 2 1 6 1 2 1 2 0
Chuan-
mai 22
Sunco x 9 5 3 2 4 6 0 2 1 1 0 4 1 1 0 0 2
Mian-
yang 11
Tasman x 0 1 2 1 1 1 1 1 0 2 0 3 3 1 0 0 8
Chuan-
mai 22
Tasman x 3 6 1 1 2 1 0 1 0 2 2 3 0 0 1 0 1
Mian-
yang 11
Total 12 18 8 9 13 9 2 9 1 7 3 16 5 4 2 2 11

Legend
* 2A – PPO QTL; 2D – PPO QTL; 3B – Xanthophyll QTL; 7A – Xanthophyll QTL.
Each column contains the number of lines with different genotypic combinations, focusing only on the markers used from each of the four QTLs for PPO
activity and Xanthophyll content. The first column containing lines with only the Chinese parent alleles; second and last column with a combination of
Chinese and Australian parental alleles; third to sixteenth column containing the Australian alleles for those particular QTLs.

71
2.4 Discussion

One of the aims of this project was to use previously identified QTLs from the

Sunco x Tasman doubled haploid population and to subject them to Marker

Assisted Selection. The field trial conducted was an important step in this

research. It allowed a true indication of phenotype to be gained when the lines

were subjected to environmental factors away from the semi-regulated

environment of a glasshouse facility. The number and types of lines planted

into this trial were selected purely on genotype. The purpose of leaving the

phenotyping until the end of the process was to obtain some understanding on

the use of molecular markers and their ability to predict a particular phenotype.

A large range of genotypes were included into the field trial to generate an

accurate view of the markers predictability and their use in wheat breeding

situations. Some of these lines planted in the field trial may still be genetically

segregating, as after two backcrosses and three generations of self-fertilisation

with a degree of molecular marker selection, the lines will have 98.44% of

alleles fixed. This will have to be taken into account when subjected to the

biochemical testing.

72
The genotypic and phenotypic data collected for the Sunco x Tasman doubled

haploid population was comprehensive enough to identify QTLs for both traits:

PPO activity and Xanthophyll content. This information is the starting point for

this project.

When attempting to use molecular markers in wheat breeding programs, they

need to be user-friendly, economic and time efficient (Francia et al. 2005).

AFLPs were used in the first phase of the AWCMMP as they produced a large

number of polymorphic markers between the parents: Sunco and Tasman,

allowing the linkage map to develop quickly. The markers produced through

AFLPs are dominant markers and they do not have the ability to identify

heterozygotes from homozygotes (Landjeva et al. 2007). Therefore the AFLPs

in the original Sunco x Tasman linkage map were not useful to track the

markers across the four backcross populations in this project and SSR markers

were used instead.

It was very important to find reliable, polymorphic and co-dominant molecular

markers on the areas of interest in the four chromosomes 2A, 2D, 3B and 7A of

the Sunco x Tasman population. For Marker-assisted selection to be successful,

the two flanking markers need to be tightly linked to the trait/gene of interest

73
(Babu et al. 2004, Francia et al. 2005). Therefore the successful refinement of

the Sunco x Tasman linkage map allowed more tightly linked SSR markers to

be used as flanking markers for the selection steps in this project. If the

markers used in the 2002 genotyping session were used throughout the entire

process, many positive lines would have been lost and negative lines could have

been carried through without any information on possible recombination

(Holland 2004).

Comparing the QTL traces generated for PPO activity and Xanthophyll content

highlights the benefits of map curation. The QTL peaks from the revised

linkage map are more sharply located on all four different chromosomes. This

results was also found by Lehmensiek et al. (2005) on other linkage maps

(CD87 x Katepwa; Cranbrook x Halberd).

There was a fall in LOD score for the PPO QTLs, whereby 2A went from LOD

of 4.05 – 3.61 (Figure 2-2) and similarly for the 2D QTL, the LOD score went

from 17.75 – 16.73 (Figure 2-3). Linkage map refinement is generally

associated with an increased LOD score, as found in the 3B and 7A

Xanthophyll QTL (Lehmensiek et al. 2005)

74
The slight drop in LOD score for the 2A and 2D PPO QTLs may indicate that

the original QTL significance was over estimated. The addition of the reliable

SSR markers underneath these QTL peaks has provided a more accurate

measure of the QTL’s significance. Therefore with these significant LOD

scores and the re-identification of the QTL peaks, choosing markers from

underneath these peaks would be the most suitable for MAS.

75
Chapter 3: Phenotyping through Biochemical Testing

3.1 Introduction

Breeding for quality attributes can be a difficult and time consuming task, but it

is core to all wheat breeding programs worldwide. The measurement of most

quality traits requires significant testing and therefore restricts the progress of

new and improved genotypes for wheat breeders (Gale 2005).

The Asian markets are very important to wheat production in Australia, as they

are large importer of Australian grown wheat and one of the largest consumers

of wheat worldwide. A high proportion of Asian noodles are produced using

bread-wheat flour, and therefore particular quality characteristics are required

for high quality noodle production. These attributes include appropriate flavour

and texture (Fu 2008-in press), protein content, dough strength and starch

pasting properties (Crosbie and Lambe 1993) and flour colour (Asenstorfer et

al. 2006, Fu 2008-in press, Mares and Campbell 2001, Zhang et al. 2008).

The colour of the flour can be affected by a number of components; these

being, the colour of the starchy endosperm, the presence of bran flakes from

76
milling and differing genetic backgrounds. Xanthophyll content is a large

contributor to the yellowness of wheat flour, which in turn determines the

colour of the noodles. This level of xanthophyll is variety dependent in wheat,

and is present in Australian hard wheat varieties (Asenstorfer et al. 2006, Fu

2008-in press). A bright yellow colour is required of Yellow alkaline noodles,

and if this can be delivered to the noodle through the flour colour, then only

minimal amounts of alkaline salts are required for the production of these

noodles (Zhang et al. 2008). In contrast, He et al. (2004) states a low yellow

pigment level is required for Chinese white noodles, where a bright white

noodle colour is required for Chinese noodles.

Colour stability is also a very important part of noodle production, and this

colour is required to remain stable for 24-48 hours after preparation. A high

level of Polyphenol oxidase (PPO) activity in wheat flour has been linked to

darkening of fresh noodles (Asenstorfer et al. 2006, Mares et al. 1997). Levels

of PPO activity present in the wheat grain can be calculated through a whole

seed assay (Bernier and Howes 1994). This allows breeders to phenotype

important breeding lines and to assist in identification of the genomic areas

responsible for this trait. These characteristics of noodle quality were discussed

fully in Chapter 1.

77
An aim of this project is to generate wheat lines with enhanced quality

attributes. As a general rule the Chinese varieties contribute a low level of PPO

activity and Xanthophyll content, therefore including the Sunco alleles for these

traits should only benefit the quality of these wheat lines. QTLs for high PPO

activity (contributed by Tasman) and high levels of Xanthophyll content

contributed by Tasman on 7A and to a lesser degree by Sunco on 3B, have been

located in the Sunco x Tasman population (Mares and Campbell 2001);

therefore transferring the Sunco alleles within these genomic areas into the

Chinese background may enhance the whiteness of flour colour and the colour

stability of the noodles produced. To research this theory high quality

phenotypic data needs to be generated using the methods used in current

breeding programs. Testing for PPO activity and Xanthophyll content are time

consuming assays; therefore identifying molecular markers associated with

these traits would benefit further molecular research and breeding endeavours.

3.2 Material and Methods

The grain produced from the field trial (Chapter 2) was retained for biochemical

testing. Both tests were carried out in the laboratory of the University of

78
Southern Queensland using laboratory statistical designs developed in

consultation with statistician Allison Kelly. The design included all test

samples from the field trial, parental material (Sunco, Tasman, Chuanmai 22

and Mianyang 11) and the biochemical standards (Batavia and Krichauff). The

laboratory designs were generated specifically for each biochemical assay,

taking into account the number of samples tested at the one time, this being five

samples for the PPO assay and eleven samples for the Xanthophyll assay. The

randomized design also incorporated duplication of test samples to remove any

laboratory or field effects from the analysis.

Polyphenol Oxidase Activity of Single Seeds

The method used to complete this testing was a modified version of the original

single seed assay by Bernier (1994), which is based on the reaction of PPO with

the substrate tyrosine. The modifications made to the original method were

carried out by Dr Daryl Mares (Mares and Panozzo 1999) to optimize the assay

for Australian germplasm and conditions. This modified method is currently

used widely throughout Australian wheat breeding programs.

79
This assay is carried out in a microplate separating the samples into three

distinct areas. The first row always served as the reagent blank. Rows 2 – 11

contained the field trial genotypes, and the last row always included Sunco

material, the internal standard. Single seeds were placed into each well of a

microplate except for the first row. Five field trial genotypes were tests on each

plate, and seeds from each of these genotypes were placed in two rows each.

This assay was conducted over a number of days. The first plate analysed each

day consisted of all parental and biochemical standard varieties. These varieties

included Tasman, Chuanmai 22, Mianyang 11, Batavia, Krichauff and Sunco.

The microplate design for the PPO assay is displayed in Figure 3-1.

Figure 3-1: Design of the microplate used for each PPO assay carried out on
the field trial material

1 2 3 4 5 6 7 8 9 10 11 12
A
B
C
Sample 1

Sample 2

Sample 3

Sample 4

Sample 5

SUNCO

D
Blank

E
F
G
H

80
Tyrosine solution (200µl) was added to each well of the plate using a multi-

channel pipette without the addition of bubbles to the well. The tyrosine stock

solution recipe and reagents necessary for this assay are displayed fully in

Appendix A.

Each plate was covered and incubated for 3 hours at 37oC. After incubation,

100µl of the substrate was removed from each well with a multichannel pipette

and placed into a clean microplate to be read in a microplate reader at the

optical density of 415nm. Each plate was assessed for outliers once all the

optical densities for each well were calculated.

For each genotype analysed per microplate, a mean value for each of the 16

replicate results was calculated. The mean of the blank row was calculated and

subtracted from each of the genotype means to produce a ‘corrected’ mean

value for each genotype. For each sample, Standard Deviations and Standard

Errors were calculated and outliers were removed if it produced a standard

deviation of >0.05.

Frequency distributions of each individual population were carried out using

the populations PPO activity results. The degree of skewness was calculated

81
for each of the four backcross populations through the computer program SPSS,

version 15. This computer program calculates skewness using the below

equation:

n(n- m3
X
n-2 m2 3/2

Where

mr = ( Xi – X

Because of the lack of normality and the degree of skewness present in the

frequency distributions, a general t-test could not be used; therefore the

conservative Welch’s t-test was used in further analysis. This test enabled

comparisons to be made between the four parental varieties and to test for

significant differences between transgressive segregants present in each

population.

82
Xanthophyll Content from Flour Samples

Samples of grain from each genotype and parental variety grown in the field

trial (Chapter 2) were used in this assay. Approximately 20 grams of field trial

grain was milled to produce the flour samples required for the Xanthophyll

content biochemical test. All milling was carried out by the staff of the Quality

Laboratory at the Department of Primary Industries and Fisheries – Leslie

Research Centre in a Quadrumat Junior Mill.

1 gram of flour was weighed into 50ml conical ended test tubes and then 5mls

of methanol was added to each tube. All tubes were placed on their sides

attached to the orbital shaker and gently mixed for 15 minutes. After being

thoroughly mixed the tubes were centrifuged for 15 minutes at 4000rpm. The

supernatant was then filtered through a 0.45µ syringe filter to ensure all flour

particles were removed. The supernatant was then transferred into clean 15ml

tubes and placed into a warm water bath to inhibit the samples becoming turbid

before measuring the optical density of the extract at 436nm compared to a

methanol blank.

