Use of Marker Assisted Selection For The Introgression of Quality Traits From Australian Into Chinese Wheats

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
linkage map……………………………………………………………….66.
.
Figure 2-4: QTL analysis carried out on chromosome 3B of Sunco x
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
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
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
a. b.
66
Figure 2-4: QTL analysis carried out on chromosome 3B of Sunco x Tasman DH population. All analysis used phenotypic data collected in
a. b.
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*
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*
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.
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.
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

Use of Marker Assisted Selection For The Introgression of Quality Traits From Australian Into Chinese Wheats

Uploaded by

Copyright:

Available Formats

Use of Marker Assisted Selection For The Introgression of Quality Traits From Australian Into Chinese Wheats

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Use of Marker Assisted Selection For The Introgression of Quality Traits From Australian Into Chinese Wheats

Uploaded by

Copyright:

Available Formats

University of Southern Queensland

Use of Marker Assisted Selection for the

For the Award of

Quantitative Trait Loci (QTLs) for polyphenol oxidase and xanthophyll

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

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

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

the Sunco x Tasman doubled haploid population. The combination of

contain different combinations of the PPO activity and Xanthophyll content

colour stability highlights the potential of marker assisted selection as a

useful tool in wheat breeding.

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

Professor Mark Sutherland

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

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

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

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

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

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

guidance from many people along the way.

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

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

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

from him throughout these last years.

and Raeleen Jennings of the Molecular Marker Laboratory.

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

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

Improvement Group and the technical staff within Biological Sciences.

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.

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

help, long days and your unlimited support.

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

1.2.4.1 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