83
Results from samples that were duplicated within the randomized statistical

design were averaged to produce a single optical density result. Standard

Deviations and Standard Errors were calculated for each sample or group of

samples. If a particular sample mean possess a standard deviation of above

0.05 it was removed from the analysis.

Frequency distributions were carried out for Xanthophyll contents of each

backcross population. The degree of skewness was calculated for each of the

distributions using the SPSS program (v15) and the equation displayed above.

Again the lack of normality in these distributions indicated the conservative

Welch’s t-test was required to perform the comparisons between parental

Xanthophyll content data.

3.3 Results

Polyphenol Oxidase Activity

The PPO assays of lines from the four Australian x Chinese crosses resulted in

the measurement of a total of 109 96-well microplates, together with nine

84
microplates of standards and parental varieties. This equates to 118 plates and

a total of 11,328 assays analysed over nine days. An example of the PPO assay

after the incubation period is displayed in Figure 3-2.

The number of tests carried out per day varied; however the consistency of the

results between days was assessed by the number of outliers observed. These

ranged from 5.63% to 8.48%. A number of outliers were removed from the raw

data based on their standard deviations of the mean. 6.48% of the total assays

conducted (734 assays of 11,328) were removed. The number of outliers

observed and removed on the separate days of testing is listed in Table 3-1.

Figure 3-2: An example of a microplate of a PPO assay after the three hour
incubation period. Refer to Figure 3.1 for sample placement pattern

85
Table 3-1: A summary of the outliers evaluated from each PPO assay
conducted on the 2005 field trial material; genotypes of all four backcrossed
populations, parental varieties and standards.

Day of PPO Number of Total Number Percentage of

testing Outliers of Assays Outliers / Total
Assays
1 57 672 8.48%
2 92 1344 6.85%
3 81 1440 5.63%
4 93 1536 6.05%
5 83 1440 5.76%
6 94 1440 6.53%
7 85 1440 5.90%
8 33 576 5.73%
9 116 1440 8.06%
TOTALS 734 11,328 6.48%

Genotypes that produced results with a high standard deviation (>0.05) were

looked at further. If the standard deviation wasn’t lowered by the removal of

the outliers these genotypes were repeated. Table 3-2 highlights the genotypes

that produced initial high standard deviations, and compares their repeat results.

These genotypes were found in all four backcross populations, with the most

genotypes being from the Sunco x Chuanmai 22 and Tasman x Chuanmai 22

populations.

The frequency distribution of the final results for PPO activity is shown in the

graphs of Figure 3-3 and Figure 3-4. These results display skewness to a higher

86
PPO activity, with the exception of the Sunco x Mianyang 11 population. The

distributions also highlight a number of lines with lower PPO activity results

than either parent. When the PPO activity data from all four populations were

investigated further, a number of significant transgressive segregants were

identified. Both of the Sunco populations contained 17 lines each producing

significantly lower PPO activity levels than either parent (14.66% of the Sunco

x Mianyang 11 population and 14.05% Sunco x Chuanmai 22 population). Of

those lines, 5 lines from the Sunco x Mianyang 11 population and 6 lines from

the Sunco x Chuanmai 22 population, produced PPO activity levels lower than

Sunco. The level of PPO activity whereby the transgressive segregants become

significantly different to either parent varies for each population. These levels

were compared to the PPO activity levels produced by each parent, and are

shown in Table 3-3. A PPO activity level (OD415nm) of 0.111 (Sunco x

Mianyang 11 population) and 0.109 (Sunco x Chuanmai 22 population) is

required to be significantly lower than the PPO level of Sunco.

The PPO activity data was analysed in a similar way for the Tasman

populations, resulting in 3 lines from the Tasman x Mianyang 11 population

and 10 lines from the Tasman x Chuanmai 22 population producing

significantly lower PPO activity levels than either Chinese parent. Again,

87
different levels of PPO activity were found to be significantly lower than the

Chinese parent in the Tasman populations. This being a PPO activity level

(OD415nm) of 0.122 (Tasman x Mianyang 11 population) and 0.130 (Tasman x

Chuanmai 22 population). The Tasman genome present in these populations

inherently increases the PPO activity levels with; 26.47% (18 lines) of the

Tasman x Mianyang 11 population, and 33.78% (25 lines) of the Tasman x

Chuanmai 22 population producing PPO activity levels significantly higher

than Tasman.

The descriptive information (means, standard deviations, sample size, standard

error and margin of error) produced by the parental varieties, parental pairs and

the entire populations can be found displayed in Table 3-3. Comparisons were

made between each of the parental varieties using the conservative Welch’s

test, and it was found that there was a significant difference between the two

Australian varieties, Sunco and Tasman. There was also a significant

difference seen between Sunco and Tasman and both Chinese varieties, but

there were no differences between the two Chinese varieties themselves.

88
Table 3-2: Summary of the genotypes reanalysed because of a high initial
Standard Deviation
Standard Deviation Standard Deviation
Genotype 1st assay Repeat assays
SC28-6 0.053 0.047; 0.044
SC33-4 0.062 0.039; 0.036
SC27-10 0.113 0.046
SC28-4 0.056 0.039; 0.042
SC32-2 0.057 0.036
SC10-9 0.059 0.049
SC32-1 0.061 0.048
SC27-3 0.053 0.042
SC10-11 0.117 0.033; 0.045 ; 0.038
SC28-8 0.158 0.030; 0.035
SC28-4 0.056 0.039; 0.042
SC33-4 0.062 0.039
SM57-14 0.056 0.029; 0.034
SM27-9 0.075 0.033
SM15-10 0.058 0.031; 0.031
SM15-10 0.058 0.031
TC28-8 0.082 0.066
TC28-11 0.056 0.034
TC35-7 0.058 0.049
TC31-6 0.095 0.052
TC28-8 0.075 0.049
TC31-11 0.056 0.038
TC31-6 0.064 0.038; 0.048
TC28-4 0.052 0.024; 0.035
TC28-4 0.052 0.024; 0.035
TC35-7 0.094 0.063
TC31-11 0.088 0.045
TM4-6 0.053 0.040; 0.039
TM28-1 0.053 0.060
TM31-11 0.055 0.074
TM29-5 0.083 0.052
TM29-6 0.099 0.027
TM20-2 0.076 0.037
TM29-15 0.054 0.039
TM29-5 0.055 0.031; 0.042
TM29-5 0.055 0.042

- bold text indicates lines with a standard deviation >0.050

89
Figure 3-3: Frequency distribution graphs of PPO activity (OD 415nm) results for each of the Sunco backcross populations

a. Sunco x Chuanmai 22 b. Sunco x Mianyang 11

20
20

MY11

15
CM22 15

Sunco
Sunco

Frequency
Frequency

10
10

5
5

Skewness 0.584 Skewness 0.096

Mean =0.16558 Mean =0.1735

Std. Dev. =0.032856 Std. Dev. =0.042222
N =116 N =121
0 0
0.100 0.125 0.150 0.175 0.200 0.225 0.100 0.150 0.200 0.250 0.300 0.350

PPO PPO

90
Figure 3-4: Frequency distribution graphs of PPO activity (OD 415nm) results for each of the Tasman backcross populations

a. Tasman x Chuanmai 22 b. Tasman x Mianyang 11

Tasman
12.5
12.5 Tasman

10.0
10.0
MY11

Frequency
Frequency

7.5
7.5

5.0 5.0

CM22
Skewness 0.794 2.5
Skewness 0.531
2.5

Mean =0.19682 Mean =0.18497

Std. Dev. =0.06232 Std. Dev. =0.038115
N =74 N =68
0.0 0.0
0.000 0.200 0.400 0.100 0.150 0.200 0.250 0.300

PPO PPO

91
 able 3-3: Descriptive information for each backcross population for PPO
Tab
activity

Sample Margin
Mean Standard size Standard of Error
Individual PPO Deviation (n) Error (at 95%)
Parents activity
OD415nm
Sunco 0.121a 0.0196 118 0.0018 ±0.0035
Chuanmai 22 0.143c 0.0151 13 0.0042 ±0.0082
c
Mianyang 11 0.137 0.0114 13 0.0032 ±0.0062
Tasman 0.192b 0.0435 14 0.0116 ±0.0228
Parental
Pairs
Sunco &
Chuanmai 22 0.132 0.0160 2 0.0109 ±0.0214
Sunco &
Mianyang 11 0.128 0.0089 2 0.0063 ±0.0124
Tasman &
Chuanmai 22 0.167 0.0336 2 0.0237 ±0.0465
Tasman &
Mianyang 11 0.162 0.0401 2 0.0284 ±0.0556
Population
Data
Sunco x
Chuanmai 22 0.172 0.0401 121 0.0037 ±0.0072
Popn
Sunco &
Mianyang 11 0.166 0.0033 116 0.0031 ±0.0060
Popn
Tasman &
Chuanmai 22 0.195 0.05919 74 0.0068 ±0.01339
Popn
Tasman &
Mianyang 11 0.188 0.0392 68 0.0047 ±0.0092
Popn

92
Table 3-3
Legend: Individual Parents: statistical data obtained for each parent
separately.
Parental Pairs: statistical data obtained for each pair.
Population Data: statistical data obtained for each
Backcross population generated in this project.
a
: statistical difference between the mean of Sunco and all
other parents
b
: statistical difference between the mean of Tasman and all
other parents
c
: statistical difference between the mean of Chuanmai 22
and Mianyang 11 and Sunco and Tasman

93
Xanthophyll Content

The Xanthophyll testing of the field material consisted of 696 individual tests

over seventeen different days. The raw data was assessed, and if particular tests

for genotypes had a high standard deviation (<0.005), then those genotypes

were repeated. The repeats were included in the complete data set if they

lowered the standard deviation between the individual field replication, and

genotype. For the data analysis, each genotype was assessed as a group of the

three field replications and any duplication/repeats carried out in the testing

process, this being termed ‘genotype group’ in the discussion. The numbers of

individual line tests that were repeated have been summarized in Table 3-4.

For some of the individual lines and genotype groups, the standard deviations

could not be lowered with a repeat test.

94
Table 3-4: Summary of the results from the repeat xanthophyll content tests

Population No. of Result Class Number of Percentage

Repeats Repeats of total
(of total repeats
100)
Sunco x a. Repeats
Chuanmai 22 25 corrected Std 14 45.2%
Dev in both
individual line 12 38.7%
and genotype
b. Repeats 5 16.1%
corrected Std-
Dev in replicates
of a individual
line
c. Repeats didn’t
correct Std Dev
for individual
line or genotype
Sunco x
Mianyang 11 31 a. 17 68%
b. 7 28%
c. 1 4%
Tasman x
Chuanmai 22 21 a. 16 76.2%
b. 4 19.0%
c. 1 4.8%
Tasman x
Mianyang 11 23 a. 7 30.4%
b. 8 34.8%
c. 8 34.8%

Frequency distribution curves were produced from the Xanthophyll content

data for each backcross population (Figure 3-5 and Figure 3-6). The Sunco

95
populations show low levels of skewness, while the Tasman populations show

skewness towards high xanthophyll content values.

When the lines of the Sunco populations are considered individually, a very

large proportion of lines performing better than Sunco and the Chinese parent

are evident, therefore conferring a trend towards low levels of Xanthophyll

content. 95.65% (Sunco x Mianyang 11) and 90% (Sunco x Chuanmai 22) of

the lines had a lower Xanthophyll content level than Sunco. Similarly, 76.52%

(Sunco x Mianyang 11) and 41.67% (Sunco x Chuanmai 22) had a lower

Xanthophyll content level than the Chinese parent. The degree of significance

for these trends has not been assessed due to the lack of multiple line testing

carried out during this assay. To test Xanthophyll content single flour samples

are tested, therefore not producing the multiple data points required of further

data analysis. Also due to the nature of the statistical laboratory design for this

assay not all lines were duplicated during the assessment of Xanthophyll

content on each of the four backcross populations.

The descriptive information generated through the Xanthophyll content assay is

displayed in Table 3-5.

96
Figure 3-5: Frequency distribution graphs of Xanthophyll content (OD 436nm) results for each of the Sunco backcross
populations

a. Sunco x Chuanmai 22 b. Sunco x Mianyang 11

20
30

CM22 Sunco

25

15

20
Frequency

Frequency
10 MY11
15

Sunco
10

Skewness -0.035 5 Skewness 0.147

Mean =0.0432 Mean =0.04057
Std. Dev. =0.007125 Std. Dev. =0.007301
N =119 N =115
0 0
0.030 0.035 0.040 0.045 0.050 0.055 0.060 0.020 0.040 0.060

XanthoOD XanthoOD

97
Figure 3-6: Frequency distribution graphs of Xanthophyll content (OD 436nm) results for each of the Tasman backcross
populations

a. Tasman x Chuanmai 22 b. Tasman x Mianyang 11

CM22 20

12.5
MY11

10.0 Tasman 15
Frequency

Frequency
7.5
10 Tasman

5.0 Skewness 0.607 Skewness 0.557

2.5

Mean =0.04807 Mean =0.04526

Std. Dev. =0.011323 Std. Dev. =0.009715
N =71 N =65
0.0 0
0.040 0.060 0.080 0.020 0.040 0.060

XanthoOD XanthoOD

98
Comparisons were made between the parental varieties with respect to their

Xanthophyll contents, and the only significant difference seen was between the

two Chinese varieties compared to the Australian varieties. There were no

significant difference produced between Sunco and Tasman, or between

Chuanmai 22 and Mianyang 11.

99
Table 3-5: Descriptive information for each backcross population for
Xanthophyll content

Mean Standard Sample Standard Margin

Xantho- Deviation size Error of Error
phyll (n) (at 95%)
Individual Parents Content
OD436n
m
Sunco 0.052a 0.0053 56 0.0007 ±0.0014
Chuanmai 22 0.041c 0.0066 3 0.0037 ±0.0074
Mianyang 11 0.046c 0.0051 4 0.0026 ±0.0050
Tasman 0.064b 0.0031 3 0.0006 ±0.0035

Parental Pairs

Sunco & Chuanmai 0.047 0.0080 2 0.0057 ±0.0112

22

Sunco & Mianyang 0.049 0.0047 2 0.0033 ±0.0065

11

Tasman & 0.052 0.0160 2 0.1133 ±0.0222

Chuanmai 22

Tasman & 0.055 0.0126 2 0.0089 ±0.0176

Mianyang 11
Population Data

Sunco x 0.043 0.0072 119 0.0007 ±0.0013

Chuanmai 22 Popn

Sunco & 0.041 0.0072 115 0.0007 ±0.0013

Mianyang 11 Popn

Tasman &
Chuanmai 22 Popn 0.048 0.0113 71 0.0013 ±0.0026

Tasman &
Mianyang 11 Popn 0.045 0.0102 65 0.0013 ±0.0025

100
Table 3-5

Legend: Individual Parents: statistical data obtained for each parent

separately.
Parental Pairs: statistical data obtained for each pair.
Population Data: statistical data obtained for each backcross
population generated in this project.
a
: statistical difference between the mean of Sunco and all other
parents
b
: statistical difference between the mean of Tasman and all
other parents
c
: statistical difference between the mean of Chuanmai 22 and
Mianyang 11 and Sunco and Tasman

101
3.4 Discussion

A large amount of biochemical testing was carried out to produce the raw data

for PPO activity levels. With any laboratory test there will always be a certain

degree of outliers produced, so it is very important to take this into account

when the raw data is analysed (Bhar and Gupta 2001). The amount of outliers

removed from the raw data was a relatively small 6.48%. Outliers can occur

throughout this process and are usually due to experimental error, for example

bubbles in the substrate, or the seed being slightly damaged. Researchers

McCaig et al. (1999) found there was a large seed-to-seed variation when using

L-tyrosine as a substrate for the PPO assay. Seed coat damage allows for more

PPO to be extracted during the assay compared to undamaged and smooth seed

coats.

For a number of genotypes PPO activities were re-analysed because they were

showing a PPO activity level too high or too low compared to the other

duplicates of that particular genotype group. To check whether this was a

reproducible result, or due to experimental error, they were re-tested. Because

the seeds used in this assay are subjected to a chemical substrate those exact

seeds cannot be used again. Therefore for re-testing, new seed has to be

102
selected from seed bulked from that particular wheat line. So this, more

correctly, could be termed re-selection and re-testing. Only 6 of the 36 lines

that were re-tested didn’t show an improved standard deviation (≤ 0.05). All of

these lines were from the populations with Tasman as the Australian parent,

three lines from each. The lines that did not show an improved standard

deviation could still be undergoing genetic segregation, since a level of

heterozygosity is still likely to be present (discussed in Chapter 2).

The descriptive information generated on the PPO activity data allows us to see

the differences between the Australian and Chinese parents. These differences

have enabled a number of lines to be generated, producing a lower PPO activity

than any of the four parental varieties. This shows that an improvement in

colour stability has been achieved. The marker haplotypes of these lines need

to be tested to discover whether the desirable QTLs from the Australian parents

have been transferred (Chapter 4). He et al. (2004) has stated that more

research needs to be directed into increasing white noodle quality. The

improvement in colour stability seen within these four backcross populations is

an important step in this process. Habernnicht et al. (2002) has also found that

cultivars that had relatively low PPO activity levels produced a high level of

colour stability and noodle brightness (L*) in fresh noodles.

103
The frequency distributions of PPO activity show a substantial degree of

skewness to high PPO activity in all but the Sunco x Mianyang 11 population.

This skewness may be indicating additive/epistatic effects between genes

regulating PPO levels in each parent and suggests that these PPO regulating

loci are different in each parent. To confirm this further genetic information

needs to be generated on the Chinese varieties. However, this genetic analysis

is outside the scope of this current project.

It was an aim of this project to assess the possibility of improving quality

attributes of the Chinese varieties by the introgression of QTLs from Australian

varieties. The marker selection applied throughout the development of the four

populations has been successful in producing lines that have PPO activity levels

significantly different from the original Chinese parent. In the populations

where the Tasman allele was selected, the majority of the transgressive

segregants were directed towards an increased level of PPO activity. In the

Sunco populations there were a number of lines found to have a lower PPO

activity level than Sunco and either of the Chinese parents. Bonnett et al.

(2005) also has success in enriching a population with the selection of the target

alleles through the backcross breeding method. A further investigation into the

genotype of these transgressive segregants has been carried out in Chapter 4.

104
When looking at the Xanthophyll content results a number of lines had to be re-

tested because of the original result having a high standard deviation. The

majority of repeat tests produced a lowered standard deviation (≤0.05) in both

the comparison with the individual line (other duplicates) and when grouped

with its genotype group. Overall, the re-testing of lines was successful in

producing reliable data for Xanthophyll contents in all but the Tasman x

Mianyang 11 population.

The frequency distributions display an obvious skew towards high Xanthophyll

content in the Tasman x Chuanmai 22 and Tasman x Mianyang 11 populations.

This suggests interaction between several loci may be required to produce low

xanthophyll content whereas the degree of skewness in the Sunco populations is

minimal. To explore the genetic implications of these distributions requires

further genetic analysis of the Chinese varieties.

As low levels of PPO activity are important to Chinese white noodle

production, flour and noodle brightness is influenced by Xanthophylls and other

yellow pigment types (Mares and Campbell 2001, Zhang et al. 2008). The

introgression of QTLs for both quality traits (PPO activity and Xanthophyll

content) into the Chinese genetic background will then produce wheat lines

105
with high noodle quality attributes adapted to the growing environment in

China.

The assessment of Xanthophyll content and PPO activity are time consuming

assays, due to the multiple steps required for each test, and the number of tests a

single operator can complete each day. Furthermore, in the case of PPO assays

damaged seed and environmental influences can give artificially high readings

that can confound results.

Difficulties in phenotyping have been one of the driving forces in integrating

molecular markers into breeding programs, as a tool for line selection and

genotype characterization (Francia et al. 2005). To test the efficacy of

molecular markers for selecting these particular traits, the next Chapter will

investigate whether the markers for PPO activity and Xanthophyll content in

the two Australian lines remain linked to these characters in BC2F6 produced in

the BC2F5 field trials.

106
Chapter 4: Analysis of the Marker Assisted Selection in the
backcross populations

4.1 Introduction

Marker assisted selection (MAS) has been reported as the next step after

identifying and validating QTLs found through comprehensive linkage analysis

between genotyping and phenotyping data. MAS is a technique by which

molecular markers are used to select lines from large populations containing

genomic regions important to particular traits. Using MAS to assist the

conventional backcrossing method has allowed genes/QTLs to be transferred

into elite cultivars (Holland 2004). Francia et al. (2005) explains the main aims

of MAS as; a) being able to trace favourable alleles (dominant or recessive)

across generations; b) to identify the most suitable individual lines among

segregating populations; and c) to break linkages with unfavourable alleles. It

has been found that applying MAS at early stages of breeding allows large

number of lines with unfavourable genotypes to be removed, thus reducing the

number of lines to be tested (Bonnett et al. 2005, Liu et al. 2004). Using

molecular markers at early generations can also reduce the number of lines

which have a recombination between the marker loci and the gene/QTL.

Recombination events as well as undesirable alleles at neighbouring loci linked

107
to the transferred gene/QTL can use cause complications with these lines in

later generations (Bonnett et al. 2005, Eathington et al. 1997, Ribaut and Betrán

1999). Linkage drag is the term given to unfavourable alleles transferred with

the target donor genomic region during backcrossing. When marker-assisted

backcrossing is carried out, the use of tightly linked flanking markers can

considerably reduce the amount of linkage drag (Stam and Zeven 1981). MAS

carried out with tightly linked markers is also important in determining the

number of progeny required in each backcross generation. The number of

progeny required for selection for particular traits is determined by the distance

between the flanking markers and the number of QTL being considered at the

one time (Chao and Ukai 2000, Edwards and Page 1994). Some researchers

state the distance between the flanking markers used in MAS should be 5cM or

less (Chao and Ukai 2000, Edwards and Page 1994), whereas others deem for

practical reasons a distance between 10-20cM is more appropriate (Visscher et

al. 1996). For MAS to have the largest impact in selecting lines with the target

genomic regions into elite cultivars, the closer the distance between the

flanking markers is to zero, the greater chance of success (Hospital and

Charcosset 1997).

108
Since the onset of this research project, it has become important to use markers

to select both for the donor/target gene/QTL (termed forward selection) and

also to select for the recurrent parents DNA (termed background selection). It

has been reported that using MAS can accelerate the recovery of the recurrent

parent genome within the Marker-assisted backcrossing method (Frisch et al.

1999, Frisch and Melchinger 2001). The use of background selection also

reduces the number of generations required to recover most of the recurrent

parent genome (see Chapter 1).

MAS has been of the most use in the assisting of backcrossing for major genes

with large effects on traits with a relatively simple inheritance (Holland 2004).

When more complex polygenic traits are being considered, limitations of MAS

can occur. To use QTL material in MAS it is important that great care has been

taken when identifying the QTLs and their position and effects, especially when

considering those with smaller effects (Francia et al. 2005). Because QTLs are

not necessarily the gene controlling a particular trait, it is very important to use

two tightly linked flanking markers, severely reducing the chances of

recombination between the target gene and both the flanking markers.

109
It is necessary to test the effectiveness and robustness of MAS before it can be

an integrated method in wheat breeding and variety development. The work

carried out in this Chapter assesses the usefulness of the molecular markers for

successfully selecting lines with the introgressed target alleles. This analysis

will indicate the ability of molecular markers to be used as predictive tools for

particular phenotypes.

4.2 Materials and Methods

The genotypic data was generated from the field trial conducted in 2005 (for

genotyping details see Chapter 2) and was entered into two different software

packages; these being SPSS (version 15) and Map Manager QTXb20 (Manly et

al. 2001). For the purpose of the statistical analysis, the small proportion of

heterozygotes present in the four populations was removed from the genotypic

data set.

The phenotypic data used for this analysis was collected through the

biochemical testing of PPO activity and Xanthophyll content (see Chapter 3).

110
Linear regressions were carried out to compare the genotype with the

phenotype (biochemical testing for PPO activity and Xanthophyll content), at a

probability of 0.05. Each population was assessed individually.

Single marker regressions were carried out to analyse Xanthophyll content

considering the SSR markers: wmc17, wmc346, gwm566, gwm285 and

gwm108 (for Tasman x Chuanmai 22 and Tasman x Mianyang 11 populations

only). The regression analysis was carried out with the PPO activity data using

the SSR markers: gwm372, gwm526, wmc18, gwm157, cfd233 and gwm301

for the Sunco x Chuanmai 22 and the Sunco x Mianyang 11 populations. The

SSR markers: gwm372, wmc170, gwm526, gwm102, wmc18, wmc181,

gwm349 and gwm301 were used on the Tasman x Chuanmai 22 and the

Tasman x Mianyang 11 populations. The marker set differs between the

populations because not all markers were polymorphic between all four parents.

111
4.3 Results

For the data to be analysed by the computer software it was required that the

heterozygotes were taken out of the genotypic data set. The numbers removed

from each population differed and has been presented in Table 4-1. The

numbers were quite low with the highest being 3.12% (from a total of 897),

therefore the overall effect on the data analysis will be limited.

Table 4-1: Proportions of heterozygote genotypic data points removed from the
genotypic data collected from each of the four backcross populations for the
SPSS data analysis.

Population Total no. of No. of Proportion of

genotypic heterozygote heterozygotes
data points genotypic data found
points found
Sunco x 1240 36 2.90%
Chuanmai 22
Sunco x 1160 33 2.84%
Mianyang 11
Tasman x 975 27 2.77%
Chuanmai 22
Tasman x 897 28 3.12%
Mianyang 11

112
The linear regressions allowed comparisons to be made between each

biochemical test and genotypic set, and the information generated from these

tests has been collated in Table 4-2 and Table 4-3. Firstly each marker was

investigated individually to understand their usefulness as predictive tools for

selection of lines for each trait. Comparisons were made between the marker

data and the PPO activities for all the population data.

113
Table 4-2: Results from the linear Regression analysis for PPO Activity
PPO activity
R2 Significance
Sunco x Chuan Mai22
gwm372 (2A) 0.080 0.010*
gwm526 (2A) 0.001 0.706
wmc18 (2D) 0.063 0.011*
gwm157 (2D) 0.066 0.007*
cfd233 (2D) 0.066 0.006*
gwm301 (2D) 0.049 0.024*

Sunco x Mianyang11
gwm372 (2A) 0.082 0.007*
gwm526 (2A) 0.054 0.022*
wmc18 (2D) 0.001 0.787
gwm157 (2D) 0.000 0.850
cfd233 (2D) 0.000 0.849
gwm301 (2D) 0.005 0.483

Tasman x Chuanmai 22
gwm372 (2A) 0.001 0.788
wmc170 (2A) 0.291 0.000*
gwm526 (2A) 0.007 0.495
gwm102 (2D) 0.025 0.220
wmc18 (2D) 0.035 0.154
wmc181 (2D) 0.216 0.000*
gwm349 (2D) 0.143 0.001*
gwm301 (2D) 0.076 0.029*

Tasman x Mianyang 11
gwm372 (2A) x x
wmc170 (2A) 0.001 0.786
gwm526 (2A) x x
gwm102 (2D) 0.012 0.403
wmc18 (2D 0.106 0.033*
wmc181 (2D) 0.196 0.000*
gwm349 (2D) 0.007 0.495
gwm301 (2D) 0.014 0.389

* - significance level P<0.05

X – Indicates where there was no differences found between the genotypic data, therefore
no regression result can be calculated.

114
The chromosomal locations being investigated are the QTLs from 2A and 2D,

originally identified with the Sunco x Tasman population PPO data. When

considering the 2A QTL, the marker: gwm372 was found to be highly

significant within both Sunco populations and explaining 8% (Sunco x

Chuanmai 22 population) and 8.2% (Sunco x Mianyang 11 population) of the

variation. The second marker: gwm526 only showed a significant result in the

Sunco x Mianyang 11 population explaining 5.4% of the variation. These

markers were not significantly linked to the trait in the Tasman populations,

whereas the marker: wmc170 on chromosome 2A was found to be highly

significant in the Tasman x Chuanmai 22 population only, explaining a

substantial 29.1% of the variation.

When considering the 2D QTL, the four markers used all produced significant

results in the Sunco x Chuanmai 22 population and two out of the four in the

Sunco x Mianyang 11 population. Microsatellite marker wmc18, gwm157,

cfd233 and gwm301 explained 6.3%, 6.6%, 6.6% and 4.9% of the variation

respectively in the Sunco x Chuanmai 22 population. This level of significance

was not found in the Sunco x Mianyang 11 population. When looking at the

Tasman populations, microsatellite wmc181 proved highly significant in both

populations explaining 21.6% (Tasman x Chuanmai 22) and 19.6% (Tasman x

115
Mianyang 11) of the variation. Similarly gwm349 explained 14.3% of the

variation, but only in the Tasman x Chuanmai 22 population and wmc18 in the

Tasman x Mianyang 11 population explaining 10.6% of the population.

Table 4-3: Results from the linear Regression analysis for Xanthophyll Content

Xanthophyll Content
2
R Significance
Sunco x Chuanmai 22
wmc17 (7A) 0.108 0.002*
wmc346 (7A) 0.038 0.053
gwm566 (3B) 0.127 0.001*
gwm285 (3B) 0.135 0.000*
Sunco x Mianyang 11
wmc17 (7A) 0.014 0.262
wmc346 (7A) 0.018 0.193
gwm566 (3B) 0.120 0.000*
gwm285 (3B) 0.158 0.000*

Tasman x Chuanmai 22
wmc17 (7A) 0.028 0.195
wmc346 (7A) 0.344 0.000*
gwm566 (3B) 0.170 0.002*
gwm285 (3B) 0.094 0.012*
gwm108 (3B) 0.104 0.020*
Tasman x Mianyang 11
wmc17 (7A) 0.003 0.686
wmc346 (7A) 0.107 0.016*
gwm566 (3B) 0.016 0.388
gwm285 (3B) 0.065 0.059
gwm108 (3B) 0.023 0.302

*
- significance level P<0.05

116
Linear regression analysis was carried out on the Xanthophyll content data in

the same manner as the PPO activity data. QTLs located on 3B and 7A were

identified in the original Sunco x Tasman population and it is the

microsatellites from these chromosomal areas that were used in the

comparisons between the genotypic data and the Xanthophyll content

phenotypic data. The 3B microsatellite markers (gwm566, gwm285 and

gwm108) were found to produce significant results in three of the four

populations explaining a range of the variation from 9.4% to 17% (Sunco x

Chuanmai 22, Sunco x Mianyang 11 and Tasman x Chuanmai 22 populations).

The microsatellite: gwm285 explained the highest amount of variation in the

Sunco populations, and it was gwm566 that produced the best results in the

Tasman x Chuanmai 22 population explaining the highest amount of variation

for this QTL at 17%. When looking at the 7A microsatellites, wmc346

produced significant results in the Tasman x Chuanmai 22 and Tasman x

Mianyang 11 populations, explaining 34.4% and 10.7% of the variation

respectively. The second marker used in the genotyping was wmc17 and it was

found to produce a significant result in the Sunco x Chuanmai 22 population

only, explaining 10.8% of the variation. Neither of the 7A microsatellites

produced significant results in the, Sunco x Mianyang 11 population.

117
Figure 4-1: Chromosomes 2A, 2D, 3B and 7A of the Sunco x Tasman linkage map,
displaying the map distances (cM) between the microsatellites used in the genotyping
of the four backcrossed populations (’03-’05)
Only SSR markers were used in this figure, the complete QTL peaks can be found in Chapter 2

14.3cM

10.2cM

30.1cM
19.2cM

1.9cM

52.4cM 26.6cM

20.5cM

0.7cM
12.6cM
16.6cM

118
Figure 4-1 displays the map distances in cM (centamorgans) between the

molecular markers used in the genotyping of the four populations (years 2003 –

2005) with the corresponding QTL traces from the Sunco x Tasman Doubled

Haploid linkage map. This diagram allows for a visual assessment of the

distances between the microsatellite markers, their distances away from the

peak and their linkage to the traits to establish which markers should be used

for flanking markers in MAS. When considering the QTL for PPO activity on

2A, the most closely linked marker is wmc170 directly under the peak, but the

distances from the other markers used is quite large at 30.1cM and 52.4cM.

The 2D QTL for PPO activity shows the markers wmc181 and cfd233 situated

under the peak, and at a distance from 1.9cM from each other. Other

microsatellites of interest in this area are gwm157 and gwm349 which are

19.2cM and 26.6cM (respectively) from the peak. Therefore these four markers

are all possibilities for use in selection of lines polymorphic for these markers.

When looking at the QTLs for Xanthophyll content in the Sunco x Tasman

population, there were a smaller number of molecular markers to be assessed,

with only three markers available under the 3B QTL and two microsatellites

available under the 7A QTL. Microsatellites gwm566 and gwm285 are located

directly under the peak of the 3B QTL and are only 0.7cM apart, while

gwm108 is only 12.6cM proximal from the peak. Therefore either marker from

119
the peak and gwm108 can be used as flanking markers for this QTL. The

distance between wmc17 and wmc346 from the 7A QTL is small at 16.6cM,

but wmc17 is quite a distance away from the peak of the QTL.

The in-depth data analysis of all phenotypic and genotypic data has identified

three lines which have produced very good results for both PPO activity and

Xanthophyll content. These three lines also have alleles present for the markers

found to be significant in the earlier linear regressions. These lines were

generated from the Sunco x Chuanmai 22 and Sunco x Mianyang 11 backcross

populations (Table 4-5).

120
Table 4-5: Three lines that produce a low result for PPO activity and
Xanthophyll content. These lines also have marker alleles contributed by
Sunco, some of which are found to be significant and able to explain a
proportion of the variation (in its population). The lines below are all
performing better than either Chinese variety (Chuanmai 22 and Mianyang 11)

Parental PPO Xanthophyll Sunco Marker

Material Activity Content Alleles
Line (OD436nm) present from
(OD415nm) chromosomes
2A,2D, 3B,7A
SM65-10 Sunco 2A: -
(6-15-2) Mianyang 0.117 0.043 2D: gwm157 a ,
11 cfd233 a
3B: gwm285 a
7A: wmc17
SM57-14 Sunco 2A: gwm526 a
(10-29-3) Mianyang 0.127 0.040 2D: -
11 3B: gwm566 a,
gwm285 a
7A: -
SC33-6 Sunco 2A: -
(7-16-2) Chuanmai 0.143 0.037 2D: cfd233 a
22 3B: gwm285 a
7A: -
a
– Markers found to be significant in the linear regression analysis

121
4.4 Discussion

Ideally this type of analysis would be carried out when all lines had finished

segregating, but for the time scale of this project the field trial had to be carried

out in 2005 with a small proportion of the lines in each backcross population

still undergoing segregation. The numbers of heterozygotes were small (a

maximum of 3.12%). At a number of generations, lines (heterozygote or

homozygous for the Australian marker alleles) from these populations have

been selected by molecular markers. Therefore the populations have been

biased towards containing a large number of heterozygotes. If the populations

had undergone random recombination through the two backcrosses and three

single seed descents, then 1.56% of the total population should have been

heterozygotes.

In the original research carried out on the Sunco x Tasman population by Mares

and Campbell (2001) the two QTLs identified for Xanthophyll content were

contributed by different parents; 3B contributed by Sunco and 7A contributed

by Tasman. Two QTLs were identified for high PPO activity on 2A and 2D

and they were contributed by Tasman. Tasman has a very high level of PPO

activity, therefore selecting for the Sunco allele is selecting for a low PPO

122
activity, which is desired in the end-use products. Therefore we need to look at

the four backcross populations considering the two Australian parents will be

contributing very different levels of PPO activity. Ideally for the generation of

lines with low PPO activities we wouldn’t select Tasman as a parent, with its

high level of PPO activity, but for this study, it was interesting to follow the

Tasman allele to observe its effect in a Chinese background which gave a

relatively low PPO activity.

Some of the molecular markers used throughout this process particularly in the

2002 selections, were more loosely linked to the four QTLs than what is ideal.

This study would have benefited from using the markers used in the 2003 to

2005 genotyping from the beginning (see Chapter 2), had they been identified

at that time. However the use of co-dominant microsatellite markers

throughout the process was essential in identifying heterozygote and

homozygous lines containing the Australian parent alleles.

An important aim of this project was to assess the usefulness of molecular

markers for predicting the phenotype of a particular trait of interest. To assess

this tool, comparisons were made between the genotypic and phenotypic data

from the materials selected from the 2005 field trial to get an indication of

123
predictability of these markers for their particular trait of interest. The

populations Sunco x Chuanmai 22 and Tasman x Chuanmai 22 gave the most

consistent results, with significant linkage results being produced for all four of

the QTLs selected. The amount of phenotypic variation explained was 34.4%

(wmc346) for Xanthophyll content and 29.1% (wmc170) for PPO activity in

the Tasman x Chuanmai 22 population. In the Sunco x Tasman population, the

microsatellites, wmc346 and wmc170, explained a smaller amount of the

phenotypic variation, this being 7% (wmc346-3B) and 11% (wmc170-2A)

(Mares and Campbell 2001). The markers that explained the highest proportion

of phenotypic variation in all of the four QTLs investigated was wmc181 (36%)

and cfd233 (38%) for the 2D PPO QTL and gwm566 (14%) and gwm285

(11%) for the 3B Xanthophyll QTL (Mares and Campbell 2001). These

markers, originally identified in the Sunco x Tasman population, were also

successfully validated in the four backcross populations used in this project.

From the regression results from the PPO activity data, the microsatellite

gwm372 from chromosome 2A is effective in selecting for low PPO activity in

both Sunco x Chuanmai 22 and Sunco x Mianyang 11 populations.

Microsatellite wmc170 produced very significant results in the Tasman x

Chuanmai 22 population, and is located at the peak of the 2A QTL. The 2A

124
QTL was contributed by Tasman in the Sunco x Tasman population; therefore

using the Tasman allele for this marker should confer a high PPO activity level.

Unfortunately this marker could not be used to select for the Sunco allele

(selecting against a high PPO activity level) in the Sunco populations because it

was not polymorphic between Sunco and the Chinese varieties. Chromosome

2D has a number of microsatellites that produced significant results and could

be used as flanking markers. In Figure 4-1, wmc181, cfd233 are the two

microsatellites situated at the most ideal position, directly under the QTL peak.

The microsatellites gwm349 and gwm157 are located further away from the

peak, but still close enough to produce very significant results in the regression

data analysis and would be very useful as alternatives in populations where the

closer SSRs were not polymorphic. Used in conjunction with wmc181 and

cfd233 they would be useful in backcrosses to select lines with lower linkage

drag. Lines containing linkage drag can occur when using more distant

flanking markers and with the markers directly under the QTL peak.

Markers on chromosome 3B showed significant linkage to lower Xanthophyll

content. Microsatellites gwm285 and gwm566 were linked to Xanthophyll

content variation in three of the populations (Sunco x Chuanmai 22; Sunco x

Mianyang 11; Tasman x Chuanmai 22). Another promising aspect of these

125
markers is that they are tightly linked at 0.7cM apart on the chromosome,

making them ideal to be used as flanking markers. In contrast the results for

markers on chromosome 7A failed to demonstrate a significant linkage with

Xanthophyll content. This may be due to the 7A markers used during the

selection process for Xanthophyll content. In the Sunco x Tasman population

these markers only explained a small proportion of the phenotypic variation, at

7% for wmc346 and 0% for wmc17. These markers were the only

microsatellite markers available under the 7A QTL; therefore they were

included in the selection process. To successful introgress this QTL into

different genetic backgrounds, markers more tightly linked to the Xanthophyll

content trait are needed.

For MAS to be successful it is essential for the markers being used to be as

tightly linked to the QTL as possible and for the map distances between

flanking markers to be as small as possible. The robustness of the markers was

increased by the work of Lehmensiek et al. (2005), but this occurred after the

first round of selections were made in 2002 (see Chapter 2). There may have

been unwanted segments of chromosomes being transferred via linkage drag

and lines carrying the QTL may have been discarded during selection due to

loss of linkage to the markers following recombination (Holland 2004). These

126
problems are also occurring when the flanking markers used in MAS are too far

apart from each other and/or distant from the peak of the QTL. The literature

supports the use of flanking markers at a distance of 5-20cM apart (Chao and

Ukai 2000, Edwards and Page 1994, Visscher et al. 1996).

Another aim of this project was to successfully introgress one or more QTLs

(2A, 2D: PPO activity; 3B, 7A: Xanthophyll Content) from an Australian

variety into the Chinese genetic background. Considering all of the analysis

carried out in this chapter, it is evident that the introgressions could have been

carried out more successfully with different or more molecular markers, but

there were four lines that were highlighted as important and worth further

investigation. These lines contained differing combinations of the QTLs, with

lines containing three of four QTLs (Line A: 2D, 3B and 7A; Line D: 2A, 2D

and 7A) and lines containing two of the four QTLs (Line B: 2A and 3B; Line

C: 2D and 3B). These lines can now be tested further in Australia and China to

ensure that the selection made from these molecular markers has conferred a

beneficial phenotype in both the Australian and Chinese environments.

127
Chapter 5: General Discussion

5.1 Research Outcomes and the Impact of this Project on Wheat

Improvement

The project objectives outlined in the rationale for the study (Chapter 1) were

all met throughout this research into Marker-assisted selection and its impact on

transferring Quantitative Trait Loci from Australian wheats into Chinese

wheats.

The objectives of this study were:

1. To use the QTLs previously identified by Mares and Campbell (2001)

to improve the flour colour characteristics in backcross populations

generated between Australian varieties (Sunco and Tasman) and

Chinese varieties (Chuanmai 22 and Mianyang 11).

2. To evaluate Marker-assisted selection as an efficient method of

transferring QTLs located in Australian germplasm to a Chinese genetic

background.

128
 3. To test the usefulness of microsatellite markers as a tool to predict

phenotype (Polyphenol oxidase activity and Xanthophyll content) in a

novel genetic background.

4. To increase awareness and usefulness of molecular markers in wheat

breeding in China.

The backcross populations generated in this project have allowed

comprehensive genotypic and phenotypic data to be produced, all of which are

essential in completing the aims of this project. The most important areas of

this study for wheat breeding and its improvement are: the identification of

microsatellites which confer a particular phenotype, and the ability to use

marker-assisted selection to improve the final colour of wheat flour and its

products.

This project was successful in highlighting microsatellites capable of

identifying wheat lines with differing levels of PPO activity and Xanthophyll

content. The microsatellites gwm372, wmc170 (2A – PPO), wmc181, cfd233

(2D – PPO) and gwm285 and gwm566 (3B – Xanthophyll) are all potential

129
markers for use in wheat breeding programs in Australia and China. This work

has also further validated these microsatellite markers which were first proved

to be useful in the Sunco x Tasman Population.

By using the above microsatellites, three wheat lines have been developed

containing different combinations of the four Sunco x Tasman QTLs. These

lines now have QTLs conferring low PPO activity and low Xanthophyll

content, both appropriate characters to produce high quality end-products. As

explained in Chapter 1, the Chinese market demands wheats that produce fresh

noodles with a bright white stable colour. Therefore wheat lines produced from

this project may, with further investigation, confer the good quality attributes

required of wheat in China.

5.2 Future Directions

There are a number of areas resulting from this project that could benefit from

continued research. These are:

a) Marker validation of the above acknowledged microsatellites needs to be

carried out under Australian and Chinese environmental conditions to ensure

130
these markers continue to predict the necessary phenotype. This current project

was only carried out in Australia and over one growing season. Willcox et al.

(2002) reports that a single season’s data is insufficient in the identification of

QTLs, therefore it is important to test the markers found linked to QTLs over a

number of years also.

b) A more comprehensive study is required to assess the potential of the four

lines generated from this project. They have been shown to contain

introgressed QTL material from Sunco, but further phenotypic data on flour and

noodle colour should be produced to test these genomic areas are conferring the

appropriate end-product quality in a range of environments.

c) A number of studies have used a method termed “background marker

selection” to enhance the recovery of the recurrent parent genotype during

backcrossing, and to limit the amount of linkage drag occurring during the QTL

introgression (Bonnett et al. 2005, Howes et al. 1998, Kuchel et al. 2007). This

type of marker selection was not carried out during this project therefore it is

not known what effect linkage drag has had on the lines developed. To better

understand the amount of genomic material surrounding the QTLs transferred,

131
markers could be screened across the genome to assess the amount of recurrent

(Chinese) parent genotype present.

d) The markers available from the 7A QTL were not able to explain the

variation in Xanthophyll content in this project. If this QTL is to be studied

further an effort to identify other markers on 7A is required.

5.3 Conclusion

Markers have been used successfully in wheat breeding programs around the

world. However, for markers to become a completely integrated tool, they

must be reliable and efficient at all times. This project has highlighted the

usefulness of molecular markers in the improvement of quality attributes

(colour) in wheat varieties.

This project has established links between Australia and China, and this

collaboration enables the sharing of elite germplasm, phenotypic and genotypic

techniques and the potential to improve the current varieties available to

farmers.

132
References

ABARE (2008) Australian Crop Report (no. 147).

http://www.abare.gov.au/publications_html/cr/cr_08/cr08_Sept.pdf

Allard RW (1960) 'Principles of Plant Breeding.' (John Wiley & Sons: New
York).

Arús P, Moreno-González J (1993) Marker-assisted selection. In 'Plant

Breeding: Principles and prospects'. (Eds MD Hayward, NO Bosemark, I
Romagosa). (Chapman & Hall: London).

Asenstorfer RE, Wang Y, Mares DJ (2006) Chemical structure of flavonoid

compounds in wheat (Triticum aestivum L.) flour that contribute to the yellow
colour of Asian alkaline noodles. Journal of Cereal Science 43, 108-119.

Asins MJ (2002) Present and future of quantitative trait locus analysis in plant
breeding. Plant Breeding 121, 281-291.

AWB (2008) Grain Production.

www.awb.com.au/aboutawb/communityeducation/grainproduction

Babu R, Nair SK, Prasanna BM, Gupta HS (2004) Integrating marker-assisted

selection in crop breeding - Prospects and challenges. Current Science 87, 607-
619.

Bernier A-M, Howes NK (1994) Quantification of Variation in Tyrosinase

Activity AMong Durum and Common Wheat Cultivars. Journal of Cereal
Science 19, 157-159.

Bhar L, Gupta VK (2001) A useful statistic for studying outliers in

experimental designs. The Indian Journal of Statistics 6, 338-350.

Bhattacharya M, Luo Q, Corke H (1999) Time-dependent changes in dough

colour in hexaploid wheat landraces differing in polyphenol oxidase activity.
Journal of Agricultural and Food Chemistry 47, 3579-3585.

Bonnett DG, Rebetzke GJ, Spielmeyer W (2002) Population size and the
importance of inbreeding with marker-assisted selection of wheat. In '12th

133
Australasian Plant Breeding Conference'. Perth. (Ed. JA McComb) pp. 655-
662. (Australasian Plant Breeding Association Inc.).

Bonnett DG, Rebetzke GJ, Spielmeyer W (2005) Strategies for efficient

implementation of molecular markers in wheat breeding. Molecular Breeding
15, 75-85.

Brennan PS, Martin DJ, et al. (1993) Triticum aestivum ssp. vulgare (bread
wheat) cv. Tasman. Canberra.
http://www.publish.csiro.au/?act=view_file&file_id=EA9940872.pdf

Brettell R (2006) A further review of wheat traits in relation to pre-breeding

and breeding objectives.
http://www.grdc.com.au/uploads/documents/traitsreport0207.pdf

Burdick EM (1956) Extraction and utilisation of carotenes and xanthophylls

Economic Botany 10, 267-279.

Bushuk W (1998) Wheat breeding for end-product use. Euphytica 100, 137-
145.

Chalmers KJ, Campbell AW, et al. (2001) Constructions of three linkage maps
in bread wheat, Triticum aestivum. Australian Journal of Agricultural Research
52, 1089-1119.

Chao Y-T, Ukai Y (2000) Sample size required for marker assisted selection in
improving quantitative traits of self-fertilizing species. Euphytica 116, 87-94.

china.org.cn (2006) China Aims for 33.5 Million Hectares of 'Super Wheat' by
2020 http://german.china.org.cn/english/scitech/182897.htm

Christopher M, Banks P (2002) Implementation of PCR-based markers in

wheat breeding for northern Australia. In '12th Australasian Plant Breeding
Conference'. Perth, Western Australia. (Ed. JA McComb) pp. 728-732.
(Australasian Plant Breeding Association Inc.).

Collard BCY, Jahufer MZZ, Brouwer JB, Pang ECK (2005) An introduction to
markers, quantitative trait loci (QTL) mapping and marker-assisted selection
for crop improvement: The basic concepts. Euphytica 142, 169-196.

134
CommodityOnline (2008) China, India lead global wheat production.
www.commodityonline.com/printnews.php?news_id=6300

Crosbie GB, Huang S, Barclay IR (1998) Wheat quality requirements of Asian

foods. Euphytica 100, 155-156.

Crosbie GB, Lambe WJ (1993) The Application of the Flour Swelling Volume
Test for Potential Noodle Quality to Wheat Breeding Lines Affected by
Sprouting. Journal of Cereal Science 18, 267-276.

Demeke T, Morris CF (2002) Molecular characterization of wheat polyphenol

oxidase (PPO). Theoretical and Applied Genetics 104, 813-818.

Demeke T, Morris CF, Campbell KG, King GE, Anderson JA, Chang H-G
(2001) Wheat polyphenol oxidase: Distribution and genetic mapping in three
inbred line populations. Crop Science 41, 1750-1757.

Dewan T (1988) Marketing implications of wheat quality.

http://www.regional.org.au/au/roc/1988/roc198807.htm

Doerksen TK, Kannenberg LW, Lee EA (2003) Effect of Recurrent selection

on combining ability in maize breeding populations. Crop Science 43, 1652-
1658.

Dubcovsky J (2004) Marker-assisted selection in public breeding porgrams: the

wheat experience. Crop Science 44, 1895-.

Eagles HA, Bariana HS, Ogbonnaya FC, Rebetzke GJ, Hollamby GJ, Henry RJ,
Henschke PH, Carter M (2001) Implementation of markers in Australian wheat
breeding. Australian Journal of Agricultural Research 52, 1349-1356.

Eathington SR, Dudley JW, Rufener II GK (1997) Usefulness of marker-QTL

associations in early generation selection. Crop Science 37, 1686-.

Edwards MD, Page NJ (1994) Evaluation of marker-assisted selection through

computer simulation. Theoretical and Applied Genetics 88, 376-382.

Elouafi I, Nachit MM, Martin LM (2001) Identification of a microsatellite on

chromsome 7B showing a strong linkage with yellow pigment in durum wheat
(Triticum turgidum L. var. durum). Hereditas 135, 255-261.

135
Feldman M (2001) Origin of cultivated wheat. In 'The World Wheat Book - A
History of Wheat Breeding'. (Eds AP Bonjean, WJ Angus). (Intercept Ltd:
Andover, UK.).

Fischer T (2007) Improvements in wheat yield. (Ed. SAAN Zealand).

http://www.sciencealert.com.au/opinions/20070608-16169.html

Francia E, Tacconi G, Crosatti C, Barabaschi D, Bulgarelli D, Dall'Aglio E,

Valé G (2005) Marker assisted selection in crop plants. Plant Cell, Tissue and
Organ Culture 82, 317-342.

Frisch M (2005) Breeding Strategies: Optimum Design of Marker-Assisted

Backcross Programs. In 'Biotechnology in Agriculture & Forestry: Molecular
Marker Systems in Plant Breeding & Crop Improvement'. (Eds H Lörz, G
Wenzel) pp. 319-334. (Springer-Verlag: Berlin).

Frisch M, Bohn M, Melchinger AE (1999) Comparison of selection strategies

for marker-assisted backcrossing of a gene. Crop Science 39, 1295-1301.

Frisch M, Melchinger AE (2001) Marker-Assisted Backcrossing for

Simultaneous Introgression of Two Genes. Crop Science 41, 1716-1725.

Fu B, X. (2008-in press) Asian noodles: History, classification, raw materials,

and processing. Food Research International in press.

Fulton TM, Nelson JC, Tanksley SD (1997) Introgression and DNA marker
analysis of Lycopersicon peruvianum, a wild relative of the cultivated tomato,
into Lycopersicon esculentum, followed through three successive backcross
generations. Theoretical and Applied Genetics 95, 895-902.

Gale KR (2005) Diagnostic DNA markers for quality traits in wheat. Journal of
Cereal Science 41, 181-192.

Grandillo S, Fulton TM (1998) Approaches to gene mapping. In 'Molecular

Plant Biology'. (Eds PM Gilmartin, C Bowler) pp. 101-117. (Oxford University
Press: U.K.).

136
GRDC (2008) Wheat Traits, Summary and Discussion Document.
http://www.grdc.com.au/uploads/documents/AWCPA%20Traits%20Summary
%2020081.pdf

Gupta PK, Balyan HS, et al. (2002) Genetic mapping of 66 new microsatellite
(SSR) loci in bread wheat. Theoretical and Applied Genetics 105, 413-422.

Habernnicht DK, Berg JE, Carlson GR, Wichman DM, Kushnak GD, Kephart
KD, Martin JM, Bruckner PL (2002) Pan bread and raw Chinese noodle
qualities in hard winter wheat genotypes grown in water-limited environments.
Crop Science 42, 1396-1403.

Hackett CA, Broadfoot LB (2003) Effects of genotyping errors, missing values

and segregation distortion in molecular marker data on the construction of
linkage maps. Heredity 90, 33-38.

Harker N, Rampling LR, Shariflou MR, Hayden MJ, Holton TA, Morell MK,
Sharp PJ, Henry RJ, Edwards KJ (2001) Microsatellites as markers for
Australian wheat improvement. Australian Journal of Agricultural Research
52, 1121-1130.

Hatcher DW, Anderson MJ, Desjardins RG, Edwards NM, Dexter JE (2002)
Effects of flour particle size and starch damage on processing and quality of
white salted noodles. Cereal Chemistry 79, 64-71.

He Z (1999) Wheat Production and Quality Requirements in China. In 'Ninth

Assembly of Wheat Breeding Society of Australia'. Toowoomba, Queensland,
Australia. (Eds P Williamson, P Banks, I Haak, J Thompson, AW Campbell)
pp. 23-27. (University of Southern Queensland).

He ZH, Rajaram S, Xin ZY (2001) The Southwestern Autumn-sown Spring

Wheat Zone. In 'A History of Wheat Breeding in China'. (Eds ZH He, S
Rajaram, ZY Xin) pp. 51-59. (CIMMYT - International Maize and Wheat
Improvement Centre: Mexico).

He ZH, Yang J, Zhang Y, Quail KJ, Pena RJ (2004) Pan bread and dry white
Chinese noodle quality in Chinese winter wheats. Euphytica 139, 257-267.

137
Holland JB (2004) Implementation of molecualr markers for quantitative traits
in breeding programs - challenges and opportunities. In '4th International Crop
Science Congress'. Brisbane, Australia. (Crop Science).

Hospital F, Charcosset A (1997) Marker-assisted introgression of quantitative

trait loci. Genetics 147, 1469-1485.

Hospital F, Moreau L, Lacoudre F, Charcosset A, Gallais A (1997) More on the

efficiency of marker-assisted selection. Theoretical and Applied Genetics 95,
1181-1189.

Howes NK, Woods SM, Townley-Smith TF (1998) Simulations and practical

problems of applying multiple marker assisted selection and doubled haploids
to wheat breeding programs. Euphytica 100, 225-230.

Hu X-Z, Wei Y-M, Kovacs MIP, Wang C (2004) Swelling index of glutenin
realted to dough characters and noodle quality. Scientia Agricultura Sinica 37,
119-124.

Huang XQ, Cöster H, Ganal MW, Röder MS (2003) Advanced backcross QTL
analysis for the identification of quantitative trait loci alleles from wild relatives
of wheat (Triticum aestivum L.). Theoretical and Applied Genetics 106, 1379-
1389.

Huang XQ, Kempf H, Ganal MW, Röder MS (2004) Advanced backcross QTL
analysis in progenies derived from a cross between a German elite winter wheat
variety and a synthetic wheat (Triticum aestivum L.). Theoretical and Applied
Genetics 109, 933-943.

Jansen RC (1993) Interval mapping of multiple quantitative trait loci. Genetics

135, 205-211.

Jones CJ, Edwards KJ, et al. (1997) Reproducibility testing of RAPD, AFLP
and SSR markers in plants by a network of European laboratories. Molecular
Breeding 3, 381-390.

Kammholz SJ, Sutherland MW, Banks PM (1996) Improving the efficiency of

haploid wheat production mediated by wide crossing. Sabrao Journal 28, 37-
46.

138
Kammholz SJ (2001) Doubled haploids and their application to wheat breeding.
University of Southern Queensland. PhD thesis.

Kammholz SJ, Campbell AW, et al. (2001) Establishment and characterisation

of wheat genetic mapping populations. Australian Journal of Agricultural
Research 52, 1079-1088.

Knapp SJ (1998) Marker-assisted selection as a strategy for increasing the

probability of selecting superior genotypes. Crop Science 38, 1164-1174.

Koebner RMD, Summers RW (2003) 21st century wheat breeding: plot

selection or plate detection? Trends in Biotechnology 21, 59-63.

Korzun V (2002) Use of molecular markers in cereal breeding. Cellular &

Molecular Biology Letters 7, 811-820.

Korzun V (2003) Molecular markers and their applications in cereal breeding.

In '10th Biotechnology in Food & Agriculture Conference'.

Kruger JE (1976) Changes in the Polyphenol Oxidases of wheat during kernel

growth and maturation. Cereal Chemistry 53, 201-213.

Kruger JE, Matsuo RR, Preston K (1992) A comparison of methods for the
prediction of Cantonese noodle colour. Canadian Journal of Plant Science 72,
1021-1029.

Kruger NL, Cooper M, Podlich DW (2002) Comparison of phenotypic, marker

and marker-assisted selection strategies in an S1 family recurrent selection
strategy. In '12th Australasian Plant Breeding Conference'. Perth, Western
Australia. (Ed. JA McComb) pp. 696-701.

Kuchel H, Fox R, Reinheimer J, Mosionek L, Willey N, Bariana HS, Jefferies S

(2007) The successful application of a marker-assisted wheat breeding strategy.
Molecular Breeding 20, 295-308.

Lande R, Thompson R (1990) Efficiency of Marker-Assisted Selection in the

Improvement of Quantitative Traits. Genetics 124, 743-756.

Lander ES, Botstein D (1989) Mapping mendelian factors underlying

quantitative traits using RFLP linkage maps. Genetics 121, 185-199.

139
Lander ES, Green P, Abrahamson J, Barlow A, Daly MJ, Lincoln SE, Newburg
L (1987) MAPMAKER: an interactive computer package for constructing
primary genetic linkage maps of experimental and natural populations.
Genomics 1, 174-181.

Landjeva S, Korzun V, Börner A (2007) Molecular markers: actual and

potential contributions to wheat genome characterization and breeding.
Euphytica 156, 271-296.

Langridge P, Chalmers K (2005) The Principle: Identification and Application

of Molecular Markers. In 'Biotechnology in Agriculture & Forestry: Molecular
Marker Systems in Plant Breeding & Crop Improvement'. (Eds H Lörz, G
Wenzel) pp. 3-21. (Springer-Verlag: Berlin).

Lehmensiek A, Eckermann PJ, Verbyla AP, Appels R, Sutherland MW,

Daggard GE (2005) Curation of wheat maps to improve map accuracry and
QTL detection. Australian Journal of Agricultural Research 56, 1347-1354.

Lepage M, Sims RPA (1968) Carotenoids of Wheat Flour: Their Identification

and Composition. Cereal Chemistry 45, 600-604.

Liu P, Zhu J, Lu Y (2004) Marker-assisted selection in segregating generations

of self-fertilizing crops. Theoretical and Applied Genetics 109, 370-376.

Manly KF, Cudmore RHJ, Meer JM (2001) Map Manager QTX, cross-platform
software for genetic mapping. Mammalian Genome 12, 930-932.

Mares DJ, Campbell AW (2001) Mapping components of flour and noodle

colour in Australian wheat. Australian Journal of Agricultural Research 52,
1297-1309.

Mares DJ, Panozzo JF (1999) Impact of selection for low grain polyphenol
oxidase activity on darkening in Asian noodles. In 'Ninth Assemby of Wheat
Breeding Society of Australia'. Toowoomba, Queensland, Australia. (Eds P
Williamson, P Banks, I Haak, J Thompson, AW Campbell) pp. 32-33.
(University of Southern Queensland).

Mares DJ, Wang Y, Baydoun M (2000) Colour of Asain Noodles: Stability of

Xanthophylls and Flavonoids and Interaction with Darkening. In '11th ICC

140
Cereal and Bread Congress and 50th Australian Cereal Chemisty Conference'.
Gold Coast, Queensland, Australia. (Eds M Wooton, IL Satey, CW Wrigley).
(Cereal Chemistry Division, Royal Australian Chemical Institute.).

Mares DJ, Wang Y, Cassidy CA (1997) Separation, Identification and Tissue

Location of Compounds Responsible for the Yellow colour of Alkaline
Noodles. In '47th Australian Cereal Chemisty Conference'. Perth, Western
Australia. (Eds AW Tarr, AS Ross, CW Wrigley) pp. 114-117. (Cereal
Chemistry Division, Royal Australian Chemical Institute.).

Martin DJ (1996) Developing quality wheat varieties for the market.

http://www.regional.org.au/au/asa/1996/contributed/409martin.htm

McCaig TN, Fenn DYK, Knox RM, DePauw RM, Clarke JM, McLeod JG
(1999) Measuring polyphenol oxidase activity in a wheat breeding program.
Canadian Journal of Plant Science 79, 507-514.

McCormick KM, Panozzo JF, Hong SH (1991) A swelling power test for
selecting potential noodle quality wheats. Australian Journal of Agricultural
Research 42, 317-323.

McIntrye L (2006) Drought-tolerant wheat quest. In 'Farming Ahead'. (Ed.

CSIRO).

Miskelly DM (1984) Flour components affecting paste and noodle colour.

Journal of the Science of Food and Agriculture 35, 463-471.

Moreau L, Charcosset A, Gallais A (2004) Experimental evaluation of several

cycles of marker-assisted selection in maize. Euphytica 137, 111-118.

Moreau L, Charcosset A, Hospital F, Gallais A (1998) Marker-assisted

selection efficiency in populations of finite size. Genetics 148, 1353-1365.

Moreno-González J, Cubero JI (1993) Selection strategies and choice of

breeding methods. In 'Plant Breeding: Principles and prospects'. (Eds MD
Hayward, NO Bosemark, I Romagosa). (Chapman & Hall: London).

Moros EE, Darnoko D, Cheryan M, Perkins EG, Jerrell J (2002) Analysis of

Xanthophylls in Corn by HPLC. Journal of Agricultural and Food Chemistry
50, 5787-5790.

141
Moss HJ (1971) The quality of noodles prepared from the flours of some
Australian wheats. Australian Journal of Experimental Agriculture & Animal
Husbandry 11, 243-247.

MOST (2006) New Targets for Super Wheat.

http://most.gov.cn/eng/newsletters/2006/200610/t20061027_37165.htm

Najafian G, Ghandi A, Abdi H (2004) Screening for late season drought

tolerance in wheat genotypes grown in Iran. In '4th International Crop Science
Congress'. Brisbane, Australia. (Ed. T Fischer). (The Regional Institute Ltd).

Nelson JC (1997) QGENE: software for marker-based genomic analysis and

breeding. Molecular Breeding 3, 239-245.

NSW-DPI (2008) Winter Crop Variety Sowing Guide 2008.

http://www.dpi.nsw.gov.au/__data/assets/pdf_file/0003/108633/wcvsg-
Part2.pdf

O'Brien L, De Pauw R (2004) Breeding. In 'WHEAT/Breeding' pp. 330-336.

Parker GD, Chalmers KJ, Rathjen AJ, Langridge P (1998) Mapping loci with
flour colour in wheat (Triticum aestivum L.). Theoretical and Applied Genetics
97, 238-245.

Pestsova E, Ganal MW, Röder MS (2000) Isolation and mapping of

microsatellite markers specidic for the D genome of bread wheat. Genome 43,
689-697.

Poehlman JM (1994) Breeding Self-Pollinating Crops. In 'Breeding Field

Crops' pp. 187-211. (Iowa State University Press: Ames, Iowa).

Poehlman JM, Sleper DA (1995) 'Breeding Field Crops.' (Iowa State University
Press: Ames).

QDPI&F (2008) Wheat - diseases, physiological disorders and frost. (Ed.

QDPI&F).
http://www.dpi.qld.gov.au/cps/rde/dpi/hs.xsl/26_3537_ENA_HTML.htm#disea
ses

142
Rajesh, Takashima W, Kaneto K (2005) Simultaneous co-immobilization of
enzyme and a redox mediator in polypyrrole film for the fabrication of an
amperometic phenol biosensor. Current Applied Physics 5, 184-188.

Raman R, Raman H, Johnstone K, Lisle C, Smith A, Martin P, Allen H (2005)

Genetic and in silico comparative mapping of the polyphenol oxidase gene in
bread wheat (Triticum aestivum L.). Functional and Integrative Genomics 5,
185-200.

Raman R, Raman H, Martin P (2007) Functional gene markers for polyphenol

oxidase locus in bread wheat (Triticum aestivum L.). Molecular Breeding 19,
315-328.

Reyes-Valdés MH (2000) A model for marker-based selection in gene

introgression breeding programs. Crop Science 40, 91-98.

Reynolds MP (1998) Increasing yield potential in wheat: complementing

conventional breeding by application of novel physiological and germplasm
strategies.

Ribaut J, Betrán J (1999) Single large-scale marker-assisted selection (SLS-

MAS). Molecular Breeding 5, 531-541.

Ribaut J, Hoisington D (1998) Marker-assisted selection: new tools and

strategies. Trends in Plant Science 3, 236-239.

Ribaut JM, Jiang C, Hoisington D (2002) Simulation Experiments on

Efficiencies of Gene Introgression by Backcrossing. Crop Science 42, 557-565.

Robert VJM, West MAL, Inai S, Caines A, Arntzen L, Smith JK, St.Clair DA
(2001) Marker-assisted introgression of blackmold resistance QTL alleles from
wild Lycopersicon cheesmanii to cultivated tomato (L.esculentum) and
evaluation of QTL phenotypic effects. Molecular Breeding 8, 217-233.

Röder MS, Korzun V, Wendehake K, Plaschke J, Tixier M, Leroy P, Ganal

MW (1998) A Microsatellite Map of Wheat. Genetics 149, 2007-2023.

Röder MS, Plaschke J, König SU, Börner A, Sorrells ME, Tanksley SD, Ganal
MW (1995) Abundance, variability and chromosomal location of
microsatellites in wheat. Molecular and General Genetics 246, 327-333.

143
Sanchez-Monge E (1993) Introduction. In 'Plant Breeding - Principles and
prospects'. (Eds MD Hayward, NO Bosemark, I Romagosa). (Chapman & Hall:
London, UK).

Servin B, Hospital F (2002) Optimal positioning of markers to control genetic

background in marker-assisted backcrossing. The Journal of Heredity 93, 214-
217.

Singh R, Rajaram S (1992) Genetics of adult-plant resistance to leaf rust in

'Frontana' and three CIMMYT wheats. Genome 35, 24-31.

Song QJ, Shi JR, Singh S, Fickus EW, Costa JM, Lewis J, Gill BS, Ward R,
Cregan PB (2005) Development and mapping of microsatellite (SSR) markers
in wheat. Theoretical and Applied Genetics 110, 550-560.

Sorrells ME, Wilson WA (1997) Direct classification and selection of superior

alleles for crop improvement. Crop Science 37, 691-.

Sourdille P, Tavaud M, Charmet G, Bernard M (2001) Transferability of wheat

microsatellites to diploid Triticeae species carrying the A, B and D genomes.
Theoretical and Applied Genetics 103, 346-352.

Spennemann DHR (2001) Wheat varieties grown in 19th century Australia - a

handlist of varieties.

Stam P (1993) Construction of integrated linkage maps by means of a new

computer package: JoinMap. The Plant Journal 3, 739-744.

Stam P, Zeven AC (1981) The theoretical proportion of the donor genome in

near-isogenic lines of self-fertilisers bred by backcrossing. Euphytica 30, 227-
238.

Steele KA, Price AH, Shashidhar HE, Witcombe JR (2006) Marker-assisted

selection to introgress rice QTLs controlling root traits into an Indian upland
rice variety. In 'Theoretical and Applied Genetics' pp. 208-221.

Steffens JC, Harel E, Hunt MD (1994) Polyphenol Oxidase. In 'Genetic

Engineering of Plant Secondary Metabolism'. (Eds BE Ellis, GW Kuroki, HA
Stafford) pp. 275-312. (Plenum Press: New York).

144
Stephenson P, Glenn B, Kirby J, Collins A, Devos KM, Busso C, Gale MD
(1998) Fifty new microsatellite loci for the wheat genetic map. Theoretical and
Applied Genetics 97, 946-949.

Storlie E, Yang EN, Zou Y, Chen DS, Sheppard JA, Martin DJ, Huang S,
Mares DJ, Sutherland M (2006) Effect of the puroindoline locus and
environment on Chinese fresh noodle texture. Australian Journal of
Agricultural Research 57, 537-542.

Stoskopf NC, Tomes DT, Christie BR (1993) 'Plant Breeding. Theory and
Practice.' (Westview Press Inc: Colorado).

Taneja SR, Abrol YP, Sachar RC (1974) Modulation of o-Diphenolase and

Monophenolase Enzymes during Wheat Grain Development. Cereal Chemistry
51, 457-465.

Tanksley SD, Grandillo S, Fulton TM, Zamir D, Eshed Y, Petiard V, Lopez J,

Beck-Bunn T (1996) Advanced backcross QTL analysis in a cross between an
elite processing line of tomato and its wild relative L. pimpinellifolium.
Theoretical and Applied Genetics 92, 213-224.

Tanksley SD, Nelson JC (1996) Advanced backcross QTL analysis: a method

for the simultaneous discovery and transfer of valuable QTLs from unadapted
germplasm into elite breeding lines. Theoretical and Applied Genetics 92, 191-
203.

Tomar SMS, Menon MK (1998) Introgression of alien genes for leaf rust
(Puccinia recondita) resistance into bread wheat (Triticum aestivum) culivars.
Indian Journal of Agricultural Sciences 68, 675-681.

Toojinda T, Baird E, Booth A, Broers L, Hayes P, Powell W, Thomas W, Vivar

H, Young G (1998) Introgression of quantitative trait loci (QTLs) determining
stripe rust resistance in barley: an example of marker-assisted line development.
Theoretical and Applied Genetics 96, 123-131.

USDA (2008a) China Winter Wheat Planting Situation - 2007/08 Season

http://www.pecad.fas.usda.gov/highlights/2006/10/ch_30oct2006/index.htm

145
USDA (2008b) Grain: World Markets and Trade.
http://www.fas.usda.gov/psdonline/circulars/grain.pdf

Van Os H, Stam P, Visser RGF, Van Eck HJ (2005) RECORD: a novel method
for ordering loci on a genetic linkage map. Theoretical and Applied Genetics
112, 30-40.

Varshney RK, Graner A, Sorrells ME (2005) Genomics-assisted breeding for

crop improvement. Trends in Plant Science 10, 621-630.

Visscher PM, Haley CS, Thompson R (1996) Marker-assisted introgression in

backcross breeding programs. Genetics 144, 1923 - 1932.

Willcox MC, Khairallah MM, et al. (2002) Selection for Resistance to

Southwestern Corn Borer Using Marker-Assisted and Convential Backcrossing.
Crop Science 42, 1516-1528.

William HM, Trethowan R, Crosby-Galvan EM (2007) Wheat breeding

assisted by markers: CIMMYT's experience. Euphytica 157, 307-319.

Wu J, Jenkins J, Zhu J, McCarty J, Watson C (2003) Monte carlo simulations

on marker grouping and ordering. Theoretical and Applied Genetics 107, 568-
573.

Young ND (1999) A cautiously optimistic vision for marker-assisted breeding.

Molecular Breeding 5, 505-510.

Yousef GG, Juvik JA (2002) Enhancement of Seedling Emergence in Sweet

Corn by Marker-Assisted Backcrossing of Beneficial QTL. Crop Science 42,
96-104.

Zeng Z-B (1994) Precision mapping of Quantitative Trait Loci. Genetics 136,
1457-1468.

Zhang Y, Nagamine T, He ZH, Ge XX, Yoshida H, Peña RJ (2005) Variation

in quality traits in common wheat as related to Chinese fresh white noodle
quality. Euphytica 141, 113-120.

146
Zhang Y, Wu Y, Xiao Y, He Z, Zhang Y, Yan J, Zhang Y, Xia X, Ma C (2008)
QTL Mapping for flour and noodle colour components and yellow pigment
content in common wheat. Euphytica online first.

147
Appendix A: Protocols used in the Genotypic and Phenotypic
assessment of the backcross populations

Protocol 1:
Phenol/Chloroform Method for genomic DNA Extraction

1. Leaf material ground using liquid nitrogen in a mortar and pestle until a fine
white powder was formed.
2. 4ml of Extraction Buffer added and mixed.
3. 4ml of Phenol/Chloroform/Iso-amylalcohol (25:24:1) added and mixed.
4. Centrifuge samples at 5000rpm for 10 minutes in a bench-top centrifuge.
5. Top Phase Layer of the supernatant transferred into a clean tube
6. 400ml of Phenol/Chloroform/Iso-amylalcohol (25:24:1) added for a re-
extraction step.
7. Centrifuge samples at 5000rpm for 10 minutes.
8. Top Phase Layer of the supernatant again transferred into a clean tube
9. 400µl 3M Sodium Acetate (pH 4.8) and 4 ml Isopropanol. Mixed very
gently
10. Centrifuge samples at 5000rpm for 10 minutes to pellet DNA.
11. Gently remove and discard supernatant. Wash DNA pellet with 4ml 70%
Ethanol.
12. Centrifuge at 5000rpm for 5 minutes.
13. Remove and discard supernatant and allow DNA pellet to air dry.
14. Add 350µl 1x TE Buffer to dried DNA pellet for resuspending the DNA.
15. Store DNA samples at 4oC for short term used or store at -20oC for long
term storage.

148
 Reagents:
DNA Extraction Buffer TE Buffer
1% Sarkosyl 10mM Tris-HCl
100mM Tris-HCl 1mM EDTA
100mM NaCl pH 8.0
10mM EDTA
pH 8.5

Protocol 2:

Wizard Genomic DNA Extraction (tubes)

Reagents for this method are commercially made therefore the exact
components are not known.
1. Into a 1.5ml microfuge tube place a sample of leaf material, add 4 ball
bearings
2. Add 600µl of Nuclei Lysis Solution
3. Place the tubes into the Mixer Mill Adapters and mix for 2 mins at 30
shakes/sec.
Repeat.
4. Incubate at 65oC for 15 mins
5. Add 10-12µl of RNase Solution to the lysate, Mix (by inverting tubes 2-5
times)
Incubate at 37 oC for 15 mins
Allow the sample to cool to Room Temp for 5 mins before proceeding

149
6. Add 200µl of Protein Precipitation Solution and Vortex vigorously at high
speed for
20 secs
7. Centrifuge for 3 mins at 13 000 – 16 000 x g (14,000 x g = 11, 481rpm)
Precipitated proteins will form a tight pellet
8. Carefully remove the supernatant containing the DNA
Transfer into a clean 1.5ml microfuge tube containing 600 µl Isopropanol
(Room Temp).
9. Gently Mix the solution by inversion until thread-like strands of DNA
form a visible mass.
10. Centrifuge at 13 000 – 16 000 x g (14,000 x g = 11, 481rpm) for 1 min at
Room Temp
11. Carefully remove the supernatant
Add 600µl 70% Ethanol (Room Temp) and gently invert the tube several
times to wash
the DNA pellet.
12. Centrifuge at 13 000 – 16 000 x g (14,000 x g = 11, 481rpm) for 1 min at
Room Temp.
13. Carefully aspirate the ethanol using a pasteur pipette
(DNA pellet is very loose at this point)
14. Invert the tube and air-dry the pellet for 15 mins
15. Add 100µl of DNA Rehydration Solution or sterile milli Q water and re-
hydrate the DNA by incubating at 65 oC for 1 hr, periodically mixing.
Or, incubation the solution overnight at Room Temp or 4oC.
16. Store DNA at 4oC

150
Protocol 3:

Wizard Genomic DNA Extraction (plates)

Reagents for this method are commercially made therefore the exact
components are not known.

1. Into each tube of the 96 well plate place a sample of leaf material, add 4
ball bearings
2. Add 500µl of Nuclei Lysis Solution
3. Place the plate into the Mixer Mill Adapters and mix for 2 mins at 30
shakes/sec.
Repeat, changing sides.
4. Incubate at 65oC for 15 mins
5. Add 10-12µl of RNase Solution to the lysate, Mix (by inverting tubes 2-5
times)
Incubate at 37 oC for 15 mins
Allow the sample to cool to Room Temp for 5 mins before proceeding
6. Add 200µl of Protein Precipitation Solution and Vortex vigorously at high
speed for
20 secs
7. Centrifuge for 3 mins at 4000 rpm, Precipitated proteins will form a tight
pellet
8. Carefully remove the supernatant containing the DNA
Transfer into a clean tubes containing 500 µl Isopropanol (Room Temp).
9. Gently Mix the solution by inversion until thread-like strands of DNA form
a visible mass.

151
10. Centrifuge at 4000rpm for 1 min at Room Temp
11. Carefully remove the supernatant
Add 500µl 70% Ethanol (Room Temp) and gently invert the tube several
times to wash
the DNA pellet.
12. Centrifuge at 4000rpm for 1 min at Room Temp.
13. Carefully remove the ethanol (DNA pellet is very loose at this point)
14. Invert the tube and air-dry the pellet for 15 mins
15. Add 100µl of DNA Rehydration Solution or sterile milli Q water and re-
hydrate the DNA by
incubating at 65 oC for 1 hr, periodically mixing.
Or, incubation the solution overnight at Room Temp or 4oC.
16. Store DNA at 2 – 8oC

Protocol 4:
Qiagen DNeasy Plant DNA Purification Kit

All Reagents supplied with the Qiagen Kit

1. Leaf tissue ground using liquid nitrogen in a mortar and pestle until a fine
white powder is produced. Transfer into a microfuge tube.
2. Add 400µl of Buffer AP1 and 4µl of RNase A solution (100mg/ml). Mix
well.
3. Incubate samples at 65oC for 10 minutes, mixing during this time.
4. Add 130µl of Buffer AP2 and mix. Incubate for a further 5 minutes on ice.
5. Centrifuge samples at 14,000rpm for 5 minutes.

152
6. Transfer the supernatant into the QIAshredder Mini spin column and
centrifuge at 14,000rpm for 2 minutes.
7. Transfer the flow-through fraction into a new tube.
8. Add 1.5 volumes of Buffer AP3/E to the cleared lysate and mix.
9. Pipette 650µl of the mixture into the DNeasy Mini spin column. Centrifuge
at 8000rpm for 1 mintue. Discard the flow-through.
10. Repeat the above step with the remaining sample and discard the flow-
through.
11. Place the DNeasy Mini spin column into a new tube and add 500µl of
Buffer AW. Centrifuge at 8000rpm for 1 mintue. Discard the flow-through.
12. Add 500µl of Buffer AW to the spin column and centrifuge at 14,000rpm for
2 mintues to dry the membrane.
13. Transfer the DNeasy Mini spin column to a new tube and add 100µl Buffer
AE directly onto the membrane. Incubate for 5 minutes at Room
Temperature. Centrifuge at 8000rpm for 1 minute.
14. Repeat Step 13.
15. Store the eluted DNA at 4oC for general use and at -20oC for long term
storage

Protocol 5:

Single Seed Assay for Polyphenol oxidase Activity

Stock Solutions:
a) Tween 80 Stock Solution: 20ml Tween 80 made up to 100ml with
water

153
 b) Tyrosine Solution: 0.225g Tyrosine (di-sodium salt) and 1.21g
Trizma base in 80ml water, pH 9.0. Add 10ml Tween 80 stock solution
and making the total volume up to 100ml with water.

Single seeds were placed into each well of a microplate except for the first row
which serves as the reagent blank. Single seeds of a particular genotype were
placed in every two rows of the plate (16 wells). Last row of the PPO standard:
cultivar Sunco

1. Tyrosine solution (200µl) was added to each well of the plate using a
multi-channel pipette.
2. Each plate was covered and incubated for 3 hours at 37oC.
3. After incubation, 100µl of the substrate was removed from each well
and placed into a clean microplate.
4. Microplate was put into a microplate reader, and optical densities
calculated at 415nm.

154